diff --git a/recipes/financial_times.recipe b/recipes/financial_times.recipe index e750b6f113..0079b2be3a 100644 --- a/recipes/financial_times.recipe +++ b/recipes/financial_times.recipe @@ -1,32 +1,41 @@ -#!/usr/bin/env python - __license__ = 'GPL v3' -__copyright__ = '2008, Darko Miletic ' +__copyright__ = '2010-2011, Darko Miletic ' ''' -ft.com +www.ft.com ''' +import datetime from calibre.web.feeds.news import BasicNewsRecipe -class FinancialTimes(BasicNewsRecipe): - title = u'Financial Times' - __author__ = 'Darko Miletic and Sujata Raman' - description = ('Financial world news. Available after 5AM ' - 'GMT, daily.') +class FinancialTimes_rss(BasicNewsRecipe): + title = 'Financial Times' + __author__ = 'Darko Miletic' + description = "The Financial Times (FT) is one of the world's leading business news and information organisations, recognised internationally for its authority, integrity and accuracy." + publisher = 'The Financial Times Ltd.' + category = 'news, finances, politics, World' oldest_article = 2 - language = 'en' - - max_articles_per_feed = 100 + language = 'en' + max_articles_per_feed = 250 no_stylesheets = True use_embedded_content = False needs_subscription = True - simultaneous_downloads= 1 - delay = 1 + encoding = 'utf8' + publication_type = 'newspaper' + masthead_url = 'http://im.media.ft.com/m/img/masthead_main.jpg' + LOGIN = 'https://registration.ft.com/registration/barrier/login' + INDEX = 'http://www.ft.com' - LOGIN = 'https://registration.ft.com/registration/barrier/login' + conversion_options = { + 'comment' : description + , 'tags' : category + , 'publisher' : publisher + , 'language' : language + , 'linearize_tables' : True + } def get_browser(self): br = BasicNewsRecipe.get_browser() + br.open(self.INDEX) if self.username is not None and self.password is not None: br.open(self.LOGIN) br.select_form(name='loginForm') @@ -35,31 +44,63 @@ class FinancialTimes(BasicNewsRecipe): br.submit() return br - keep_only_tags = [ dict(name='div', attrs={'id':'cont'}) ] - remove_tags_after = dict(name='p', attrs={'class':'copyright'}) + keep_only_tags = [dict(name='div', attrs={'class':['fullstory fullstoryHeader','fullstory fullstoryBody','ft-story-header','ft-story-body','index-detail']})] remove_tags = [ - dict(name='div', attrs={'id':'floating-con'}) + dict(name='div', attrs={'id':'floating-con'}) + ,dict(name=['meta','iframe','base','object','embed','link']) + ,dict(attrs={'class':['storyTools','story-package','screen-copy','story-package separator','expandable-image']}) ] + remove_attributes = ['width','height','lang'] - extra_css = ''' - body{font-family:Arial,Helvetica,sans-serif;} - h2(font-size:large;} - .ft-story-header(font-size:xx-small;} - .ft-story-body(font-size:small;} - a{color:#003399;} + extra_css = """ + body{font-family: Georgia,Times,"Times New Roman",serif} + h2{font-size:large} + .ft-story-header{font-size: x-small} .container{font-size:x-small;} h3{font-size:x-small;color:#003399;} - ''' + .copyright{font-size: x-small} + img{margin-top: 0.8em; display: block} + .lastUpdated{font-family: Arial,Helvetica,sans-serif; font-size: x-small} + .byline,.ft-story-body,.ft-story-header{font-family: Arial,Helvetica,sans-serif} + """ + feeds = [ (u'UK' , u'http://www.ft.com/rss/home/uk' ) ,(u'US' , u'http://www.ft.com/rss/home/us' ) - ,(u'Europe' , u'http://www.ft.com/rss/home/europe' ) ,(u'Asia' , u'http://www.ft.com/rss/home/asia' ) ,(u'Middle East', u'http://www.ft.com/rss/home/middleeast') ] def preprocess_html(self, soup): - content_type = soup.find('meta', {'http-equiv':'Content-Type'}) - if content_type: - content_type['content'] = 'text/html; charset=utf-8' + items = ['promo-box','promo-title', + 'promo-headline','promo-image', + 'promo-intro','promo-link','subhead'] + for item in items: + for it in soup.findAll(item): + it.name = 'div' + it.attrs = [] + for item in soup.findAll(style=True): + del item['style'] + for item in soup.findAll('a'): + limg = item.find('img') + if item.string is not None: + str = item.string + item.replaceWith(str) + else: + if limg: + item.name = 'div' + item.attrs = [] + else: + str = self.tag_to_string(item) + item.replaceWith(str) + for item in soup.findAll('img'): + if not item.has_key('alt'): + item['alt'] = 'image' return soup + + def get_cover_url(self): + cdate = datetime.date.today() + if cdate.isoweekday() == 7: + cdate -= datetime.timedelta(days=1) + return cdate.strftime('http://specials.ft.com/vtf_pdf/%d%m%y_FRONT1_USA.pdf') + diff --git a/recipes/financial_times_uk.recipe b/recipes/financial_times_uk.recipe index 6fe1ac6acd..e06eb0dc77 100644 --- a/recipes/financial_times_uk.recipe +++ b/recipes/financial_times_uk.recipe @@ -3,6 +3,8 @@ __copyright__ = '2010-2011, Darko Miletic ' ''' www.ft.com/uk-edition ''' + +import datetime from calibre import strftime from calibre.web.feeds.news import BasicNewsRecipe @@ -20,7 +22,6 @@ class FinancialTimes(BasicNewsRecipe): needs_subscription = True encoding = 'utf8' publication_type = 'newspaper' - cover_url = strftime('http://specials.ft.com/vtf_pdf/%d%m%y_FRONT1_LON.pdf') masthead_url = 'http://im.media.ft.com/m/img/masthead_main.jpg' LOGIN = 'https://registration.ft.com/registration/barrier/login' INDEX = 'http://www.ft.com/uk-edition' @@ -128,3 +129,10 @@ class FinancialTimes(BasicNewsRecipe): if not item.has_key('alt'): item['alt'] = 'image' return soup + + def get_cover_url(self): + cdate = datetime.date.today() + if cdate.isoweekday() == 7: + cdate -= datetime.timedelta(days=1) + return cdate.strftime('http://specials.ft.com/vtf_pdf/%d%m%y_FRONT1_LON.pdf') + \ No newline at end of file diff --git a/recipes/icons/financial_times.png b/recipes/icons/financial_times.png new file mode 100644 index 0000000000..2a769d9dbb Binary files /dev/null and b/recipes/icons/financial_times.png differ diff --git a/recipes/ming_pao.recipe b/recipes/ming_pao.recipe index 08ee20cb15..3566fca667 100644 --- a/recipes/ming_pao.recipe +++ b/recipes/ming_pao.recipe @@ -1,17 +1,23 @@ -# -*- coding: utf-8 -*- __license__ = 'GPL v3' __copyright__ = '2010-2011, Eddie Lau' +# Region - Hong Kong, Vancouver, Toronto +__Region__ = 'Hong Kong' # Users of Kindle 3 with limited system-level CJK support # please replace the following "True" with "False". __MakePeriodical__ = True # Turn below to true if your device supports display of CJK titles __UseChineseTitle__ = False -# Trun below to true if you wish to use life.mingpao.com as the main article source +# Set it to False if you want to skip images +__KeepImages__ = True +# (HK only) Turn below to true if you wish to use life.mingpao.com as the main article source __UseLife__ = True + ''' Change Log: +2011/06/26: add fetching Vancouver and Toronto versions of the paper, also provide captions for images using life.mingpao fetch source + provide options to remove all images in the file 2011/05/12: switch the main parse source to life.mingpao.com, which has more photos on the article pages 2011/03/06: add new articles for finance section, also a new section "Columns" 2011/02/28: rearrange the sections @@ -34,21 +40,96 @@ Change Log: import os, datetime, re from calibre.web.feeds.recipes import BasicNewsRecipe from contextlib import nested - - from calibre.ebooks.BeautifulSoup import BeautifulSoup from calibre.ebooks.metadata.opf2 import OPFCreator from calibre.ebooks.metadata.toc import TOC from calibre.ebooks.metadata import MetaInformation -class MPHKRecipe(BasicNewsRecipe): - title = 'Ming Pao - Hong Kong' +# MAIN CLASS +class MPRecipe(BasicNewsRecipe): + if __Region__ == 'Hong Kong': + title = 'Ming Pao - Hong Kong' + description = 'Hong Kong Chinese Newspaper (http://news.mingpao.com)' + category = 'Chinese, News, Hong Kong' + extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} font>b {font-size:200%; font-weight:bold;}' + masthead_url = 'http://news.mingpao.com/image/portals_top_logo_news.gif' + keep_only_tags = [dict(name='h1'), + dict(name='font', attrs={'style':['font-size:14pt; line-height:160%;']}), # for entertainment page title + dict(name='font', attrs={'color':['AA0000']}), # for column articles title + dict(attrs={'id':['newscontent']}), # entertainment and column page content + dict(attrs={'id':['newscontent01','newscontent02']}), + dict(attrs={'class':['photo']}), + dict(name='table', attrs={'width':['100%'], 'border':['0'], 'cellspacing':['5'], 'cellpadding':['0']}), # content in printed version of life.mingpao.com + dict(name='img', attrs={'width':['180'], 'alt':['按圖放大']}) # images for source from life.mingpao.com + ] + if __KeepImages__: + remove_tags = [dict(name='style'), + dict(attrs={'id':['newscontent135']}), # for the finance page from mpfinance.com + dict(name='font', attrs={'size':['2'], 'color':['666666']}), # article date in life.mingpao.com article + #dict(name='table') # for content fetched from life.mingpao.com + ] + else: + remove_tags = [dict(name='style'), + dict(attrs={'id':['newscontent135']}), # for the finance page from mpfinance.com + dict(name='font', attrs={'size':['2'], 'color':['666666']}), # article date in life.mingpao.com article + dict(name='img'), + #dict(name='table') # for content fetched from life.mingpao.com + ] + remove_attributes = ['width'] + preprocess_regexps = [ + (re.compile(r'
', re.DOTALL|re.IGNORECASE), + lambda match: '

'), + (re.compile(r'

', re.DOTALL|re.IGNORECASE), + lambda match: ''), + (re.compile(r'

', re.DOTALL|re.IGNORECASE), # for entertainment page + lambda match: ''), + # skip
after title in life.mingpao.com fetched article + (re.compile(r"

", re.DOTALL|re.IGNORECASE), + lambda match: "
"), + (re.compile(r"

", re.DOTALL|re.IGNORECASE), + lambda match: "") + ] + elif __Region__ == 'Vancouver': + title = 'Ming Pao - Vancouver' + description = 'Vancouver Chinese Newspaper (http://www.mingpaovan.com)' + category = 'Chinese, News, Vancouver' + extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}' + masthead_url = 'http://www.mingpaovan.com/image/mainlogo2_VAN2.gif' + keep_only_tags = [dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['1']}), + dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['3'], 'cellpadding':['3'], 'id':['tblContent3']}), + dict(name='table', attrs={'width':['180'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['0'], 'bgcolor':['F0F0F0']}), + ] + if __KeepImages__: + remove_tags = [dict(name='img', attrs={'src':['../../../image/magnifier.gif']})] # the magnifier icon + else: + remove_tags = [dict(name='img')] + remove_attributes = ['width'] + preprocess_regexps = [(re.compile(r' ', re.DOTALL|re.IGNORECASE), + lambda match: ''), + ] + elif __Region__ == 'Toronto': + title = 'Ming Pao - Toronto' + description = 'Toronto Chinese Newspaper (http://www.mingpaotor.com)' + category = 'Chinese, News, Toronto' + extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}' + masthead_url = 'http://www.mingpaotor.com/image/mainlogo2_TOR2.gif' + keep_only_tags = [dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['1']}), + dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['3'], 'cellpadding':['3'], 'id':['tblContent3']}), + dict(name='table', attrs={'width':['180'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['0'], 'bgcolor':['F0F0F0']}), + ] + if __KeepImages__: + remove_tags = [dict(name='img', attrs={'src':['../../../image/magnifier.gif']})] # the magnifier icon + else: + remove_tags = [dict(name='img')] + remove_attributes = ['width'] + preprocess_regexps = [(re.compile(r' ', re.DOTALL|re.IGNORECASE), + lambda match: ''), + ] + oldest_article = 1 max_articles_per_feed = 100 __author__ = 'Eddie Lau' - description = 'Hong Kong Chinese Newspaper (http://news.mingpao.com)' publisher = 'MingPao' - category = 'Chinese, News, Hong Kong' remove_javascript = True use_embedded_content = False no_stylesheets = True @@ -57,33 +138,6 @@ class MPHKRecipe(BasicNewsRecipe): recursions = 0 conversion_options = {'linearize_tables':True} timefmt = '' - extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} font>b {font-size:200%; font-weight:bold;}' - masthead_url = 'http://news.mingpao.com/image/portals_top_logo_news.gif' - keep_only_tags = [dict(name='h1'), - dict(name='font', attrs={'style':['font-size:14pt; line-height:160%;']}), # for entertainment page title - dict(name='font', attrs={'color':['AA0000']}), # for column articles title - dict(attrs={'id':['newscontent']}), # entertainment and column page content - dict(attrs={'id':['newscontent01','newscontent02']}), - dict(attrs={'class':['photo']}), - dict(name='img', attrs={'width':['180'], 'alt':['按圖放大']}) # images for source from life.mingpao.com - ] - remove_tags = [dict(name='style'), - dict(attrs={'id':['newscontent135']}), # for the finance page from mpfinance.com - dict(name='table')] # for content fetched from life.mingpao.com - remove_attributes = ['width'] - preprocess_regexps = [ - (re.compile(r'
', re.DOTALL|re.IGNORECASE), - lambda match: '

'), - (re.compile(r'

', re.DOTALL|re.IGNORECASE), - lambda match: ''), - (re.compile(r'

', re.DOTALL|re.IGNORECASE), # for entertainment page - lambda match: ''), - # skip
after title in life.mingpao.com fetched article - (re.compile(r"

", re.DOTALL|re.IGNORECASE), - lambda match: "
"), - (re.compile(r"

", re.DOTALL|re.IGNORECASE), - lambda match: "") - ] def image_url_processor(cls, baseurl, url): # trick: break the url at the first occurance of digit, add an additional @@ -124,8 +178,18 @@ class MPHKRecipe(BasicNewsRecipe): def get_dtlocal(self): dt_utc = datetime.datetime.utcnow() - # convert UTC to local hk time - at around HKT 6.00am, all news are available - dt_local = dt_utc - datetime.timedelta(-2.0/24) + if __Region__ == 'Hong Kong': + # convert UTC to local hk time - at HKT 4.30am, all news are available + dt_local = dt_utc + datetime.timedelta(8.0/24) - datetime.timedelta(4.5/24) + # dt_local = dt_utc.astimezone(pytz.timezone('Asia/Hong_Kong')) - datetime.timedelta(4.5/24) + elif __Region__ == 'Vancouver': + # convert UTC to local Vancouver time - at PST time 4.30am, all news are available + dt_local = dt_utc + datetime.timedelta(-8.0/24) - datetime.timedelta(4.5/24) + #dt_local = dt_utc.astimezone(pytz.timezone('America/Vancouver')) - datetime.timedelta(4.5/24) + elif __Region__ == 'Toronto': + # convert UTC to local Toronto time - at EST time 4.30am, all news are available + dt_local = dt_utc + datetime.timedelta(-5.0/24) - datetime.timedelta(4.5/24) + #dt_local = dt_utc.astimezone(pytz.timezone('America/Toronto')) - datetime.timedelta(4.5/24) return dt_local def get_fetchdate(self): @@ -135,13 +199,15 @@ class MPHKRecipe(BasicNewsRecipe): return self.get_dtlocal().strftime("%Y-%m-%d") def get_fetchday(self): - # dt_utc = datetime.datetime.utcnow() - # convert UTC to local hk time - at around HKT 6.00am, all news are available - # dt_local = dt_utc - datetime.timedelta(-2.0/24) return self.get_dtlocal().strftime("%d") def get_cover_url(self): - cover = 'http://news.mingpao.com/' + self.get_fetchdate() + '/' + self.get_fetchdate() + '_' + self.get_fetchday() + 'gacov.jpg' + if __Region__ == 'Hong Kong': + cover = 'http://news.mingpao.com/' + self.get_fetchdate() + '/' + self.get_fetchdate() + '_' + self.get_fetchday() + 'gacov.jpg' + elif __Region__ == 'Vancouver': + cover = 'http://www.mingpaovan.com/ftp/News/' + self.get_fetchdate() + '/' + self.get_fetchday() + 'pgva1s.jpg' + elif __Region__ == 'Toronto': + cover = 'http://www.mingpaotor.com/ftp/News/' + self.get_fetchdate() + '/' + self.get_fetchday() + 'pgtas.jpg' br = BasicNewsRecipe.get_browser() try: br.open(cover) @@ -153,76 +219,104 @@ class MPHKRecipe(BasicNewsRecipe): feeds = [] dateStr = self.get_fetchdate() - if __UseLife__: - for title, url, keystr in [(u'\u8981\u805e Headline', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalga', 'nal'), - (u'\u6e2f\u805e Local', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalgb', 'nal'), - (u'\u6559\u80b2 Education', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalgf', 'nal'), - (u'\u793e\u8a55/\u7b46\u9663 Editorial', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr', 'nal'), - (u'\u8ad6\u58c7 Forum', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalfa', 'nal'), - (u'\u4e2d\u570b China', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalca', 'nal'), - (u'\u570b\u969b World', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalta', 'nal'), - (u'\u7d93\u6fdf Finance', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea', 'nal'), - (u'\u9ad4\u80b2 Sport', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalsp', 'nal'), - (u'\u5f71\u8996 Film/TV', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalma', 'nal'), - (u'\u5c08\u6b04 Columns', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn', 'ncl')]: - articles = self.parse_section2(url, keystr) + if __Region__ == 'Hong Kong': + if __UseLife__: + for title, url, keystr in [(u'\u8981\u805e Headline', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalga', 'nal'), + (u'\u6e2f\u805e Local', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalgb', 'nal'), + (u'\u6559\u80b2 Education', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalgf', 'nal'), + (u'\u793e\u8a55/\u7b46\u9663 Editorial', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr', 'nal'), + (u'\u8ad6\u58c7 Forum', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalfa', 'nal'), + (u'\u4e2d\u570b China', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalca', 'nal'), + (u'\u570b\u969b World', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalta', 'nal'), + (u'\u7d93\u6fdf Finance', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea', 'nal'), + (u'\u9ad4\u80b2 Sport', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalsp', 'nal'), + (u'\u5f71\u8996 Film/TV', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalma', 'nal'), + (u'\u5c08\u6b04 Columns', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn', 'ncl')]: + articles = self.parse_section2(url, keystr) + if articles: + feeds.append((title, articles)) + + for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'), + (u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + else: + for title, url in [(u'\u8981\u805e Headline', 'http://news.mingpao.com/' + dateStr + '/gaindex.htm'), + (u'\u6e2f\u805e Local', 'http://news.mingpao.com/' + dateStr + '/gbindex.htm'), + (u'\u6559\u80b2 Education', 'http://news.mingpao.com/' + dateStr + '/gfindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + # special- editorial + ed_articles = self.parse_ed_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr') + if ed_articles: + feeds.append((u'\u793e\u8a55/\u7b46\u9663 Editorial', ed_articles)) + + for title, url in [(u'\u8ad6\u58c7 Forum', 'http://news.mingpao.com/' + dateStr + '/faindex.htm'), + (u'\u4e2d\u570b China', 'http://news.mingpao.com/' + dateStr + '/caindex.htm'), + (u'\u570b\u969b World', 'http://news.mingpao.com/' + dateStr + '/taindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + # special - finance + #fin_articles = self.parse_fin_section('http://www.mpfinance.com/htm/Finance/' + dateStr + '/News/ea,eb,ecindex.htm') + fin_articles = self.parse_fin_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea') + if fin_articles: + feeds.append((u'\u7d93\u6fdf Finance', fin_articles)) + + for title, url in [('Tech News', 'http://news.mingpao.com/' + dateStr + '/naindex.htm'), + (u'\u9ad4\u80b2 Sport', 'http://news.mingpao.com/' + dateStr + '/spindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + # special - entertainment + ent_articles = self.parse_ent_section('http://ol.mingpao.com/cfm/star1.cfm') + if ent_articles: + feeds.append((u'\u5f71\u8996 Film/TV', ent_articles)) + + for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'), + (u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + + # special- columns + col_articles = self.parse_col_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn') + if col_articles: + feeds.append((u'\u5c08\u6b04 Columns', col_articles)) + elif __Region__ == 'Vancouver': + for title, url in [(u'\u8981\u805e Headline', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VAindex.htm'), + (u'\u52a0\u570b Canada', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VBindex.htm'), + (u'\u793e\u5340 Local', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VDindex.htm'), + (u'\u6e2f\u805e Hong Kong', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/HK-VGindex.htm'), + (u'\u570b\u969b World', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VTindex.htm'), + (u'\u4e2d\u570b China', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VCindex.htm'), + (u'\u7d93\u6fdf Economics', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VEindex.htm'), + (u'\u9ad4\u80b2 Sports', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VSindex.htm'), + (u'\u5f71\u8996 Film/TV', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/HK-MAindex.htm'), + (u'\u526f\u520a Supplements', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/WWindex.htm'),]: + articles = self.parse_section3(url, 'http://www.mingpaovan.com/') if articles: feeds.append((title, articles)) - - for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'), - (u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]: - articles = self.parse_section(url) + elif __Region__ == 'Toronto': + for title, url in [(u'\u8981\u805e Headline', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TAindex.htm'), + (u'\u52a0\u570b Canada', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TDindex.htm'), + (u'\u793e\u5340 Local', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TFindex.htm'), + (u'\u4e2d\u570b China', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TCAindex.htm'), + (u'\u570b\u969b World', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TTAindex.htm'), + (u'\u6e2f\u805e Hong Kong', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/HK-GAindex.htm'), + (u'\u7d93\u6fdf Economics', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/THindex.htm'), + (u'\u9ad4\u80b2 Sports', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TSindex.htm'), + (u'\u5f71\u8996 Film/TV', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/HK-MAindex.htm'), + (u'\u526f\u520a Supplements', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/WWindex.htm'),]: + articles = self.parse_section3(url, 'http://www.mingpaotor.com/') if articles: feeds.append((title, articles)) - else: - for title, url in [(u'\u8981\u805e Headline', 'http://news.mingpao.com/' + dateStr + '/gaindex.htm'), - (u'\u6e2f\u805e Local', 'http://news.mingpao.com/' + dateStr + '/gbindex.htm'), - (u'\u6559\u80b2 Education', 'http://news.mingpao.com/' + dateStr + '/gfindex.htm')]: - articles = self.parse_section(url) - if articles: - feeds.append((title, articles)) - - # special- editorial - ed_articles = self.parse_ed_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr') - if ed_articles: - feeds.append((u'\u793e\u8a55/\u7b46\u9663 Editorial', ed_articles)) - - for title, url in [(u'\u8ad6\u58c7 Forum', 'http://news.mingpao.com/' + dateStr + '/faindex.htm'), - (u'\u4e2d\u570b China', 'http://news.mingpao.com/' + dateStr + '/caindex.htm'), - (u'\u570b\u969b World', 'http://news.mingpao.com/' + dateStr + '/taindex.htm')]: - articles = self.parse_section(url) - if articles: - feeds.append((title, articles)) - - # special - finance - #fin_articles = self.parse_fin_section('http://www.mpfinance.com/htm/Finance/' + dateStr + '/News/ea,eb,ecindex.htm') - fin_articles = self.parse_fin_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea') - if fin_articles: - feeds.append((u'\u7d93\u6fdf Finance', fin_articles)) - - for title, url in [('Tech News', 'http://news.mingpao.com/' + dateStr + '/naindex.htm'), - (u'\u9ad4\u80b2 Sport', 'http://news.mingpao.com/' + dateStr + '/spindex.htm')]: - articles = self.parse_section(url) - if articles: - feeds.append((title, articles)) - - # special - entertainment - ent_articles = self.parse_ent_section('http://ol.mingpao.com/cfm/star1.cfm') - if ent_articles: - feeds.append((u'\u5f71\u8996 Film/TV', ent_articles)) - - for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'), - (u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]: - articles = self.parse_section(url) - if articles: - feeds.append((title, articles)) - - - # special- columns - col_articles = self.parse_col_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn') - if col_articles: - feeds.append((u'\u5c08\u6b04 Columns', col_articles)) - return feeds # parse from news.mingpao.com @@ -256,11 +350,30 @@ class MPHKRecipe(BasicNewsRecipe): title = self.tag_to_string(i) url = 'http://life.mingpao.com/cfm/' + i.get('href', False) if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind(keystr) == -1): + url = url.replace('dailynews3.cfm', 'dailynews3a.cfm') # use printed version of the article current_articles.append({'title': title, 'url': url, 'description': ''}) included_urls.append(url) current_articles.reverse() return current_articles + # parse from www.mingpaovan.com + def parse_section3(self, url, baseUrl): + self.get_fetchdate() + soup = self.index_to_soup(url) + divs = soup.findAll(attrs={'class': ['ListContentLargeLink']}) + current_articles = [] + included_urls = [] + divs.reverse() + for i in divs: + title = self.tag_to_string(i) + urlstr = i.get('href', False) + urlstr = baseUrl + '/' + urlstr.replace('../../../', '') + if urlstr not in included_urls: + current_articles.append({'title': title, 'url': urlstr, 'description': '', 'date': ''}) + included_urls.append(urlstr) + current_articles.reverse() + return current_articles + def parse_ed_section(self, url): self.get_fetchdate() soup = self.index_to_soup(url) @@ -338,7 +451,12 @@ class MPHKRecipe(BasicNewsRecipe): if dir is None: dir = self.output_dir if __UseChineseTitle__ == True: - title = u'\u660e\u5831 (\u9999\u6e2f)' + if __Region__ == 'Hong Kong': + title = u'\u660e\u5831 (\u9999\u6e2f)' + elif __Region__ == 'Vancouver': + title = u'\u660e\u5831 (\u6eab\u54e5\u83ef)' + elif __Region__ == 'Toronto': + title = u'\u660e\u5831 (\u591a\u502b\u591a)' else: title = self.short_title() # if not generating a periodical, force date to apply in title diff --git a/recipes/ming_pao_toronto.recipe b/recipes/ming_pao_toronto.recipe new file mode 100644 index 0000000000..677a8272b0 --- /dev/null +++ b/recipes/ming_pao_toronto.recipe @@ -0,0 +1,594 @@ +__license__ = 'GPL v3' +__copyright__ = '2010-2011, Eddie Lau' + +# Region - Hong Kong, Vancouver, Toronto +__Region__ = 'Toronto' +# Users of Kindle 3 with limited system-level CJK support +# please replace the following "True" with "False". +__MakePeriodical__ = True +# Turn below to true if your device supports display of CJK titles +__UseChineseTitle__ = False +# Set it to False if you want to skip images +__KeepImages__ = True +# (HK only) Turn below to true if you wish to use life.mingpao.com as the main article source +__UseLife__ = True + + +''' +Change Log: +2011/06/26: add fetching Vancouver and Toronto versions of the paper, also provide captions for images using life.mingpao fetch source + provide options to remove all images in the file +2011/05/12: switch the main parse source to life.mingpao.com, which has more photos on the article pages +2011/03/06: add new articles for finance section, also a new section "Columns" +2011/02/28: rearrange the sections + [Disabled until Kindle has better CJK support and can remember last (section,article) read in Sections & Articles + View] make it the same title if generating a periodical, so past issue will be automatically put into "Past Issues" + folder in Kindle 3 +2011/02/20: skip duplicated links in finance section, put photos which may extend a whole page to the back of the articles + clean up the indentation +2010/12/07: add entertainment section, use newspaper front page as ebook cover, suppress date display in section list + (to avoid wrong date display in case the user generates the ebook in a time zone different from HKT) +2010/11/22: add English section, remove eco-news section which is not updated daily, correct + ordering of articles +2010/11/12: add news image and eco-news section +2010/11/08: add parsing of finance section +2010/11/06: temporary work-around for Kindle device having no capability to display unicode + in section/article list. +2010/10/31: skip repeated articles in section pages +''' + +import os, datetime, re +from calibre.web.feeds.recipes import BasicNewsRecipe +from contextlib import nested +from calibre.ebooks.BeautifulSoup import BeautifulSoup +from calibre.ebooks.metadata.opf2 import OPFCreator +from calibre.ebooks.metadata.toc import TOC +from calibre.ebooks.metadata import MetaInformation + +# MAIN CLASS +class MPRecipe(BasicNewsRecipe): + if __Region__ == 'Hong Kong': + title = 'Ming Pao - Hong Kong' + description = 'Hong Kong Chinese Newspaper (http://news.mingpao.com)' + category = 'Chinese, News, Hong Kong' + extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} font>b {font-size:200%; font-weight:bold;}' + masthead_url = 'http://news.mingpao.com/image/portals_top_logo_news.gif' + keep_only_tags = [dict(name='h1'), + dict(name='font', attrs={'style':['font-size:14pt; line-height:160%;']}), # for entertainment page title + dict(name='font', attrs={'color':['AA0000']}), # for column articles title + dict(attrs={'id':['newscontent']}), # entertainment and column page content + dict(attrs={'id':['newscontent01','newscontent02']}), + dict(attrs={'class':['photo']}), + dict(name='table', attrs={'width':['100%'], 'border':['0'], 'cellspacing':['5'], 'cellpadding':['0']}), # content in printed version of life.mingpao.com + dict(name='img', attrs={'width':['180'], 'alt':['按圖放大']}) # images for source from life.mingpao.com + ] + if __KeepImages__: + remove_tags = [dict(name='style'), + dict(attrs={'id':['newscontent135']}), # for the finance page from mpfinance.com + dict(name='font', attrs={'size':['2'], 'color':['666666']}), # article date in life.mingpao.com article + #dict(name='table') # for content fetched from life.mingpao.com + ] + else: + remove_tags = [dict(name='style'), + dict(attrs={'id':['newscontent135']}), # for the finance page from mpfinance.com + dict(name='font', attrs={'size':['2'], 'color':['666666']}), # article date in life.mingpao.com article + dict(name='img'), + #dict(name='table') # for content fetched from life.mingpao.com + ] + remove_attributes = ['width'] + preprocess_regexps = [ + (re.compile(r'
', re.DOTALL|re.IGNORECASE), + lambda match: '

'), + (re.compile(r'

', re.DOTALL|re.IGNORECASE), + lambda match: ''), + (re.compile(r'

', re.DOTALL|re.IGNORECASE), # for entertainment page + lambda match: ''), + # skip
after title in life.mingpao.com fetched article + (re.compile(r"

", re.DOTALL|re.IGNORECASE), + lambda match: "
"), + (re.compile(r"

", re.DOTALL|re.IGNORECASE), + lambda match: "") + ] + elif __Region__ == 'Vancouver': + title = 'Ming Pao - Vancouver' + description = 'Vancouver Chinese Newspaper (http://www.mingpaovan.com)' + category = 'Chinese, News, Vancouver' + extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}' + masthead_url = 'http://www.mingpaovan.com/image/mainlogo2_VAN2.gif' + keep_only_tags = [dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['1']}), + dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['3'], 'cellpadding':['3'], 'id':['tblContent3']}), + dict(name='table', attrs={'width':['180'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['0'], 'bgcolor':['F0F0F0']}), + ] + if __KeepImages__: + remove_tags = [dict(name='img', attrs={'src':['../../../image/magnifier.gif']})] # the magnifier icon + else: + remove_tags = [dict(name='img')] + remove_attributes = ['width'] + preprocess_regexps = [(re.compile(r' ', re.DOTALL|re.IGNORECASE), + lambda match: ''), + ] + elif __Region__ == 'Toronto': + title = 'Ming Pao - Toronto' + description = 'Toronto Chinese Newspaper (http://www.mingpaotor.com)' + category = 'Chinese, News, Toronto' + extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}' + masthead_url = 'http://www.mingpaotor.com/image/mainlogo2_TOR2.gif' + keep_only_tags = [dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['1']}), + dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['3'], 'cellpadding':['3'], 'id':['tblContent3']}), + dict(name='table', attrs={'width':['180'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['0'], 'bgcolor':['F0F0F0']}), + ] + if __KeepImages__: + remove_tags = [dict(name='img', attrs={'src':['../../../image/magnifier.gif']})] # the magnifier icon + else: + remove_tags = [dict(name='img')] + remove_attributes = ['width'] + preprocess_regexps = [(re.compile(r' ', re.DOTALL|re.IGNORECASE), + lambda match: ''), + ] + + oldest_article = 1 + max_articles_per_feed = 100 + __author__ = 'Eddie Lau' + publisher = 'MingPao' + remove_javascript = True + use_embedded_content = False + no_stylesheets = True + language = 'zh' + encoding = 'Big5-HKSCS' + recursions = 0 + conversion_options = {'linearize_tables':True} + timefmt = '' + + def image_url_processor(cls, baseurl, url): + # trick: break the url at the first occurance of digit, add an additional + # '_' at the front + # not working, may need to move this to preprocess_html() method +# minIdx = 10000 +# i0 = url.find('0') +# if i0 >= 0 and i0 < minIdx: +# minIdx = i0 +# i1 = url.find('1') +# if i1 >= 0 and i1 < minIdx: +# minIdx = i1 +# i2 = url.find('2') +# if i2 >= 0 and i2 < minIdx: +# minIdx = i2 +# i3 = url.find('3') +# if i3 >= 0 and i0 < minIdx: +# minIdx = i3 +# i4 = url.find('4') +# if i4 >= 0 and i4 < minIdx: +# minIdx = i4 +# i5 = url.find('5') +# if i5 >= 0 and i5 < minIdx: +# minIdx = i5 +# i6 = url.find('6') +# if i6 >= 0 and i6 < minIdx: +# minIdx = i6 +# i7 = url.find('7') +# if i7 >= 0 and i7 < minIdx: +# minIdx = i7 +# i8 = url.find('8') +# if i8 >= 0 and i8 < minIdx: +# minIdx = i8 +# i9 = url.find('9') +# if i9 >= 0 and i9 < minIdx: +# minIdx = i9 + return url + + def get_dtlocal(self): + dt_utc = datetime.datetime.utcnow() + if __Region__ == 'Hong Kong': + # convert UTC to local hk time - at HKT 4.30am, all news are available + dt_local = dt_utc + datetime.timedelta(8.0/24) - datetime.timedelta(4.5/24) + # dt_local = dt_utc.astimezone(pytz.timezone('Asia/Hong_Kong')) - datetime.timedelta(4.5/24) + elif __Region__ == 'Vancouver': + # convert UTC to local Vancouver time - at PST time 4.30am, all news are available + dt_local = dt_utc + datetime.timedelta(-8.0/24) - datetime.timedelta(4.5/24) + #dt_local = dt_utc.astimezone(pytz.timezone('America/Vancouver')) - datetime.timedelta(4.5/24) + elif __Region__ == 'Toronto': + # convert UTC to local Toronto time - at EST time 4.30am, all news are available + dt_local = dt_utc + datetime.timedelta(-5.0/24) - datetime.timedelta(4.5/24) + #dt_local = dt_utc.astimezone(pytz.timezone('America/Toronto')) - datetime.timedelta(4.5/24) + return dt_local + + def get_fetchdate(self): + return self.get_dtlocal().strftime("%Y%m%d") + + def get_fetchformatteddate(self): + return self.get_dtlocal().strftime("%Y-%m-%d") + + def get_fetchday(self): + return self.get_dtlocal().strftime("%d") + + def get_cover_url(self): + if __Region__ == 'Hong Kong': + cover = 'http://news.mingpao.com/' + self.get_fetchdate() + '/' + self.get_fetchdate() + '_' + self.get_fetchday() + 'gacov.jpg' + elif __Region__ == 'Vancouver': + cover = 'http://www.mingpaovan.com/ftp/News/' + self.get_fetchdate() + '/' + self.get_fetchday() + 'pgva1s.jpg' + elif __Region__ == 'Toronto': + cover = 'http://www.mingpaotor.com/ftp/News/' + self.get_fetchdate() + '/' + self.get_fetchday() + 'pgtas.jpg' + br = BasicNewsRecipe.get_browser() + try: + br.open(cover) + except: + cover = None + return cover + + def parse_index(self): + feeds = [] + dateStr = self.get_fetchdate() + + if __Region__ == 'Hong Kong': + if __UseLife__: + for title, url, keystr in [(u'\u8981\u805e Headline', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalga', 'nal'), + (u'\u6e2f\u805e Local', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalgb', 'nal'), + (u'\u6559\u80b2 Education', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalgf', 'nal'), + (u'\u793e\u8a55/\u7b46\u9663 Editorial', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr', 'nal'), + (u'\u8ad6\u58c7 Forum', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalfa', 'nal'), + (u'\u4e2d\u570b China', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalca', 'nal'), + (u'\u570b\u969b World', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalta', 'nal'), + (u'\u7d93\u6fdf Finance', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea', 'nal'), + (u'\u9ad4\u80b2 Sport', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalsp', 'nal'), + (u'\u5f71\u8996 Film/TV', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalma', 'nal'), + (u'\u5c08\u6b04 Columns', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn', 'ncl')]: + articles = self.parse_section2(url, keystr) + if articles: + feeds.append((title, articles)) + + for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'), + (u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + else: + for title, url in [(u'\u8981\u805e Headline', 'http://news.mingpao.com/' + dateStr + '/gaindex.htm'), + (u'\u6e2f\u805e Local', 'http://news.mingpao.com/' + dateStr + '/gbindex.htm'), + (u'\u6559\u80b2 Education', 'http://news.mingpao.com/' + dateStr + '/gfindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + # special- editorial + ed_articles = self.parse_ed_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr') + if ed_articles: + feeds.append((u'\u793e\u8a55/\u7b46\u9663 Editorial', ed_articles)) + + for title, url in [(u'\u8ad6\u58c7 Forum', 'http://news.mingpao.com/' + dateStr + '/faindex.htm'), + (u'\u4e2d\u570b China', 'http://news.mingpao.com/' + dateStr + '/caindex.htm'), + (u'\u570b\u969b World', 'http://news.mingpao.com/' + dateStr + '/taindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + # special - finance + #fin_articles = self.parse_fin_section('http://www.mpfinance.com/htm/Finance/' + dateStr + '/News/ea,eb,ecindex.htm') + fin_articles = self.parse_fin_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea') + if fin_articles: + feeds.append((u'\u7d93\u6fdf Finance', fin_articles)) + + for title, url in [('Tech News', 'http://news.mingpao.com/' + dateStr + '/naindex.htm'), + (u'\u9ad4\u80b2 Sport', 'http://news.mingpao.com/' + dateStr + '/spindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + # special - entertainment + ent_articles = self.parse_ent_section('http://ol.mingpao.com/cfm/star1.cfm') + if ent_articles: + feeds.append((u'\u5f71\u8996 Film/TV', ent_articles)) + + for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'), + (u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + + # special- columns + col_articles = self.parse_col_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn') + if col_articles: + feeds.append((u'\u5c08\u6b04 Columns', col_articles)) + elif __Region__ == 'Vancouver': + for title, url in [(u'\u8981\u805e Headline', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VAindex.htm'), + (u'\u52a0\u570b Canada', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VBindex.htm'), + (u'\u793e\u5340 Local', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VDindex.htm'), + (u'\u6e2f\u805e Hong Kong', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/HK-VGindex.htm'), + (u'\u570b\u969b World', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VTindex.htm'), + (u'\u4e2d\u570b China', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VCindex.htm'), + (u'\u7d93\u6fdf Economics', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VEindex.htm'), + (u'\u9ad4\u80b2 Sports', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VSindex.htm'), + (u'\u5f71\u8996 Film/TV', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/HK-MAindex.htm'), + (u'\u526f\u520a Supplements', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/WWindex.htm'),]: + articles = self.parse_section3(url, 'http://www.mingpaovan.com/') + if articles: + feeds.append((title, articles)) + elif __Region__ == 'Toronto': + for title, url in [(u'\u8981\u805e Headline', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TAindex.htm'), + (u'\u52a0\u570b Canada', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TDindex.htm'), + (u'\u793e\u5340 Local', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TFindex.htm'), + (u'\u4e2d\u570b China', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TCAindex.htm'), + (u'\u570b\u969b World', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TTAindex.htm'), + (u'\u6e2f\u805e Hong Kong', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/HK-GAindex.htm'), + (u'\u7d93\u6fdf Economics', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/THindex.htm'), + (u'\u9ad4\u80b2 Sports', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TSindex.htm'), + (u'\u5f71\u8996 Film/TV', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/HK-MAindex.htm'), + (u'\u526f\u520a Supplements', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/WWindex.htm'),]: + articles = self.parse_section3(url, 'http://www.mingpaotor.com/') + if articles: + feeds.append((title, articles)) + return feeds + + # parse from news.mingpao.com + def parse_section(self, url): + dateStr = self.get_fetchdate() + soup = self.index_to_soup(url) + divs = soup.findAll(attrs={'class': ['bullet','bullet_grey']}) + current_articles = [] + included_urls = [] + divs.reverse() + for i in divs: + a = i.find('a', href = True) + title = self.tag_to_string(a) + url = a.get('href', False) + url = 'http://news.mingpao.com/' + dateStr + '/' +url + if url not in included_urls and url.rfind('Redirect') == -1: + current_articles.append({'title': title, 'url': url, 'description':'', 'date':''}) + included_urls.append(url) + current_articles.reverse() + return current_articles + + # parse from life.mingpao.com + def parse_section2(self, url, keystr): + self.get_fetchdate() + soup = self.index_to_soup(url) + a = soup.findAll('a', href=True) + a.reverse() + current_articles = [] + included_urls = [] + for i in a: + title = self.tag_to_string(i) + url = 'http://life.mingpao.com/cfm/' + i.get('href', False) + if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind(keystr) == -1): + url = url.replace('dailynews3.cfm', 'dailynews3a.cfm') # use printed version of the article + current_articles.append({'title': title, 'url': url, 'description': ''}) + included_urls.append(url) + current_articles.reverse() + return current_articles + + # parse from www.mingpaovan.com + def parse_section3(self, url, baseUrl): + self.get_fetchdate() + soup = self.index_to_soup(url) + divs = soup.findAll(attrs={'class': ['ListContentLargeLink']}) + current_articles = [] + included_urls = [] + divs.reverse() + for i in divs: + title = self.tag_to_string(i) + urlstr = i.get('href', False) + urlstr = baseUrl + '/' + urlstr.replace('../../../', '') + if urlstr not in included_urls: + current_articles.append({'title': title, 'url': urlstr, 'description': '', 'date': ''}) + included_urls.append(urlstr) + current_articles.reverse() + return current_articles + + def parse_ed_section(self, url): + self.get_fetchdate() + soup = self.index_to_soup(url) + a = soup.findAll('a', href=True) + a.reverse() + current_articles = [] + included_urls = [] + for i in a: + title = self.tag_to_string(i) + url = 'http://life.mingpao.com/cfm/' + i.get('href', False) + if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind('nal') == -1): + current_articles.append({'title': title, 'url': url, 'description': ''}) + included_urls.append(url) + current_articles.reverse() + return current_articles + + def parse_fin_section(self, url): + self.get_fetchdate() + soup = self.index_to_soup(url) + a = soup.findAll('a', href= True) + current_articles = [] + included_urls = [] + for i in a: + #url = 'http://www.mpfinance.com/cfm/' + i.get('href', False) + url = 'http://life.mingpao.com/cfm/' + i.get('href', False) + #if url not in included_urls and not url.rfind(dateStr) == -1 and url.rfind('index') == -1: + if url not in included_urls and (not url.rfind('txt') == -1) and (not url.rfind('nal') == -1): + title = self.tag_to_string(i) + current_articles.append({'title': title, 'url': url, 'description':''}) + included_urls.append(url) + return current_articles + + def parse_ent_section(self, url): + self.get_fetchdate() + soup = self.index_to_soup(url) + a = soup.findAll('a', href=True) + a.reverse() + current_articles = [] + included_urls = [] + for i in a: + title = self.tag_to_string(i) + url = 'http://ol.mingpao.com/cfm/' + i.get('href', False) + if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind('star') == -1): + current_articles.append({'title': title, 'url': url, 'description': ''}) + included_urls.append(url) + current_articles.reverse() + return current_articles + + def parse_col_section(self, url): + self.get_fetchdate() + soup = self.index_to_soup(url) + a = soup.findAll('a', href=True) + a.reverse() + current_articles = [] + included_urls = [] + for i in a: + title = self.tag_to_string(i) + url = 'http://life.mingpao.com/cfm/' + i.get('href', False) + if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind('ncl') == -1): + current_articles.append({'title': title, 'url': url, 'description': ''}) + included_urls.append(url) + current_articles.reverse() + return current_articles + + def preprocess_html(self, soup): + for item in soup.findAll(style=True): + del item['style'] + for item in soup.findAll(style=True): + del item['width'] + for item in soup.findAll(stype=True): + del item['absmiddle'] + return soup + + def create_opf(self, feeds, dir=None): + if dir is None: + dir = self.output_dir + if __UseChineseTitle__ == True: + if __Region__ == 'Hong Kong': + title = u'\u660e\u5831 (\u9999\u6e2f)' + elif __Region__ == 'Vancouver': + title = u'\u660e\u5831 (\u6eab\u54e5\u83ef)' + elif __Region__ == 'Toronto': + title = u'\u660e\u5831 (\u591a\u502b\u591a)' + else: + title = self.short_title() + # if not generating a periodical, force date to apply in title + if __MakePeriodical__ == False: + title = title + ' ' + self.get_fetchformatteddate() + if True: + mi = MetaInformation(title, [self.publisher]) + mi.publisher = self.publisher + mi.author_sort = self.publisher + if __MakePeriodical__ == True: + mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title() + else: + mi.publication_type = self.publication_type+':'+self.short_title() + #mi.timestamp = nowf() + mi.timestamp = self.get_dtlocal() + mi.comments = self.description + if not isinstance(mi.comments, unicode): + mi.comments = mi.comments.decode('utf-8', 'replace') + #mi.pubdate = nowf() + mi.pubdate = self.get_dtlocal() + opf_path = os.path.join(dir, 'index.opf') + ncx_path = os.path.join(dir, 'index.ncx') + opf = OPFCreator(dir, mi) + # Add mastheadImage entry to section + mp = getattr(self, 'masthead_path', None) + if mp is not None and os.access(mp, os.R_OK): + from calibre.ebooks.metadata.opf2 import Guide + ref = Guide.Reference(os.path.basename(self.masthead_path), os.getcwdu()) + ref.type = 'masthead' + ref.title = 'Masthead Image' + opf.guide.append(ref) + + manifest = [os.path.join(dir, 'feed_%d'%i) for i in range(len(feeds))] + manifest.append(os.path.join(dir, 'index.html')) + manifest.append(os.path.join(dir, 'index.ncx')) + + # Get cover + cpath = getattr(self, 'cover_path', None) + if cpath is None: + pf = open(os.path.join(dir, 'cover.jpg'), 'wb') + if self.default_cover(pf): + cpath = pf.name + if cpath is not None and os.access(cpath, os.R_OK): + opf.cover = cpath + manifest.append(cpath) + + # Get masthead + mpath = getattr(self, 'masthead_path', None) + if mpath is not None and os.access(mpath, os.R_OK): + manifest.append(mpath) + + opf.create_manifest_from_files_in(manifest) + for mani in opf.manifest: + if mani.path.endswith('.ncx'): + mani.id = 'ncx' + if mani.path.endswith('mastheadImage.jpg'): + mani.id = 'masthead-image' + entries = ['index.html'] + toc = TOC(base_path=dir) + self.play_order_counter = 0 + self.play_order_map = {} + + def feed_index(num, parent): + f = feeds[num] + for j, a in enumerate(f): + if getattr(a, 'downloaded', False): + adir = 'feed_%d/article_%d/'%(num, j) + auth = a.author + if not auth: + auth = None + desc = a.text_summary + if not desc: + desc = None + else: + desc = self.description_limiter(desc) + entries.append('%sindex.html'%adir) + po = self.play_order_map.get(entries[-1], None) + if po is None: + self.play_order_counter += 1 + po = self.play_order_counter + parent.add_item('%sindex.html'%adir, None, a.title if a.title else _('Untitled Article'), + play_order=po, author=auth, description=desc) + last = os.path.join(self.output_dir, ('%sindex.html'%adir).replace('/', os.sep)) + for sp in a.sub_pages: + prefix = os.path.commonprefix([opf_path, sp]) + relp = sp[len(prefix):] + entries.append(relp.replace(os.sep, '/')) + last = sp + + if os.path.exists(last): + with open(last, 'rb') as fi: + src = fi.read().decode('utf-8') + soup = BeautifulSoup(src) + body = soup.find('body') + if body is not None: + prefix = '/'.join('..'for i in range(2*len(re.findall(r'link\d+', last)))) + templ = self.navbar.generate(True, num, j, len(f), + not self.has_single_feed, + a.orig_url, self.publisher, prefix=prefix, + center=self.center_navbar) + elem = BeautifulSoup(templ.render(doctype='xhtml').decode('utf-8')).find('div') + body.insert(len(body.contents), elem) + with open(last, 'wb') as fi: + fi.write(unicode(soup).encode('utf-8')) + if len(feeds) == 0: + raise Exception('All feeds are empty, aborting.') + + if len(feeds) > 1: + for i, f in enumerate(feeds): + entries.append('feed_%d/index.html'%i) + po = self.play_order_map.get(entries[-1], None) + if po is None: + self.play_order_counter += 1 + po = self.play_order_counter + auth = getattr(f, 'author', None) + if not auth: + auth = None + desc = getattr(f, 'description', None) + if not desc: + desc = None + feed_index(i, toc.add_item('feed_%d/index.html'%i, None, + f.title, play_order=po, description=desc, author=auth)) + + else: + entries.append('feed_%d/index.html'%0) + feed_index(0, toc) + + for i, p in enumerate(entries): + entries[i] = os.path.join(dir, p.replace('/', os.sep)) + opf.create_spine(entries) + opf.set_toc(toc) + + with nested(open(opf_path, 'wb'), open(ncx_path, 'wb')) as (opf_file, ncx_file): + opf.render(opf_file, ncx_file) + diff --git a/recipes/ming_pao_vancouver.recipe b/recipes/ming_pao_vancouver.recipe new file mode 100644 index 0000000000..3312c8f7b8 --- /dev/null +++ b/recipes/ming_pao_vancouver.recipe @@ -0,0 +1,594 @@ +__license__ = 'GPL v3' +__copyright__ = '2010-2011, Eddie Lau' + +# Region - Hong Kong, Vancouver, Toronto +__Region__ = 'Vancouver' +# Users of Kindle 3 with limited system-level CJK support +# please replace the following "True" with "False". +__MakePeriodical__ = True +# Turn below to true if your device supports display of CJK titles +__UseChineseTitle__ = False +# Set it to False if you want to skip images +__KeepImages__ = True +# (HK only) Turn below to true if you wish to use life.mingpao.com as the main article source +__UseLife__ = True + + +''' +Change Log: +2011/06/26: add fetching Vancouver and Toronto versions of the paper, also provide captions for images using life.mingpao fetch source + provide options to remove all images in the file +2011/05/12: switch the main parse source to life.mingpao.com, which has more photos on the article pages +2011/03/06: add new articles for finance section, also a new section "Columns" +2011/02/28: rearrange the sections + [Disabled until Kindle has better CJK support and can remember last (section,article) read in Sections & Articles + View] make it the same title if generating a periodical, so past issue will be automatically put into "Past Issues" + folder in Kindle 3 +2011/02/20: skip duplicated links in finance section, put photos which may extend a whole page to the back of the articles + clean up the indentation +2010/12/07: add entertainment section, use newspaper front page as ebook cover, suppress date display in section list + (to avoid wrong date display in case the user generates the ebook in a time zone different from HKT) +2010/11/22: add English section, remove eco-news section which is not updated daily, correct + ordering of articles +2010/11/12: add news image and eco-news section +2010/11/08: add parsing of finance section +2010/11/06: temporary work-around for Kindle device having no capability to display unicode + in section/article list. +2010/10/31: skip repeated articles in section pages +''' + +import os, datetime, re +from calibre.web.feeds.recipes import BasicNewsRecipe +from contextlib import nested +from calibre.ebooks.BeautifulSoup import BeautifulSoup +from calibre.ebooks.metadata.opf2 import OPFCreator +from calibre.ebooks.metadata.toc import TOC +from calibre.ebooks.metadata import MetaInformation + +# MAIN CLASS +class MPRecipe(BasicNewsRecipe): + if __Region__ == 'Hong Kong': + title = 'Ming Pao - Hong Kong' + description = 'Hong Kong Chinese Newspaper (http://news.mingpao.com)' + category = 'Chinese, News, Hong Kong' + extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} font>b {font-size:200%; font-weight:bold;}' + masthead_url = 'http://news.mingpao.com/image/portals_top_logo_news.gif' + keep_only_tags = [dict(name='h1'), + dict(name='font', attrs={'style':['font-size:14pt; line-height:160%;']}), # for entertainment page title + dict(name='font', attrs={'color':['AA0000']}), # for column articles title + dict(attrs={'id':['newscontent']}), # entertainment and column page content + dict(attrs={'id':['newscontent01','newscontent02']}), + dict(attrs={'class':['photo']}), + dict(name='table', attrs={'width':['100%'], 'border':['0'], 'cellspacing':['5'], 'cellpadding':['0']}), # content in printed version of life.mingpao.com + dict(name='img', attrs={'width':['180'], 'alt':['按圖放大']}) # images for source from life.mingpao.com + ] + if __KeepImages__: + remove_tags = [dict(name='style'), + dict(attrs={'id':['newscontent135']}), # for the finance page from mpfinance.com + dict(name='font', attrs={'size':['2'], 'color':['666666']}), # article date in life.mingpao.com article + #dict(name='table') # for content fetched from life.mingpao.com + ] + else: + remove_tags = [dict(name='style'), + dict(attrs={'id':['newscontent135']}), # for the finance page from mpfinance.com + dict(name='font', attrs={'size':['2'], 'color':['666666']}), # article date in life.mingpao.com article + dict(name='img'), + #dict(name='table') # for content fetched from life.mingpao.com + ] + remove_attributes = ['width'] + preprocess_regexps = [ + (re.compile(r'
', re.DOTALL|re.IGNORECASE), + lambda match: '

'), + (re.compile(r'

', re.DOTALL|re.IGNORECASE), + lambda match: ''), + (re.compile(r'

', re.DOTALL|re.IGNORECASE), # for entertainment page + lambda match: ''), + # skip
after title in life.mingpao.com fetched article + (re.compile(r"

", re.DOTALL|re.IGNORECASE), + lambda match: "
"), + (re.compile(r"

", re.DOTALL|re.IGNORECASE), + lambda match: "") + ] + elif __Region__ == 'Vancouver': + title = 'Ming Pao - Vancouver' + description = 'Vancouver Chinese Newspaper (http://www.mingpaovan.com)' + category = 'Chinese, News, Vancouver' + extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}' + masthead_url = 'http://www.mingpaovan.com/image/mainlogo2_VAN2.gif' + keep_only_tags = [dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['1']}), + dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['3'], 'cellpadding':['3'], 'id':['tblContent3']}), + dict(name='table', attrs={'width':['180'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['0'], 'bgcolor':['F0F0F0']}), + ] + if __KeepImages__: + remove_tags = [dict(name='img', attrs={'src':['../../../image/magnifier.gif']})] # the magnifier icon + else: + remove_tags = [dict(name='img')] + remove_attributes = ['width'] + preprocess_regexps = [(re.compile(r' ', re.DOTALL|re.IGNORECASE), + lambda match: ''), + ] + elif __Region__ == 'Toronto': + title = 'Ming Pao - Toronto' + description = 'Toronto Chinese Newspaper (http://www.mingpaotor.com)' + category = 'Chinese, News, Toronto' + extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}' + masthead_url = 'http://www.mingpaotor.com/image/mainlogo2_TOR2.gif' + keep_only_tags = [dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['1']}), + dict(name='table', attrs={'width':['450'], 'border':['0'], 'cellspacing':['3'], 'cellpadding':['3'], 'id':['tblContent3']}), + dict(name='table', attrs={'width':['180'], 'border':['0'], 'cellspacing':['0'], 'cellpadding':['0'], 'bgcolor':['F0F0F0']}), + ] + if __KeepImages__: + remove_tags = [dict(name='img', attrs={'src':['../../../image/magnifier.gif']})] # the magnifier icon + else: + remove_tags = [dict(name='img')] + remove_attributes = ['width'] + preprocess_regexps = [(re.compile(r' ', re.DOTALL|re.IGNORECASE), + lambda match: ''), + ] + + oldest_article = 1 + max_articles_per_feed = 100 + __author__ = 'Eddie Lau' + publisher = 'MingPao' + remove_javascript = True + use_embedded_content = False + no_stylesheets = True + language = 'zh' + encoding = 'Big5-HKSCS' + recursions = 0 + conversion_options = {'linearize_tables':True} + timefmt = '' + + def image_url_processor(cls, baseurl, url): + # trick: break the url at the first occurance of digit, add an additional + # '_' at the front + # not working, may need to move this to preprocess_html() method +# minIdx = 10000 +# i0 = url.find('0') +# if i0 >= 0 and i0 < minIdx: +# minIdx = i0 +# i1 = url.find('1') +# if i1 >= 0 and i1 < minIdx: +# minIdx = i1 +# i2 = url.find('2') +# if i2 >= 0 and i2 < minIdx: +# minIdx = i2 +# i3 = url.find('3') +# if i3 >= 0 and i0 < minIdx: +# minIdx = i3 +# i4 = url.find('4') +# if i4 >= 0 and i4 < minIdx: +# minIdx = i4 +# i5 = url.find('5') +# if i5 >= 0 and i5 < minIdx: +# minIdx = i5 +# i6 = url.find('6') +# if i6 >= 0 and i6 < minIdx: +# minIdx = i6 +# i7 = url.find('7') +# if i7 >= 0 and i7 < minIdx: +# minIdx = i7 +# i8 = url.find('8') +# if i8 >= 0 and i8 < minIdx: +# minIdx = i8 +# i9 = url.find('9') +# if i9 >= 0 and i9 < minIdx: +# minIdx = i9 + return url + + def get_dtlocal(self): + dt_utc = datetime.datetime.utcnow() + if __Region__ == 'Hong Kong': + # convert UTC to local hk time - at HKT 4.30am, all news are available + dt_local = dt_utc + datetime.timedelta(8.0/24) - datetime.timedelta(4.5/24) + # dt_local = dt_utc.astimezone(pytz.timezone('Asia/Hong_Kong')) - datetime.timedelta(4.5/24) + elif __Region__ == 'Vancouver': + # convert UTC to local Vancouver time - at PST time 4.30am, all news are available + dt_local = dt_utc + datetime.timedelta(-8.0/24) - datetime.timedelta(4.5/24) + #dt_local = dt_utc.astimezone(pytz.timezone('America/Vancouver')) - datetime.timedelta(4.5/24) + elif __Region__ == 'Toronto': + # convert UTC to local Toronto time - at EST time 4.30am, all news are available + dt_local = dt_utc + datetime.timedelta(-5.0/24) - datetime.timedelta(4.5/24) + #dt_local = dt_utc.astimezone(pytz.timezone('America/Toronto')) - datetime.timedelta(4.5/24) + return dt_local + + def get_fetchdate(self): + return self.get_dtlocal().strftime("%Y%m%d") + + def get_fetchformatteddate(self): + return self.get_dtlocal().strftime("%Y-%m-%d") + + def get_fetchday(self): + return self.get_dtlocal().strftime("%d") + + def get_cover_url(self): + if __Region__ == 'Hong Kong': + cover = 'http://news.mingpao.com/' + self.get_fetchdate() + '/' + self.get_fetchdate() + '_' + self.get_fetchday() + 'gacov.jpg' + elif __Region__ == 'Vancouver': + cover = 'http://www.mingpaovan.com/ftp/News/' + self.get_fetchdate() + '/' + self.get_fetchday() + 'pgva1s.jpg' + elif __Region__ == 'Toronto': + cover = 'http://www.mingpaotor.com/ftp/News/' + self.get_fetchdate() + '/' + self.get_fetchday() + 'pgtas.jpg' + br = BasicNewsRecipe.get_browser() + try: + br.open(cover) + except: + cover = None + return cover + + def parse_index(self): + feeds = [] + dateStr = self.get_fetchdate() + + if __Region__ == 'Hong Kong': + if __UseLife__: + for title, url, keystr in [(u'\u8981\u805e Headline', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalga', 'nal'), + (u'\u6e2f\u805e Local', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalgb', 'nal'), + (u'\u6559\u80b2 Education', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalgf', 'nal'), + (u'\u793e\u8a55/\u7b46\u9663 Editorial', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr', 'nal'), + (u'\u8ad6\u58c7 Forum', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalfa', 'nal'), + (u'\u4e2d\u570b China', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalca', 'nal'), + (u'\u570b\u969b World', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalta', 'nal'), + (u'\u7d93\u6fdf Finance', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea', 'nal'), + (u'\u9ad4\u80b2 Sport', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalsp', 'nal'), + (u'\u5f71\u8996 Film/TV', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalma', 'nal'), + (u'\u5c08\u6b04 Columns', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn', 'ncl')]: + articles = self.parse_section2(url, keystr) + if articles: + feeds.append((title, articles)) + + for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'), + (u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + else: + for title, url in [(u'\u8981\u805e Headline', 'http://news.mingpao.com/' + dateStr + '/gaindex.htm'), + (u'\u6e2f\u805e Local', 'http://news.mingpao.com/' + dateStr + '/gbindex.htm'), + (u'\u6559\u80b2 Education', 'http://news.mingpao.com/' + dateStr + '/gfindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + # special- editorial + ed_articles = self.parse_ed_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr') + if ed_articles: + feeds.append((u'\u793e\u8a55/\u7b46\u9663 Editorial', ed_articles)) + + for title, url in [(u'\u8ad6\u58c7 Forum', 'http://news.mingpao.com/' + dateStr + '/faindex.htm'), + (u'\u4e2d\u570b China', 'http://news.mingpao.com/' + dateStr + '/caindex.htm'), + (u'\u570b\u969b World', 'http://news.mingpao.com/' + dateStr + '/taindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + # special - finance + #fin_articles = self.parse_fin_section('http://www.mpfinance.com/htm/Finance/' + dateStr + '/News/ea,eb,ecindex.htm') + fin_articles = self.parse_fin_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea') + if fin_articles: + feeds.append((u'\u7d93\u6fdf Finance', fin_articles)) + + for title, url in [('Tech News', 'http://news.mingpao.com/' + dateStr + '/naindex.htm'), + (u'\u9ad4\u80b2 Sport', 'http://news.mingpao.com/' + dateStr + '/spindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + # special - entertainment + ent_articles = self.parse_ent_section('http://ol.mingpao.com/cfm/star1.cfm') + if ent_articles: + feeds.append((u'\u5f71\u8996 Film/TV', ent_articles)) + + for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'), + (u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]: + articles = self.parse_section(url) + if articles: + feeds.append((title, articles)) + + + # special- columns + col_articles = self.parse_col_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn') + if col_articles: + feeds.append((u'\u5c08\u6b04 Columns', col_articles)) + elif __Region__ == 'Vancouver': + for title, url in [(u'\u8981\u805e Headline', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VAindex.htm'), + (u'\u52a0\u570b Canada', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VBindex.htm'), + (u'\u793e\u5340 Local', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VDindex.htm'), + (u'\u6e2f\u805e Hong Kong', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/HK-VGindex.htm'), + (u'\u570b\u969b World', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VTindex.htm'), + (u'\u4e2d\u570b China', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VCindex.htm'), + (u'\u7d93\u6fdf Economics', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VEindex.htm'), + (u'\u9ad4\u80b2 Sports', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VSindex.htm'), + (u'\u5f71\u8996 Film/TV', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/HK-MAindex.htm'), + (u'\u526f\u520a Supplements', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/WWindex.htm'),]: + articles = self.parse_section3(url, 'http://www.mingpaovan.com/') + if articles: + feeds.append((title, articles)) + elif __Region__ == 'Toronto': + for title, url in [(u'\u8981\u805e Headline', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TAindex.htm'), + (u'\u52a0\u570b Canada', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TDindex.htm'), + (u'\u793e\u5340 Local', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TFindex.htm'), + (u'\u4e2d\u570b China', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TCAindex.htm'), + (u'\u570b\u969b World', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TTAindex.htm'), + (u'\u6e2f\u805e Hong Kong', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/HK-GAindex.htm'), + (u'\u7d93\u6fdf Economics', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/THindex.htm'), + (u'\u9ad4\u80b2 Sports', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/TSindex.htm'), + (u'\u5f71\u8996 Film/TV', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/HK-MAindex.htm'), + (u'\u526f\u520a Supplements', 'http://www.mingpaotor.com/htm/News/' + dateStr + '/WWindex.htm'),]: + articles = self.parse_section3(url, 'http://www.mingpaotor.com/') + if articles: + feeds.append((title, articles)) + return feeds + + # parse from news.mingpao.com + def parse_section(self, url): + dateStr = self.get_fetchdate() + soup = self.index_to_soup(url) + divs = soup.findAll(attrs={'class': ['bullet','bullet_grey']}) + current_articles = [] + included_urls = [] + divs.reverse() + for i in divs: + a = i.find('a', href = True) + title = self.tag_to_string(a) + url = a.get('href', False) + url = 'http://news.mingpao.com/' + dateStr + '/' +url + if url not in included_urls and url.rfind('Redirect') == -1: + current_articles.append({'title': title, 'url': url, 'description':'', 'date':''}) + included_urls.append(url) + current_articles.reverse() + return current_articles + + # parse from life.mingpao.com + def parse_section2(self, url, keystr): + self.get_fetchdate() + soup = self.index_to_soup(url) + a = soup.findAll('a', href=True) + a.reverse() + current_articles = [] + included_urls = [] + for i in a: + title = self.tag_to_string(i) + url = 'http://life.mingpao.com/cfm/' + i.get('href', False) + if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind(keystr) == -1): + url = url.replace('dailynews3.cfm', 'dailynews3a.cfm') # use printed version of the article + current_articles.append({'title': title, 'url': url, 'description': ''}) + included_urls.append(url) + current_articles.reverse() + return current_articles + + # parse from www.mingpaovan.com + def parse_section3(self, url, baseUrl): + self.get_fetchdate() + soup = self.index_to_soup(url) + divs = soup.findAll(attrs={'class': ['ListContentLargeLink']}) + current_articles = [] + included_urls = [] + divs.reverse() + for i in divs: + title = self.tag_to_string(i) + urlstr = i.get('href', False) + urlstr = baseUrl + '/' + urlstr.replace('../../../', '') + if urlstr not in included_urls: + current_articles.append({'title': title, 'url': urlstr, 'description': '', 'date': ''}) + included_urls.append(urlstr) + current_articles.reverse() + return current_articles + + def parse_ed_section(self, url): + self.get_fetchdate() + soup = self.index_to_soup(url) + a = soup.findAll('a', href=True) + a.reverse() + current_articles = [] + included_urls = [] + for i in a: + title = self.tag_to_string(i) + url = 'http://life.mingpao.com/cfm/' + i.get('href', False) + if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind('nal') == -1): + current_articles.append({'title': title, 'url': url, 'description': ''}) + included_urls.append(url) + current_articles.reverse() + return current_articles + + def parse_fin_section(self, url): + self.get_fetchdate() + soup = self.index_to_soup(url) + a = soup.findAll('a', href= True) + current_articles = [] + included_urls = [] + for i in a: + #url = 'http://www.mpfinance.com/cfm/' + i.get('href', False) + url = 'http://life.mingpao.com/cfm/' + i.get('href', False) + #if url not in included_urls and not url.rfind(dateStr) == -1 and url.rfind('index') == -1: + if url not in included_urls and (not url.rfind('txt') == -1) and (not url.rfind('nal') == -1): + title = self.tag_to_string(i) + current_articles.append({'title': title, 'url': url, 'description':''}) + included_urls.append(url) + return current_articles + + def parse_ent_section(self, url): + self.get_fetchdate() + soup = self.index_to_soup(url) + a = soup.findAll('a', href=True) + a.reverse() + current_articles = [] + included_urls = [] + for i in a: + title = self.tag_to_string(i) + url = 'http://ol.mingpao.com/cfm/' + i.get('href', False) + if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind('star') == -1): + current_articles.append({'title': title, 'url': url, 'description': ''}) + included_urls.append(url) + current_articles.reverse() + return current_articles + + def parse_col_section(self, url): + self.get_fetchdate() + soup = self.index_to_soup(url) + a = soup.findAll('a', href=True) + a.reverse() + current_articles = [] + included_urls = [] + for i in a: + title = self.tag_to_string(i) + url = 'http://life.mingpao.com/cfm/' + i.get('href', False) + if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind('ncl') == -1): + current_articles.append({'title': title, 'url': url, 'description': ''}) + included_urls.append(url) + current_articles.reverse() + return current_articles + + def preprocess_html(self, soup): + for item in soup.findAll(style=True): + del item['style'] + for item in soup.findAll(style=True): + del item['width'] + for item in soup.findAll(stype=True): + del item['absmiddle'] + return soup + + def create_opf(self, feeds, dir=None): + if dir is None: + dir = self.output_dir + if __UseChineseTitle__ == True: + if __Region__ == 'Hong Kong': + title = u'\u660e\u5831 (\u9999\u6e2f)' + elif __Region__ == 'Vancouver': + title = u'\u660e\u5831 (\u6eab\u54e5\u83ef)' + elif __Region__ == 'Toronto': + title = u'\u660e\u5831 (\u591a\u502b\u591a)' + else: + title = self.short_title() + # if not generating a periodical, force date to apply in title + if __MakePeriodical__ == False: + title = title + ' ' + self.get_fetchformatteddate() + if True: + mi = MetaInformation(title, [self.publisher]) + mi.publisher = self.publisher + mi.author_sort = self.publisher + if __MakePeriodical__ == True: + mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title() + else: + mi.publication_type = self.publication_type+':'+self.short_title() + #mi.timestamp = nowf() + mi.timestamp = self.get_dtlocal() + mi.comments = self.description + if not isinstance(mi.comments, unicode): + mi.comments = mi.comments.decode('utf-8', 'replace') + #mi.pubdate = nowf() + mi.pubdate = self.get_dtlocal() + opf_path = os.path.join(dir, 'index.opf') + ncx_path = os.path.join(dir, 'index.ncx') + opf = OPFCreator(dir, mi) + # Add mastheadImage entry to section + mp = getattr(self, 'masthead_path', None) + if mp is not None and os.access(mp, os.R_OK): + from calibre.ebooks.metadata.opf2 import Guide + ref = Guide.Reference(os.path.basename(self.masthead_path), os.getcwdu()) + ref.type = 'masthead' + ref.title = 'Masthead Image' + opf.guide.append(ref) + + manifest = [os.path.join(dir, 'feed_%d'%i) for i in range(len(feeds))] + manifest.append(os.path.join(dir, 'index.html')) + manifest.append(os.path.join(dir, 'index.ncx')) + + # Get cover + cpath = getattr(self, 'cover_path', None) + if cpath is None: + pf = open(os.path.join(dir, 'cover.jpg'), 'wb') + if self.default_cover(pf): + cpath = pf.name + if cpath is not None and os.access(cpath, os.R_OK): + opf.cover = cpath + manifest.append(cpath) + + # Get masthead + mpath = getattr(self, 'masthead_path', None) + if mpath is not None and os.access(mpath, os.R_OK): + manifest.append(mpath) + + opf.create_manifest_from_files_in(manifest) + for mani in opf.manifest: + if mani.path.endswith('.ncx'): + mani.id = 'ncx' + if mani.path.endswith('mastheadImage.jpg'): + mani.id = 'masthead-image' + entries = ['index.html'] + toc = TOC(base_path=dir) + self.play_order_counter = 0 + self.play_order_map = {} + + def feed_index(num, parent): + f = feeds[num] + for j, a in enumerate(f): + if getattr(a, 'downloaded', False): + adir = 'feed_%d/article_%d/'%(num, j) + auth = a.author + if not auth: + auth = None + desc = a.text_summary + if not desc: + desc = None + else: + desc = self.description_limiter(desc) + entries.append('%sindex.html'%adir) + po = self.play_order_map.get(entries[-1], None) + if po is None: + self.play_order_counter += 1 + po = self.play_order_counter + parent.add_item('%sindex.html'%adir, None, a.title if a.title else _('Untitled Article'), + play_order=po, author=auth, description=desc) + last = os.path.join(self.output_dir, ('%sindex.html'%adir).replace('/', os.sep)) + for sp in a.sub_pages: + prefix = os.path.commonprefix([opf_path, sp]) + relp = sp[len(prefix):] + entries.append(relp.replace(os.sep, '/')) + last = sp + + if os.path.exists(last): + with open(last, 'rb') as fi: + src = fi.read().decode('utf-8') + soup = BeautifulSoup(src) + body = soup.find('body') + if body is not None: + prefix = '/'.join('..'for i in range(2*len(re.findall(r'link\d+', last)))) + templ = self.navbar.generate(True, num, j, len(f), + not self.has_single_feed, + a.orig_url, self.publisher, prefix=prefix, + center=self.center_navbar) + elem = BeautifulSoup(templ.render(doctype='xhtml').decode('utf-8')).find('div') + body.insert(len(body.contents), elem) + with open(last, 'wb') as fi: + fi.write(unicode(soup).encode('utf-8')) + if len(feeds) == 0: + raise Exception('All feeds are empty, aborting.') + + if len(feeds) > 1: + for i, f in enumerate(feeds): + entries.append('feed_%d/index.html'%i) + po = self.play_order_map.get(entries[-1], None) + if po is None: + self.play_order_counter += 1 + po = self.play_order_counter + auth = getattr(f, 'author', None) + if not auth: + auth = None + desc = getattr(f, 'description', None) + if not desc: + desc = None + feed_index(i, toc.add_item('feed_%d/index.html'%i, None, + f.title, play_order=po, description=desc, author=auth)) + + else: + entries.append('feed_%d/index.html'%0) + feed_index(0, toc) + + for i, p in enumerate(entries): + entries[i] = os.path.join(dir, p.replace('/', os.sep)) + opf.create_spine(entries) + opf.set_toc(toc) + + with nested(open(opf_path, 'wb'), open(ncx_path, 'wb')) as (opf_file, ncx_file): + opf.render(opf_file, ncx_file) + diff --git a/recipes/wprost.recipe b/recipes/wprost.recipe index b317571981..b271665125 100644 --- a/recipes/wprost.recipe +++ b/recipes/wprost.recipe @@ -2,90 +2,92 @@ __license__ = 'GPL v3' __copyright__ = '2010, matek09, matek09@gmail.com' +__copyright__ = 'Modified 2011, Mariusz Wolek ' from calibre.web.feeds.news import BasicNewsRecipe import re class Wprost(BasicNewsRecipe): - EDITION = 0 - FIND_LAST_FULL_ISSUE = True - EXCLUDE_LOCKED = True - ICO_BLOCKED = 'http://www.wprost.pl/G/icons/ico_blocked.gif' + EDITION = 0 + FIND_LAST_FULL_ISSUE = True + EXCLUDE_LOCKED = True + ICO_BLOCKED = 'http://www.wprost.pl/G/icons/ico_blocked.gif' - title = u'Wprost' - __author__ = 'matek09' - description = 'Weekly magazine' - encoding = 'ISO-8859-2' - no_stylesheets = True - language = 'pl' - remove_javascript = True + title = u'Wprost' + __author__ = 'matek09' + description = 'Weekly magazine' + encoding = 'ISO-8859-2' + no_stylesheets = True + language = 'pl' + remove_javascript = True - remove_tags_before = dict(dict(name = 'div', attrs = {'id' : 'print-layer'})) - remove_tags_after = dict(dict(name = 'div', attrs = {'id' : 'print-layer'})) + remove_tags_before = dict(dict(name = 'div', attrs = {'id' : 'print-layer'})) + remove_tags_after = dict(dict(name = 'div', attrs = {'id' : 'print-layer'})) - '''keep_only_tags =[] - keep_only_tags.append(dict(name = 'table', attrs = {'id' : 'title-table'})) - keep_only_tags.append(dict(name = 'div', attrs = {'class' : 'div-header'})) - keep_only_tags.append(dict(name = 'div', attrs = {'class' : 'div-content'})) - keep_only_tags.append(dict(name = 'div', attrs = {'class' : 'def element-autor'}))''' + '''keep_only_tags =[] + keep_only_tags.append(dict(name = 'table', attrs = {'id' : 'title-table'})) + keep_only_tags.append(dict(name = 'div', attrs = {'class' : 'div-header'})) + keep_only_tags.append(dict(name = 'div', attrs = {'class' : 'div-content'})) + keep_only_tags.append(dict(name = 'div', attrs = {'class' : 'def element-autor'}))''' - preprocess_regexps = [(re.compile(r'style="display: none;"'), lambda match: ''), - (re.compile(r'display: block;'), lambda match: '')] + preprocess_regexps = [(re.compile(r'style="display: none;"'), lambda match: ''), + (re.compile(r'display: block;'), lambda match: ''), + (re.compile(r'\\\<\/table\>'), lambda match: ''), + (re.compile(r'\'), lambda match: ''), + (re.compile(r'\'), lambda match: ''), + (re.compile(r'\
'), lambda match: '')] + remove_tags =[] + remove_tags.append(dict(name = 'div', attrs = {'class' : 'def element-date'})) + remove_tags.append(dict(name = 'div', attrs = {'class' : 'def silver'})) + remove_tags.append(dict(name = 'div', attrs = {'id' : 'content-main-column-right'})) - remove_tags =[] - remove_tags.append(dict(name = 'div', attrs = {'class' : 'def element-date'})) - remove_tags.append(dict(name = 'div', attrs = {'class' : 'def silver'})) - remove_tags.append(dict(name = 'div', attrs = {'id' : 'content-main-column-right'})) - - - extra_css = ''' - .div-header {font-size: x-small; font-weight: bold} - ''' + extra_css = ''' + .div-header {font-size: x-small; font-weight: bold} + ''' #h2 {font-size: x-large; font-weight: bold} - def is_blocked(self, a): - if a.findNextSibling('img') is None: - return False - else: - return True + def is_blocked(self, a): + if a.findNextSibling('img') is None: + return False + else: + return True - def find_last_issue(self): - soup = self.index_to_soup('http://www.wprost.pl/archiwum/') - a = 0 - if self.FIND_LAST_FULL_ISSUE: - ico_blocked = soup.findAll('img', attrs={'src' : self.ICO_BLOCKED}) - a = ico_blocked[-1].findNext('a', attrs={'title' : re.compile('Zobacz spis tre.ci')}) - else: - a = soup.find('a', attrs={'title' : re.compile('Zobacz spis tre.ci')}) - self.EDITION = a['href'].replace('/tygodnik/?I=', '') - self.cover_url = a.img['src'] + def find_last_issue(self): + soup = self.index_to_soup('http://www.wprost.pl/archiwum/') + a = 0 + if self.FIND_LAST_FULL_ISSUE: + ico_blocked = soup.findAll('img', attrs={'src' : self.ICO_BLOCKED}) + a = ico_blocked[-1].findNext('a', attrs={'title' : re.compile('Zobacz spis tre.ci')}) + else: + a = soup.find('a', attrs={'title' : re.compile('Zobacz spis tre.ci')}) + self.EDITION = a['href'].replace('/tygodnik/?I=', '') + self.cover_url = a.img['src'] - def parse_index(self): - self.find_last_issue() - soup = self.index_to_soup('http://www.wprost.pl/tygodnik/?I=' + self.EDITION) - feeds = [] - for main_block in soup.findAll(attrs={'class':'main-block-s3 s3-head head-red3'}): - articles = list(self.find_articles(main_block)) - if len(articles) > 0: - section = self.tag_to_string(main_block) - feeds.append((section, articles)) - return feeds - - def find_articles(self, main_block): - for a in main_block.findAllNext( attrs={'style':['','padding-top: 15px;']}): - if a.name in "td": - break - if self.EXCLUDE_LOCKED & self.is_blocked(a): - continue - yield { - 'title' : self.tag_to_string(a), - 'url' : 'http://www.wprost.pl' + a['href'], - 'date' : '', - 'description' : '' - } + def parse_index(self): + self.find_last_issue() + soup = self.index_to_soup('http://www.wprost.pl/tygodnik/?I=' + self.EDITION) + feeds = [] + for main_block in soup.findAll(attrs={'class':'main-block-s3 s3-head head-red3'}): + articles = list(self.find_articles(main_block)) + if len(articles) > 0: + section = self.tag_to_string(main_block) + feeds.append((section, articles)) + return feeds + def find_articles(self, main_block): + for a in main_block.findAllNext( attrs={'style':['','padding-top: 15px;']}): + if a.name in "td": + break + if self.EXCLUDE_LOCKED & self.is_blocked(a): + continue + yield { + 'title' : self.tag_to_string(a), + 'url' : 'http://www.wprost.pl' + a['href'], + 'date' : '', + 'description' : '' + } diff --git a/src/calibre/customize/builtins.py b/src/calibre/customize/builtins.py index e3b7bef5d8..333a5baaa4 100644 --- a/src/calibre/customize/builtins.py +++ b/src/calibre/customize/builtins.py @@ -1148,7 +1148,7 @@ plugins += [LookAndFeel, Behavior, Columns, Toolbar, Search, InputOptions, class StoreAmazonKindleStore(StoreBase): name = 'Amazon Kindle' description = u'Kindle books from Amazon.' - actual_plugin = 'calibre.gui2.store.amazon_plugin:AmazonKindleStore' + actual_plugin = 'calibre.gui2.store.stores.amazon_plugin:AmazonKindleStore' headquarters = 'US' formats = ['KINDLE'] @@ -1158,7 +1158,7 @@ class StoreAmazonDEKindleStore(StoreBase): name = 'Amazon DE Kindle' author = 'Charles Haley' description = u'Kindle Bücher von Amazon.' - actual_plugin = 'calibre.gui2.store.amazon_de_plugin:AmazonDEKindleStore' + actual_plugin = 'calibre.gui2.store.stores.amazon_de_plugin:AmazonDEKindleStore' headquarters = 'DE' formats = ['KINDLE'] @@ -1168,7 +1168,7 @@ class StoreAmazonUKKindleStore(StoreBase): name = 'Amazon UK Kindle' author = 'Charles Haley' description = u'Kindle books from Amazon\'s UK web site. Also, includes French language ebooks.' - actual_plugin = 'calibre.gui2.store.amazon_uk_plugin:AmazonUKKindleStore' + actual_plugin = 'calibre.gui2.store.stores.amazon_uk_plugin:AmazonUKKindleStore' headquarters = 'UK' formats = ['KINDLE'] @@ -1177,7 +1177,7 @@ class StoreAmazonUKKindleStore(StoreBase): class StoreArchiveOrgStore(StoreBase): name = 'Archive.org' description = u'An Internet library offering permanent access for researchers, historians, scholars, people with disabilities, and the general public to historical collections that exist in digital format.' - actual_plugin = 'calibre.gui2.store.archive_org_plugin:ArchiveOrgStore' + actual_plugin = 'calibre.gui2.store.stores.archive_org_plugin:ArchiveOrgStore' drm_free_only = True headquarters = 'US' @@ -1186,7 +1186,7 @@ class StoreArchiveOrgStore(StoreBase): class StoreBaenWebScriptionStore(StoreBase): name = 'Baen WebScription' description = u'Sci-Fi & Fantasy brought to you by Jim Baen.' - actual_plugin = 'calibre.gui2.store.baen_webscription_plugin:BaenWebScriptionStore' + actual_plugin = 'calibre.gui2.store.stores.baen_webscription_plugin:BaenWebScriptionStore' drm_free_only = True headquarters = 'US' @@ -1195,7 +1195,7 @@ class StoreBaenWebScriptionStore(StoreBase): class StoreBNStore(StoreBase): name = 'Barnes and Noble' description = u'The world\'s largest book seller. As the ultimate destination for book lovers, Barnes & Noble.com offers an incredible array of content.' - actual_plugin = 'calibre.gui2.store.bn_plugin:BNStore' + actual_plugin = 'calibre.gui2.store.stores.bn_plugin:BNStore' headquarters = 'US' formats = ['NOOK'] @@ -1205,7 +1205,7 @@ class StoreBeamEBooksDEStore(StoreBase): name = 'Beam EBooks DE' author = 'Charles Haley' description = u'Bei uns finden Sie: Tausende deutschsprachige eBooks; Alle eBooks ohne hartes DRM; PDF, ePub und Mobipocket Format; Sofortige Verfügbarkeit - 24 Stunden am Tag; Günstige Preise; eBooks für viele Lesegeräte, PC,Mac und Smartphones; Viele Gratis eBooks' - actual_plugin = 'calibre.gui2.store.beam_ebooks_de_plugin:BeamEBooksDEStore' + actual_plugin = 'calibre.gui2.store.stores.beam_ebooks_de_plugin:BeamEBooksDEStore' drm_free_only = True headquarters = 'DE' @@ -1215,7 +1215,7 @@ class StoreBeamEBooksDEStore(StoreBase): class StoreBeWriteStore(StoreBase): name = 'BeWrite Books' description = u'Publishers of fine books. Highly selective and editorially driven. Does not offer: books for children or exclusively YA, erotica, swords-and-sorcery fantasy and space-opera-style science fiction. All other genres are represented.' - actual_plugin = 'calibre.gui2.store.bewrite_plugin:BeWriteStore' + actual_plugin = 'calibre.gui2.store.stores.bewrite_plugin:BeWriteStore' drm_free_only = True headquarters = 'US' @@ -1224,7 +1224,7 @@ class StoreBeWriteStore(StoreBase): class StoreDieselEbooksStore(StoreBase): name = 'Diesel eBooks' description = u'Instant access to over 2.4 million titles from hundreds of publishers including Harlequin, HarperCollins, John Wiley & Sons, McGraw-Hill, Simon & Schuster and Random House.' - actual_plugin = 'calibre.gui2.store.diesel_ebooks_plugin:DieselEbooksStore' + actual_plugin = 'calibre.gui2.store.stores.diesel_ebooks_plugin:DieselEbooksStore' headquarters = 'US' formats = ['EPUB', 'PDF'] @@ -1233,7 +1233,7 @@ class StoreDieselEbooksStore(StoreBase): class StoreEbookscomStore(StoreBase): name = 'eBooks.com' description = u'Sells books in multiple electronic formats in all categories. Technical infrastructure is cutting edge, robust and scalable, with servers in the US and Europe.' - actual_plugin = 'calibre.gui2.store.ebooks_com_plugin:EbookscomStore' + actual_plugin = 'calibre.gui2.store.stores.ebooks_com_plugin:EbookscomStore' headquarters = 'US' formats = ['EPUB', 'LIT', 'MOBI', 'PDF'] @@ -1243,7 +1243,7 @@ class StoreEPubBuyDEStore(StoreBase): name = 'EPUBBuy DE' author = 'Charles Haley' description = u'Bei EPUBBuy.com finden Sie ausschliesslich eBooks im weitverbreiteten EPUB-Format und ohne DRM. So haben Sie die freie Wahl, wo Sie Ihr eBook lesen: Tablet, eBook-Reader, Smartphone oder einfach auf Ihrem PC. So macht eBook-Lesen Spaß!' - actual_plugin = 'calibre.gui2.store.epubbuy_de_plugin:EPubBuyDEStore' + actual_plugin = 'calibre.gui2.store.stores.epubbuy_de_plugin:EPubBuyDEStore' drm_free_only = True headquarters = 'DE' @@ -1254,7 +1254,7 @@ class StoreEBookShoppeUKStore(StoreBase): name = 'ebookShoppe UK' author = u'Charles Haley' description = u'We made this website in an attempt to offer the widest range of UK eBooks possible across and as many formats as we could manage.' - actual_plugin = 'calibre.gui2.store.ebookshoppe_uk_plugin:EBookShoppeUKStore' + actual_plugin = 'calibre.gui2.store.stores.ebookshoppe_uk_plugin:EBookShoppeUKStore' headquarters = 'UK' formats = ['EPUB', 'PDF'] @@ -1263,7 +1263,7 @@ class StoreEBookShoppeUKStore(StoreBase): class StoreEHarlequinStore(StoreBase): name = 'eHarlequin' description = u'A global leader in series romance and one of the world\'s leading publishers of books for women. Offers women a broad range of reading from romance to bestseller fiction, from young adult novels to erotic literature, from nonfiction to fantasy, from African-American novels to inspirational romance, and more.' - actual_plugin = 'calibre.gui2.store.eharlequin_plugin:EHarlequinStore' + actual_plugin = 'calibre.gui2.store.stores.eharlequin_plugin:EHarlequinStore' headquarters = 'CA' formats = ['EPUB', 'PDF'] @@ -1272,7 +1272,7 @@ class StoreEHarlequinStore(StoreBase): class StoreEpubBudStore(StoreBase): name = 'ePub Bud' description = 'Well, it\'s pretty much just "YouTube for Children\'s eBooks. A not-for-profit organization devoted to brining self published childrens books to the world.' - actual_plugin = 'calibre.gui2.store.epubbud_plugin:EpubBudStore' + actual_plugin = 'calibre.gui2.store.stores.epubbud_plugin:EpubBudStore' drm_free_only = True headquarters = 'US' @@ -1281,7 +1281,7 @@ class StoreEpubBudStore(StoreBase): class StoreFeedbooksStore(StoreBase): name = 'Feedbooks' description = u'Feedbooks is a cloud publishing and distribution service, connected to a large ecosystem of reading systems and social networks. Provides a variety of genres from independent and classic books.' - actual_plugin = 'calibre.gui2.store.feedbooks_plugin:FeedbooksStore' + actual_plugin = 'calibre.gui2.store.stores.feedbooks_plugin:FeedbooksStore' headquarters = 'FR' formats = ['EPUB', 'MOBI', 'PDF'] @@ -1290,7 +1290,7 @@ class StoreFoylesUKStore(StoreBase): name = 'Foyles UK' author = 'Charles Haley' description = u'Foyles of London\'s ebook store. Provides extensive range covering all subjects.' - actual_plugin = 'calibre.gui2.store.foyles_uk_plugin:FoylesUKStore' + actual_plugin = 'calibre.gui2.store.stores.foyles_uk_plugin:FoylesUKStore' headquarters = 'UK' formats = ['EPUB', 'PDF'] @@ -1300,7 +1300,7 @@ class StoreGandalfStore(StoreBase): name = 'Gandalf' author = u'Tomasz Długosz' description = u'Księgarnia internetowa Gandalf.' - actual_plugin = 'calibre.gui2.store.gandalf_plugin:GandalfStore' + actual_plugin = 'calibre.gui2.store.stores.gandalf_plugin:GandalfStore' headquarters = 'PL' formats = ['EPUB', 'PDF'] @@ -1308,7 +1308,7 @@ class StoreGandalfStore(StoreBase): class StoreGoogleBooksStore(StoreBase): name = 'Google Books' description = u'Google Books' - actual_plugin = 'calibre.gui2.store.google_books_plugin:GoogleBooksStore' + actual_plugin = 'calibre.gui2.store.stores.google_books_plugin:GoogleBooksStore' headquarters = 'US' formats = ['EPUB', 'PDF', 'TXT'] @@ -1316,7 +1316,7 @@ class StoreGoogleBooksStore(StoreBase): class StoreGutenbergStore(StoreBase): name = 'Project Gutenberg' description = u'The first producer of free ebooks. Free in the United States because their copyright has expired. They may not be free of copyright in other countries. Readers outside of the United States must check the copyright laws of their countries before downloading or redistributing our ebooks.' - actual_plugin = 'calibre.gui2.store.gutenberg_plugin:GutenbergStore' + actual_plugin = 'calibre.gui2.store.stores.gutenberg_plugin:GutenbergStore' drm_free_only = True headquarters = 'US' @@ -1325,7 +1325,7 @@ class StoreGutenbergStore(StoreBase): class StoreKoboStore(StoreBase): name = 'Kobo' description = u'With over 2.3 million eBooks to browse we have engaged readers in over 200 countries in Kobo eReading. Our eBook listings include New York Times Bestsellers, award winners, classics and more!' - actual_plugin = 'calibre.gui2.store.kobo_plugin:KoboStore' + actual_plugin = 'calibre.gui2.store.stores.kobo_plugin:KoboStore' headquarters = 'CA' formats = ['EPUB'] @@ -1335,7 +1335,7 @@ class StoreLegimiStore(StoreBase): name = 'Legimi' author = u'Tomasz Długosz' description = u'Tanie oraz darmowe ebooki, egazety i blogi w formacie EPUB, wprost na Twój e-czytnik, iPhone, iPad, Android i komputer' - actual_plugin = 'calibre.gui2.store.legimi_plugin:LegimiStore' + actual_plugin = 'calibre.gui2.store.stores.legimi_plugin:LegimiStore' headquarters = 'PL' formats = ['EPUB'] @@ -1344,7 +1344,7 @@ class StoreLibreDEStore(StoreBase): name = 'Libri DE' author = 'Charles Haley' description = u'Sicher Bücher, Hörbücher und Downloads online bestellen.' - actual_plugin = 'calibre.gui2.store.libri_de_plugin:LibreDEStore' + actual_plugin = 'calibre.gui2.store.stores.libri_de_plugin:LibreDEStore' headquarters = 'DE' formats = ['EPUB', 'PDF'] @@ -1353,7 +1353,7 @@ class StoreLibreDEStore(StoreBase): class StoreManyBooksStore(StoreBase): name = 'ManyBooks' description = u'Public domain and creative commons works from many sources.' - actual_plugin = 'calibre.gui2.store.manybooks_plugin:ManyBooksStore' + actual_plugin = 'calibre.gui2.store.stores.manybooks_plugin:ManyBooksStore' drm_free_only = True headquarters = 'US' @@ -1362,7 +1362,7 @@ class StoreManyBooksStore(StoreBase): class StoreMobileReadStore(StoreBase): name = 'MobileRead' description = u'Ebooks handcrafted with the utmost care.' - actual_plugin = 'calibre.gui2.store.mobileread.mobileread_plugin:MobileReadStore' + actual_plugin = 'calibre.gui2.store.stores.mobileread.mobileread_plugin:MobileReadStore' drm_free_only = True headquarters = 'CH' @@ -1372,16 +1372,24 @@ class StoreNextoStore(StoreBase): name = 'Nexto' author = u'Tomasz Długosz' description = u'Największy w Polsce sklep internetowy z audiobookami mp3, ebookami pdf oraz prasą do pobrania on-line.' - actual_plugin = 'calibre.gui2.store.nexto_plugin:NextoStore' + actual_plugin = 'calibre.gui2.store.stores.nexto_plugin:NextoStore' headquarters = 'PL' formats = ['EPUB', 'PDF'] affiliate = True +class StoreOpenBooksStore(StoreBase): + name = 'Open Books' + description = u'Comprehensive listing of DRM free ebooks from a variety of sources provided by users of calibre.' + actual_plugin = 'calibre.gui2.store.stores.open_books_plugin:OpenBooksStore' + + drm_free_only = True + headquarters = 'US' + class StoreOpenLibraryStore(StoreBase): name = 'Open Library' description = u'One web page for every book ever published. The goal is to be a true online library. Over 20 million records from a variety of large catalogs as well as single contributions, with more on the way.' - actual_plugin = 'calibre.gui2.store.open_library_plugin:OpenLibraryStore' + actual_plugin = 'calibre.gui2.store.stores.open_library_plugin:OpenLibraryStore' drm_free_only = True headquarters = 'US' @@ -1390,7 +1398,7 @@ class StoreOpenLibraryStore(StoreBase): class StoreOReillyStore(StoreBase): name = 'OReilly' description = u'Programming and tech ebooks from OReilly.' - actual_plugin = 'calibre.gui2.store.oreilly_plugin:OReillyStore' + actual_plugin = 'calibre.gui2.store.stores.oreilly_plugin:OReillyStore' drm_free_only = True headquarters = 'US' @@ -1399,7 +1407,7 @@ class StoreOReillyStore(StoreBase): class StorePragmaticBookshelfStore(StoreBase): name = 'Pragmatic Bookshelf' description = u'The Pragmatic Bookshelf\'s collection of programming and tech books avaliable as ebooks.' - actual_plugin = 'calibre.gui2.store.pragmatic_bookshelf_plugin:PragmaticBookshelfStore' + actual_plugin = 'calibre.gui2.store.stores.pragmatic_bookshelf_plugin:PragmaticBookshelfStore' drm_free_only = True headquarters = 'US' @@ -1408,7 +1416,7 @@ class StorePragmaticBookshelfStore(StoreBase): class StoreSmashwordsStore(StoreBase): name = 'Smashwords' description = u'An ebook publishing and distribution platform for ebook authors, publishers and readers. Covers many genres and formats.' - actual_plugin = 'calibre.gui2.store.smashwords_plugin:SmashwordsStore' + actual_plugin = 'calibre.gui2.store.stores.smashwords_plugin:SmashwordsStore' drm_free_only = True headquarters = 'US' @@ -1419,7 +1427,7 @@ class StoreVirtualoStore(StoreBase): name = 'Virtualo' author = u'Tomasz Długosz' description = u'Księgarnia internetowa, która oferuje bezpieczny i szeroki dostęp do książek w formie cyfrowej.' - actual_plugin = 'calibre.gui2.store.virtualo_plugin:VirtualoStore' + actual_plugin = 'calibre.gui2.store.stores.virtualo_plugin:VirtualoStore' headquarters = 'PL' formats = ['EPUB', 'PDF'] @@ -1428,7 +1436,7 @@ class StoreWaterstonesUKStore(StoreBase): name = 'Waterstones UK' author = 'Charles Haley' description = u'Waterstone\'s mission is to be the leading Bookseller on the High Street and online providing customers the widest choice, great value and expert advice from a team passionate about Bookselling.' - actual_plugin = 'calibre.gui2.store.waterstones_uk_plugin:WaterstonesUKStore' + actual_plugin = 'calibre.gui2.store.stores.waterstones_uk_plugin:WaterstonesUKStore' headquarters = 'UK' formats = ['EPUB', 'PDF'] @@ -1436,7 +1444,7 @@ class StoreWaterstonesUKStore(StoreBase): class StoreWeightlessBooksStore(StoreBase): name = 'Weightless Books' description = u'An independent DRM-free ebooksite devoted to ebooks of all sorts.' - actual_plugin = 'calibre.gui2.store.weightless_books_plugin:WeightlessBooksStore' + actual_plugin = 'calibre.gui2.store.stores.weightless_books_plugin:WeightlessBooksStore' drm_free_only = True headquarters = 'US' @@ -1446,7 +1454,7 @@ class StoreWHSmithUKStore(StoreBase): name = 'WH Smith UK' author = 'Charles Haley' description = u"Shop for savings on Books, discounted Magazine subscriptions and great prices on Stationery, Toys & Games" - actual_plugin = 'calibre.gui2.store.whsmith_uk_plugin:WHSmithUKStore' + actual_plugin = 'calibre.gui2.store.stores.whsmith_uk_plugin:WHSmithUKStore' headquarters = 'UK' formats = ['EPUB', 'PDF'] @@ -1454,7 +1462,7 @@ class StoreWHSmithUKStore(StoreBase): class StoreWizardsTowerBooksStore(StoreBase): name = 'Wizards Tower Books' description = u'A science fiction and fantasy publisher. Concentrates mainly on making out-of-print works available once more as e-books, and helping other small presses exploit the e-book market. Also publishes a small number of limited-print-run anthologies with a view to encouraging diversity in the science fiction and fantasy field.' - actual_plugin = 'calibre.gui2.store.wizards_tower_books_plugin:WizardsTowerBooksStore' + actual_plugin = 'calibre.gui2.store.stores.wizards_tower_books_plugin:WizardsTowerBooksStore' drm_free_only = True headquarters = 'UK' @@ -1464,7 +1472,7 @@ class StoreWoblinkStore(StoreBase): name = 'Woblink' author = u'Tomasz Długosz' description = u'Czytanie zdarza się wszędzie!' - actual_plugin = 'calibre.gui2.store.woblink_plugin:WoblinkStore' + actual_plugin = 'calibre.gui2.store.stores.woblink_plugin:WoblinkStore' headquarters = 'PL' formats = ['EPUB'] @@ -1473,7 +1481,7 @@ class StoreZixoStore(StoreBase): name = 'Zixo' author = u'Tomasz Długosz' description = u'Księgarnia z ebookami oraz książkami audio. Aby otwierać książki w formacie Zixo należy zainstalować program dostępny na stronie księgarni. Umożliwia on m.in. dodawanie zakładek i dostosowywanie rozmiaru czcionki.' - actual_plugin = 'calibre.gui2.store.zixo_plugin:ZixoStore' + actual_plugin = 'calibre.gui2.store.stores.zixo_plugin:ZixoStore' headquarters = 'PL' formats = ['PDF, ZIXO'] @@ -1504,6 +1512,7 @@ plugins += [ StoreManyBooksStore, StoreMobileReadStore, StoreNextoStore, + StoreOpenBooksStore, StoreOpenLibraryStore, StoreOReillyStore, StorePragmaticBookshelfStore, diff --git a/src/calibre/devices/android/driver.py b/src/calibre/devices/android/driver.py index 40cc3a34dc..2c840c644a 100644 --- a/src/calibre/devices/android/driver.py +++ b/src/calibre/devices/android/driver.py @@ -45,8 +45,11 @@ class ANDROID(USBMS): 0xfce : { 0xd12e : [0x0100]}, # Google - 0x18d1 : { 0x4e11 : [0x0100, 0x226, 0x227], 0x4e12: [0x0100, 0x226, - 0x227], 0x4e21: [0x0100, 0x226, 0x227], 0xb058: [0x0222]}, + 0x18d1 : { + 0x4e11 : [0x0100, 0x226, 0x227], + 0x4e12: [0x0100, 0x226, 0x227], + 0x4e21: [0x0100, 0x226, 0x227], + 0xb058: [0x0222, 0x226, 0x227]}, # Samsung 0x04e8 : { 0x681d : [0x0222, 0x0223, 0x0224, 0x0400], @@ -107,7 +110,7 @@ class ANDROID(USBMS): VENDOR_NAME = ['HTC', 'MOTOROLA', 'GOOGLE_', 'ANDROID', 'ACER', 'GT-I5700', 'SAMSUNG', 'DELL', 'LINUX', 'GOOGLE', 'ARCHOS', 'TELECHIP', 'HUAWEI', 'T-MOBILE', 'SEMC', 'LGE', 'NVIDIA', - 'GENERIC-', 'ZTE'] + 'GENERIC-', 'ZTE', 'MID'] WINDOWS_MAIN_MEM = ['ANDROID_PHONE', 'A855', 'A853', 'INC.NEXUS_ONE', '__UMS_COMPOSITE', '_MB200', 'MASS_STORAGE', '_-_CARD', 'SGH-I897', 'GT-I9000', 'FILE-STOR_GADGET', 'SGH-T959', 'SAMSUNG_ANDROID', @@ -116,7 +119,7 @@ class ANDROID(USBMS): 'IDEOS_TABLET', 'MYTOUCH_4G', 'UMS_COMPOSITE', 'SCH-I800_CARD', '7', 'A956', 'A955', 'A43', 'ANDROID_PLATFORM', 'TEGRA_2', 'MB860', 'MULTI-CARD', 'MID7015A', 'INCREDIBLE', 'A7EB', 'STREAK', - 'MB525'] + 'MB525', 'ANDROID2.3'] WINDOWS_CARD_A_MEM = ['ANDROID_PHONE', 'GT-I9000_CARD', 'SGH-I897', 'FILE-STOR_GADGET', 'SGH-T959', 'SAMSUNG_ANDROID', 'GT-P1000_CARD', 'A70S', 'A101IT', '7', 'INCREDIBLE', 'A7EB', 'SGH-T849_CARD', diff --git a/src/calibre/devices/iriver/driver.py b/src/calibre/devices/iriver/driver.py index 0ad540f8a3..21b188e031 100644 --- a/src/calibre/devices/iriver/driver.py +++ b/src/calibre/devices/iriver/driver.py @@ -20,11 +20,11 @@ class IRIVER_STORY(USBMS): FORMATS = ['epub', 'fb2', 'pdf', 'djvu', 'txt'] VENDOR_ID = [0x1006] - PRODUCT_ID = [0x4023, 0x4024, 0x4025] - BCD = [0x0323] + PRODUCT_ID = [0x4023, 0x4024, 0x4025, 0x4034] + BCD = [0x0323, 0x0326] VENDOR_NAME = 'IRIVER' - WINDOWS_MAIN_MEM = ['STORY', 'STORY_EB05', 'STORY_WI-FI'] + WINDOWS_MAIN_MEM = ['STORY', 'STORY_EB05', 'STORY_WI-FI', 'STORY_EB07'] WINDOWS_CARD_A_MEM = ['STORY', 'STORY_SD'] #OSX_MAIN_MEM = 'Kindle Internal Storage Media' diff --git a/src/calibre/devices/usbms/books.py b/src/calibre/devices/usbms/books.py index 731d3e2b49..4d726e5bde 100644 --- a/src/calibre/devices/usbms/books.py +++ b/src/calibre/devices/usbms/books.py @@ -14,7 +14,7 @@ from calibre.constants import preferred_encoding from calibre import isbytestring, force_unicode from calibre.utils.config import prefs, tweaks from calibre.utils.icu import strcmp -from calibre.utils.formatter import eval_formatter +from calibre.utils.formatter import EvalFormatter class Book(Metadata): def __init__(self, prefix, lpath, size=None, other=None): @@ -116,7 +116,7 @@ class CollectionsBookList(BookList): field_name = field_meta['name'] else: field_name = '' - cat_name = eval_formatter.safe_format( + cat_name = EvalFormatter().safe_format( fmt=tweaks['sony_collection_name_template'], kwargs={'category':field_name, 'value':field_value}, error_value='GET_CATEGORY', book=None) diff --git a/src/calibre/ebooks/comic/input.py b/src/calibre/ebooks/comic/input.py index 5203e3698d..f6833ca197 100755 --- a/src/calibre/ebooks/comic/input.py +++ b/src/calibre/ebooks/comic/input.py @@ -351,7 +351,9 @@ class ComicInput(InputFormatPlugin): comics = [] with CurrentDir(tdir): if not os.path.exists('comics.txt'): - raise ValueError('%s is not a valid comic collection' + raise ValueError(( + '%s is not a valid comic collection' + ' no comics.txt was found in the file') %stream.name) raw = open('comics.txt', 'rb').read() if raw.startswith(codecs.BOM_UTF16_BE): diff --git a/src/calibre/ebooks/compression/palmdoc.py b/src/calibre/ebooks/compression/palmdoc.py index 90dabcb5a8..6069777fab 100644 --- a/src/calibre/ebooks/compression/palmdoc.py +++ b/src/calibre/ebooks/compression/palmdoc.py @@ -17,6 +17,8 @@ def decompress_doc(data): return cPalmdoc.decompress(data) def compress_doc(data): + if not data: + return u'' return cPalmdoc.compress(data) def test(): diff --git a/src/calibre/ebooks/metadata/book/base.py b/src/calibre/ebooks/metadata/book/base.py index 382cb6c5a2..c3af8bea07 100644 --- a/src/calibre/ebooks/metadata/book/base.py +++ b/src/calibre/ebooks/metadata/book/base.py @@ -70,6 +70,7 @@ class SafeFormat(TemplateFormatter): return '' return v +# DEPRECATED. This is not thread safe. Do not use. composite_formatter = SafeFormat() class Metadata(object): @@ -110,6 +111,7 @@ class Metadata(object): # List of strings or [] self.author = list(authors) if authors else []# Needed for backward compatibility self.authors = list(authors) if authors else [] + self.formatter = SafeFormat() def is_null(self, field): ''' @@ -146,7 +148,7 @@ class Metadata(object): return val if val is None: d['#value#'] = 'RECURSIVE_COMPOSITE FIELD (Metadata) ' + field - val = d['#value#'] = composite_formatter.safe_format( + val = d['#value#'] = self.formatter.safe_format( d['display']['composite_template'], self, _('TEMPLATE ERROR'), @@ -423,11 +425,12 @@ class Metadata(object): ''' if not ops: return + formatter = SafeFormat() for op in ops: try: src = op[0] dest = op[1] - val = composite_formatter.safe_format\ + val = formatter.safe_format\ (src, other, 'PLUGBOARD TEMPLATE ERROR', other) if dest == 'tags': self.set(dest, [f.strip() for f in val.split(',') if f.strip()]) diff --git a/src/calibre/gui2/__init__.py b/src/calibre/gui2/__init__.py index 0b21502327..8dbc72ab98 100644 --- a/src/calibre/gui2/__init__.py +++ b/src/calibre/gui2/__init__.py @@ -7,12 +7,13 @@ from urllib import unquote from PyQt4.Qt import (QVariant, QFileInfo, QObject, SIGNAL, QBuffer, Qt, QByteArray, QTranslator, QCoreApplication, QThread, QEvent, QTimer, pyqtSignal, QDate, QDesktopServices, - QFileDialog, QFileIconProvider, + QFileDialog, QFileIconProvider, QSettings, QIcon, QApplication, QDialog, QUrl, QFont) ORG_NAME = 'KovidsBrain' APP_UID = 'libprs500' -from calibre.constants import islinux, iswindows, isbsd, isfrozen, isosx +from calibre.constants import (islinux, iswindows, isbsd, isfrozen, isosx, + config_dir) from calibre.utils.config import Config, ConfigProxy, dynamic, JSONConfig from calibre.utils.localization import set_qt_translator from calibre.ebooks.metadata import MetaInformation @@ -192,6 +193,11 @@ def _config(): # {{{ config = _config() # }}} +QSettings.setPath(QSettings.IniFormat, QSettings.UserScope, config_dir) +QSettings.setPath(QSettings.IniFormat, QSettings.SystemScope, + config_dir) +QSettings.setDefaultFormat(QSettings.IniFormat) + # Turn off DeprecationWarnings in windows GUI if iswindows: import warnings diff --git a/src/calibre/gui2/actions/choose_library.py b/src/calibre/gui2/actions/choose_library.py index 32460b6bec..f96a261790 100644 --- a/src/calibre/gui2/actions/choose_library.py +++ b/src/calibre/gui2/actions/choose_library.py @@ -5,7 +5,7 @@ __license__ = 'GPL v3' __copyright__ = '2010, Kovid Goyal ' __docformat__ = 'restructuredtext en' -import os, shutil +import os from functools import partial from PyQt4.Qt import QMenu, Qt, QInputDialog, QToolButton @@ -14,7 +14,7 @@ from calibre import isbytestring from calibre.constants import filesystem_encoding, iswindows from calibre.utils.config import prefs from calibre.gui2 import (gprefs, warning_dialog, Dispatcher, error_dialog, - question_dialog, info_dialog) + question_dialog, info_dialog, open_local_file) from calibre.library.database2 import LibraryDatabase2 from calibre.gui2.actions import InterfaceAction @@ -107,7 +107,7 @@ class ChooseLibraryAction(InterfaceAction): self.quick_menu_action = self.choose_menu.addMenu(self.quick_menu) self.rename_menu = QMenu(_('Rename library')) self.rename_menu_action = self.choose_menu.addMenu(self.rename_menu) - self.delete_menu = QMenu(_('Delete library')) + self.delete_menu = QMenu(_('Remove library')) self.delete_menu_action = self.choose_menu.addMenu(self.delete_menu) ac = self.create_action(spec=(_('Pick a random book'), 'catalog.png', @@ -252,22 +252,15 @@ class ChooseLibraryAction(InterfaceAction): def delete_requested(self, name, location): loc = location.replace('/', os.sep) - if not question_dialog(self.gui, _('Are you sure?'), - _('

WARNING

')+ - _('All files (not just ebooks) ' - 'from

%s

will be ' - 'permanently deleted. Are you sure?') % loc, - show_copy_button=False, default_yes=False): - return - exists = self.gui.library_view.model().db.exists_at(loc) - if exists: - try: - shutil.rmtree(loc, ignore_errors=True) - except: - pass self.stats.remove(location) self.build_menus() self.gui.iactions['Copy To Library'].build_menus() + info_dialog(self.gui, _('Library removed'), + _('The library %s has been removed from calibre. ' + 'The files remain on your computer, if you want ' + 'to delete them, you will have to do so manually.') % loc, + show=True) + open_local_file(loc) def backup_status(self, location): dirty_text = 'no' diff --git a/src/calibre/gui2/convert/regex_builder.py b/src/calibre/gui2/convert/regex_builder.py index f79e6df3fe..c7b03c9bb4 100644 --- a/src/calibre/gui2/convert/regex_builder.py +++ b/src/calibre/gui2/convert/regex_builder.py @@ -139,7 +139,12 @@ class RegexBuilder(QDialog, Ui_RegexBuilder): try: self.open_book(fpath) finally: - os.remove(fpath) + try: + os.remove(fpath) + except: + # Fails on windows if the input plugin for this format keeps the file open + # Happens for LIT files + pass return True def open_book(self, pathtoebook): @@ -148,7 +153,8 @@ class RegexBuilder(QDialog, Ui_RegexBuilder): text = [u''] preprocessor = HTMLPreProcessor(None, False) for path in self.iterator.spine: - html = open(path, 'rb').read().decode('utf-8', 'replace') + with open(path, 'rb') as f: + html = f.read().decode('utf-8', 'replace') html = preprocessor(html, get_preprocess_html=True) text.append(html) self.preview.setPlainText('\n---\n'.join(text)) diff --git a/src/calibre/gui2/dialogs/metadata_bulk.py b/src/calibre/gui2/dialogs/metadata_bulk.py index 7c7c78629c..22dfb98956 100644 --- a/src/calibre/gui2/dialogs/metadata_bulk.py +++ b/src/calibre/gui2/dialogs/metadata_bulk.py @@ -12,7 +12,7 @@ from PyQt4.Qt import Qt, QDialog, QGridLayout, QVBoxLayout, QFont, QLabel, \ from calibre.gui2.dialogs.metadata_bulk_ui import Ui_MetadataBulkDialog from calibre.gui2.dialogs.tag_editor import TagEditor from calibre.ebooks.metadata import string_to_authors, authors_to_string, title_sort -from calibre.ebooks.metadata.book.base import composite_formatter +from calibre.ebooks.metadata.book.base import SafeFormat from calibre.gui2.custom_column_widgets import populate_metadata_page from calibre.gui2 import error_dialog, ResizableDialog, UNDEFINED_QDATE, \ gprefs, question_dialog @@ -499,7 +499,7 @@ class MetadataBulkDialog(ResizableDialog, Ui_MetadataBulkDialog): def s_r_get_field(self, mi, field): if field: if field == '{template}': - v = composite_formatter.safe_format\ + v = SafeFormat().safe_format\ (unicode(self.s_r_template.text()), mi, _('S/R TEMPLATE ERROR'), mi) return [v] fm = self.db.metadata_for_field(field) diff --git a/src/calibre/gui2/dialogs/template_dialog.py b/src/calibre/gui2/dialogs/template_dialog.py index f78e7a7383..7d30f37bc1 100644 --- a/src/calibre/gui2/dialogs/template_dialog.py +++ b/src/calibre/gui2/dialogs/template_dialog.py @@ -11,7 +11,7 @@ from PyQt4.Qt import (Qt, QDialog, QDialogButtonBox, QSyntaxHighlighter, QFont, from calibre.gui2 import error_dialog from calibre.gui2.dialogs.template_dialog_ui import Ui_TemplateDialog from calibre.utils.formatter_functions import formatter_functions -from calibre.ebooks.metadata.book.base import composite_formatter, Metadata +from calibre.ebooks.metadata.book.base import SafeFormat, Metadata from calibre.library.coloring import (displayable_columns) @@ -270,7 +270,7 @@ class TemplateDialog(QDialog, Ui_TemplateDialog): self.highlighter.regenerate_paren_positions() self.text_cursor_changed() self.template_value.setText( - composite_formatter.safe_format(cur_text, self.mi, + SafeFormat().safe_format(cur_text, self.mi, _('EXCEPTION: '), self.mi)) def text_cursor_changed(self): diff --git a/src/calibre/gui2/init.py b/src/calibre/gui2/init.py index 874be6832f..67b4d5edd6 100644 --- a/src/calibre/gui2/init.py +++ b/src/calibre/gui2/init.py @@ -16,7 +16,7 @@ from calibre.constants import isosx, __appname__, preferred_encoding, \ from calibre.gui2 import config, is_widescreen, gprefs from calibre.gui2.library.views import BooksView, DeviceBooksView from calibre.gui2.widgets import Splitter -from calibre.gui2.tag_view import TagBrowserWidget +from calibre.gui2.tag_browser.ui import TagBrowserWidget from calibre.gui2.book_details import BookDetails from calibre.gui2.notify import get_notifier diff --git a/src/calibre/gui2/library/models.py b/src/calibre/gui2/library/models.py index 40d6e2b6cf..8cbc2e1979 100644 --- a/src/calibre/gui2/library/models.py +++ b/src/calibre/gui2/library/models.py @@ -14,7 +14,7 @@ from PyQt4.Qt import (QAbstractTableModel, Qt, pyqtSignal, QIcon, QImage, from calibre.gui2 import NONE, UNDEFINED_QDATE from calibre.utils.pyparsing import ParseException from calibre.ebooks.metadata import fmt_sidx, authors_to_string, string_to_authors -from calibre.ebooks.metadata.book.base import composite_formatter +from calibre.ebooks.metadata.book.base import SafeFormat from calibre.ptempfile import PersistentTemporaryFile from calibre.utils.config import tweaks, prefs from calibre.utils.date import dt_factory, qt_to_dt @@ -91,6 +91,7 @@ class BooksModel(QAbstractTableModel): # {{{ self.current_highlighted_idx = None self.highlight_only = False self.colors = frozenset([unicode(c) for c in QColor.colorNames()]) + self.formatter = SafeFormat() self.read_config() def change_alignment(self, colname, alignment): @@ -711,7 +712,7 @@ class BooksModel(QAbstractTableModel): # {{{ try: if mi is None: mi = self.db.get_metadata(id_, index_is_id=True) - color = composite_formatter.safe_format(fmt, mi, '', mi) + color = self.formatter.safe_format(fmt, mi, '', mi) if color in self.colors: color = QColor(color) if color.isValid(): diff --git a/src/calibre/gui2/metadata/basic_widgets.py b/src/calibre/gui2/metadata/basic_widgets.py index e00af37d33..2d6c79d0e3 100644 --- a/src/calibre/gui2/metadata/basic_widgets.py +++ b/src/calibre/gui2/metadata/basic_widgets.py @@ -173,9 +173,20 @@ class TitleSortEdit(TitleEdit): def auto_generate(self, *args): self.current_val = title_sort(self.title_edit.current_val) - self.title_edit.textChanged.disconnect() - self.textChanged.disconnect() - self.autogen_button.clicked.disconnect() + + def break_cycles(self): + try: + self.title_edit.textChanged.disconnect() + except: + pass + try: + self.textChanged.disconnect() + except: + pass + try: + self.autogen_button.clicked.disconnect() + except: + pass # }}} @@ -280,7 +291,10 @@ class AuthorsEdit(MultiCompleteComboBox): def break_cycles(self): self.db = self.dialog = None - self.manage_authors_signal.triggered.disconnect() + try: + self.manage_authors_signal.triggered.disconnect() + except: + pass class AuthorSortEdit(EnLineEdit): @@ -387,11 +401,26 @@ class AuthorSortEdit(EnLineEdit): def break_cycles(self): self.db = None - self.authors_edit.editTextChanged.disconnect() - self.textChanged.disconnect() - self.autogen_button.clicked.disconnect() - self.copy_a_to_as_action.triggered.disconnect() - self.copy_as_to_a_action.triggered.disconnect() + try: + self.authors_edit.editTextChanged.disconnect() + except: + pass + try: + self.textChanged.disconnect() + except: + pass + try: + self.autogen_button.clicked.disconnect() + except: + pass + try: + self.copy_a_to_as_action.triggered.disconnect() + except: + pass + try: + self.copy_as_to_a_action.triggered.disconnect() + except: + pass self.authors_edit = None # }}} @@ -519,9 +548,18 @@ class SeriesIndexEdit(QDoubleSpinBox): traceback.print_exc() def break_cycles(self): - self.series_edit.currentIndexChanged.disconnect() - self.series_edit.editTextChanged.disconnect() - self.series_edit.lineEdit().editingFinished.disconnect() + try: + self.series_edit.currentIndexChanged.disconnect() + except: + pass + try: + self.series_edit.editTextChanged.disconnect() + except: + pass + try: + self.series_edit.lineEdit().editingFinished.disconnect() + except: + pass self.db = self.series_edit = self.dialog = None # }}} @@ -898,7 +936,10 @@ class Cover(ImageView): # {{{ return True def break_cycles(self): - self.cover_changed.disconnect() + try: + self.cover_changed.disconnect() + except: + pass self.dialog = self._cdata = self.current_val = self.original_val = None # }}} diff --git a/src/calibre/gui2/store/archive_org_plugin.py b/src/calibre/gui2/store/archive_org_plugin.py deleted file mode 100644 index e8e96b3839..0000000000 --- a/src/calibre/gui2/store/archive_org_plugin.py +++ /dev/null @@ -1,89 +0,0 @@ -# -*- coding: utf-8 -*- - -from __future__ import (unicode_literals, division, absolute_import, print_function) - -__license__ = 'GPL 3' -__copyright__ = '2011, John Schember ' -__docformat__ = 'restructuredtext en' - -import urllib -from contextlib import closing - -from lxml import html - -from PyQt4.Qt import QUrl - -from calibre import browser, url_slash_cleaner -from calibre.gui2 import open_url -from calibre.gui2.store import StorePlugin -from calibre.gui2.store.basic_config import BasicStoreConfig -from calibre.gui2.store.search_result import SearchResult -from calibre.gui2.store.web_store_dialog import WebStoreDialog - -class ArchiveOrgStore(BasicStoreConfig, StorePlugin): - - def open(self, parent=None, detail_item=None, external=False): - url = 'http://www.archive.org/details/texts' - - if detail_item: - detail_item = url_slash_cleaner('http://www.archive.org' + detail_item) - - if external or self.config.get('open_external', False): - open_url(QUrl(url_slash_cleaner(detail_item if detail_item else url))) - else: - d = WebStoreDialog(self.gui, url, parent, detail_item) - d.setWindowTitle(self.name) - d.set_tags(self.config.get('tags', '')) - d.exec_() - - def search(self, query, max_results=10, timeout=60): - query = query + ' AND mediatype:texts' - url = 'http://www.archive.org/search.php?query=' + urllib.quote(query) - - br = browser() - - counter = max_results - with closing(br.open(url, timeout=timeout)) as f: - doc = html.fromstring(f.read()) - for data in doc.xpath('//td[@class="hitCell"]'): - if counter <= 0: - break - - id = ''.join(data.xpath('.//a[@class="titleLink"]/@href')) - if not id: - continue - - title = ''.join(data.xpath('.//a[@class="titleLink"]//text()')) - authors = data.xpath('.//text()') - if not authors: - continue - author = None - for a in authors: - if '-' in a: - author = a.replace('-', ' ').strip() - if author: - break - if not author: - continue - - counter -= 1 - - s = SearchResult() - s.title = title.strip() - s.author = author.strip() - s.price = '$0.00' - s.detail_item = id.strip() - s.drm = SearchResult.DRM_UNLOCKED - - yield s - - def get_details(self, search_result, timeout): - url = url_slash_cleaner('http://www.archive.org' + search_result.detail_item) - - br = browser() - with closing(br.open(url, timeout=timeout)) as nf: - idata = html.fromstring(nf.read()) - formats = ', '.join(idata.xpath('//p[@id="dl" and @class="content"]//a/text()')) - search_result.formats = formats.upper() - - return True diff --git a/src/calibre/gui2/store/config/chooser/models.py b/src/calibre/gui2/store/config/chooser/models.py index dbda367fae..d2c6bcd8d3 100644 --- a/src/calibre/gui2/store/config/chooser/models.py +++ b/src/calibre/gui2/store/config/chooser/models.py @@ -133,7 +133,7 @@ class Matches(QAbstractItemModel): return QVariant('

%s

' % result.description) elif col == 2: if result.drm_free_only: - return QVariant('

' + _('This store only distributes ebooks with DRM.') + '

') + return QVariant('

' + _('This store only distributes ebooks without DRM.') + '

') else: return QVariant('

' + _('This store distributes ebooks with DRM. It may have some titles without DRM, but you will need to check on a per title basis.') + '

') elif col == 3: diff --git a/src/calibre/gui2/store/declined.txt b/src/calibre/gui2/store/declined.txt index ada839662d..3e553f2dc8 100644 --- a/src/calibre/gui2/store/declined.txt +++ b/src/calibre/gui2/store/declined.txt @@ -1,9 +1,6 @@ This is a list of stores that objected, declined or asked not to be included in the store integration. -* Borders (http://www.borders.com/) -* Indigo (http://www.chapters.indigo.ca/) +* Borders (http://www.borders.com/). +* Indigo (http://www.chapters.indigo.ca/). * Libraria Rizzoli (http://libreriarizzoli.corriere.it/). - No reply with two attempts over 2 weeks -* WH Smith (http://www.whsmith.co.uk/) - Refused to permit signing up for the affiliate program \ No newline at end of file diff --git a/src/calibre/gui2/store/feedbooks_plugin.py b/src/calibre/gui2/store/feedbooks_plugin.py deleted file mode 100644 index 9b1f7f6574..0000000000 --- a/src/calibre/gui2/store/feedbooks_plugin.py +++ /dev/null @@ -1,106 +0,0 @@ -# -*- coding: utf-8 -*- - -from __future__ import (unicode_literals, division, absolute_import, print_function) - -__license__ = 'GPL 3' -__copyright__ = '2011, John Schember ' -__docformat__ = 'restructuredtext en' - -import urllib2 -from contextlib import closing - -from lxml import html - -from PyQt4.Qt import QUrl - -from calibre import browser, url_slash_cleaner -from calibre.gui2 import open_url -from calibre.gui2.store import StorePlugin -from calibre.gui2.store.basic_config import BasicStoreConfig -from calibre.gui2.store.search_result import SearchResult -from calibre.gui2.store.web_store_dialog import WebStoreDialog - -class FeedbooksStore(BasicStoreConfig, StorePlugin): - - def open(self, parent=None, detail_item=None, external=False): - url = 'http://m.feedbooks.com/' - ext_url = 'http://feedbooks.com/' - - if external or self.config.get('open_external', False): - if detail_item: - ext_url = ext_url + detail_item - open_url(QUrl(url_slash_cleaner(ext_url))) - else: - detail_url = None - if detail_item: - detail_url = url + detail_item - d = WebStoreDialog(self.gui, url, parent, detail_url) - d.setWindowTitle(self.name) - d.set_tags(self.config.get('tags', '')) - d.exec_() - - def search(self, query, max_results=10, timeout=60): - url = 'http://m.feedbooks.com/search?query=' + urllib2.quote(query) - - br = browser() - - counter = max_results - with closing(br.open(url, timeout=timeout)) as f: - doc = html.fromstring(f.read()) - for data in doc.xpath('//ul[@class="m-list"]//li'): - if counter <= 0: - break - data = html.fromstring(html.tostring(data)) - - id = '' - id_a = data.xpath('//a[@class="buy"]') - if id_a: - id = id_a[0].get('href', None) - id = id.split('/')[-2] - id = '/item/' + id - else: - id_a = data.xpath('//a[@class="download"]') - if id_a: - id = id_a[0].get('href', None) - id = id.split('/')[-1] - id = id.split('.')[0] - id = '/book/' + id - if not id: - continue - - title = ''.join(data.xpath('//h5//a/text()')) - author = ''.join(data.xpath('//h6//a/text()')) - price = ''.join(data.xpath('//a[@class="buy"]/text()')) - formats = 'EPUB' - if not price: - price = '$0.00' - formats = 'EPUB, MOBI, PDF' - cover_url = '' - cover_url_img = data.xpath('//img') - if cover_url_img: - cover_url = cover_url_img[0].get('src') - cover_url.split('?')[0] - - counter -= 1 - - s = SearchResult() - s.cover_url = cover_url - s.title = title.strip() - s.author = author.strip() - s.price = price.replace(' ', '').strip() - s.detail_item = id.strip() - s.formats = formats - - yield s - - def get_details(self, search_result, timeout): - url = 'http://m.feedbooks.com/' - - br = browser() - with closing(br.open(url_slash_cleaner(url + search_result.detail_item), timeout=timeout)) as nf: - idata = html.fromstring(nf.read()) - if idata.xpath('boolean(//div[contains(@class, "m-description-long")]//p[contains(., "DRM") or contains(b, "Protection")])'): - search_result.drm = SearchResult.DRM_LOCKED - else: - search_result.drm = SearchResult.DRM_UNLOCKED - return True diff --git a/src/calibre/gui2/store/opensearch_store.py b/src/calibre/gui2/store/opensearch_store.py new file mode 100644 index 0000000000..54fedbd002 --- /dev/null +++ b/src/calibre/gui2/store/opensearch_store.py @@ -0,0 +1,101 @@ +# -*- coding: utf-8 -*- + +from __future__ import (unicode_literals, division, absolute_import, print_function) + +__license__ = 'GPL 3' +__copyright__ = '2011, John Schember ' +__docformat__ = 'restructuredtext en' + +import mimetypes +import urllib +from contextlib import closing + +from lxml import etree + +from PyQt4.Qt import QUrl + +from calibre import browser +from calibre.gui2 import open_url +from calibre.gui2.store import StorePlugin +from calibre.gui2.store.search_result import SearchResult +from calibre.gui2.store.web_store_dialog import WebStoreDialog +from calibre.utils.opensearch.description import Description +from calibre.utils.opensearch.query import Query + +class OpenSearchStore(StorePlugin): + + open_search_url = '' + web_url = '' + + def open(self, parent=None, detail_item=None, external=False): + if not hasattr(self, 'web_url'): + return + + if external or self.config.get('open_external', False): + open_url(QUrl(detail_item if detail_item else self.web_url)) + else: + d = WebStoreDialog(self.gui, self.web_url, parent, detail_item) + d.setWindowTitle(self.name) + d.set_tags(self.config.get('tags', '')) + d.exec_() + + def search(self, query, max_results=10, timeout=60): + if not hasattr(self, 'open_search_url'): + return + + description = Description(self.open_search_url) + url_template = description.get_best_template() + if not url_template: + return + oquery = Query(url_template) + + # set up initial values + oquery.searchTerms = urllib.quote_plus(query) + oquery.count = max_results + url = oquery.url() + + counter = max_results + br = browser() + with closing(br.open(url, timeout=timeout)) as f: + doc = etree.fromstring(f.read()) + for data in doc.xpath('//*[local-name() = "entry"]'): + if counter <= 0: + break + + counter -= 1 + + s = SearchResult() + + s.detail_item = ''.join(data.xpath('./*[local-name() = "id"]/text()')).strip() + + for link in data.xpath('./*[local-name() = "link"]'): + rel = link.get('rel') + href = link.get('href') + type = link.get('type') + + if rel and href and type: + if rel in ('http://opds-spec.org/thumbnail', 'http://opds-spec.org/image/thumbnail'): + s.cover_url = href + elif rel == u'http://opds-spec.org/acquisition/buy': + s.detail_item = href + elif rel == u'http://opds-spec.org/acquisition': + if type: + ext = mimetypes.guess_extension(type) + if ext: + ext = ext[1:].upper().strip() + s.downloads[ext] = href + s.formats = ', '.join(s.downloads.keys()).strip() + + s.title = ' '.join(data.xpath('./*[local-name() = "title"]//text()')).strip() + s.author = ', '.join(data.xpath('./*[local-name() = "author"]//*[local-name() = "name"]//text()')).strip() + + price_e = data.xpath('.//*[local-name() = "price"][1]') + if price_e: + price_e = price_e[0] + currency_code = price_e.get('currencycode', '') + price = ''.join(price_e.xpath('.//text()')).strip() + s.price = currency_code + ' ' + price + s.price = s.price.strip() + + + yield s diff --git a/src/calibre/gui2/store/pragmatic_bookshelf_plugin.py b/src/calibre/gui2/store/pragmatic_bookshelf_plugin.py deleted file mode 100644 index f3803bbcea..0000000000 --- a/src/calibre/gui2/store/pragmatic_bookshelf_plugin.py +++ /dev/null @@ -1,84 +0,0 @@ -# -*- coding: utf-8 -*- - -from __future__ import (unicode_literals, division, absolute_import, print_function) - -__license__ = 'GPL 3' -__copyright__ = '2011, John Schember ' -__docformat__ = 'restructuredtext en' - -import urllib -from contextlib import closing - -from lxml import html - -from PyQt4.Qt import QUrl - -from calibre import browser, url_slash_cleaner -from calibre.gui2 import open_url -from calibre.gui2.store import StorePlugin -from calibre.gui2.store.basic_config import BasicStoreConfig -from calibre.gui2.store.search_result import SearchResult -from calibre.gui2.store.web_store_dialog import WebStoreDialog - -class PragmaticBookshelfStore(BasicStoreConfig, StorePlugin): - - def open(self, parent=None, detail_item=None, external=False): - url = 'http://pragprog.com/' - - if external or self.config.get('open_external', False): - open_url(QUrl(url_slash_cleaner(detail_item if detail_item else url))) - else: - d = WebStoreDialog(self.gui, url, parent, detail_item) - d.setWindowTitle(self.name) - d.set_tags(self.config.get('tags', '')) - d.exec_() - - def search(self, query, max_results=10, timeout=60): - ''' - OPDS based search. - - We really should get the catelog from http://pragprog.com/catalog.opds - and look for the application/opensearchdescription+xml entry. - Then get the opensearch description to get the search url and - format. However, we are going to be lazy and hard code it. - ''' - url = 'http://pragprog.com/catalog/search?q=' + urllib.quote_plus(query) - - br = browser() - - counter = max_results - with closing(br.open(url, timeout=timeout)) as f: - # Use html instead of etree as html allows us - # to ignore the namespace easily. - doc = html.fromstring(f.read()) - for data in doc.xpath('//entry'): - if counter <= 0: - break - - id = ''.join(data.xpath('.//link[@rel="http://opds-spec.org/acquisition/buy"]/@href')) - if not id: - continue - - price = ''.join(data.xpath('.//price/@currencycode')).strip() - price += ' ' - price += ''.join(data.xpath('.//price/text()')).strip() - if not price.strip(): - continue - - cover_url = ''.join(data.xpath('.//link[@rel="http://opds-spec.org/cover"]/@href')) - - title = ''.join(data.xpath('.//title/text()')) - author = ''.join(data.xpath('.//author//text()')) - - counter -= 1 - - s = SearchResult() - s.cover_url = cover_url - s.title = title.strip() - s.author = author.strip() - s.price = price.strip() - s.detail_item = id.strip() - s.drm = SearchResult.DRM_UNLOCKED - s.formats = 'EPUB, PDF, MOBI' - - yield s diff --git a/src/calibre/gui2/store/search/adv_search_builder.py b/src/calibre/gui2/store/search/adv_search_builder.py index cc89ca4eb7..127ac27acb 100644 --- a/src/calibre/gui2/store/search/adv_search_builder.py +++ b/src/calibre/gui2/store/search/adv_search_builder.py @@ -45,6 +45,7 @@ class AdvSearchBuilderDialog(QDialog, Ui_Dialog): self.author_box.setText('') self.price_box.setText('') self.format_box.setText('') + self.download_combo.setCurrentIndex(0) self.affiliate_combo.setCurrentIndex(0) def tokens(self, raw): @@ -119,6 +120,9 @@ class AdvSearchBuilderDialog(QDialog, Ui_Dialog): format = unicode(self.format_box.text()).strip() if format: ans.append('format:"' + self.mc + format + '"') + download = unicode(self.download_combo.currentText()).strip() + if download: + ans.append('download:' + download) affiliate = unicode(self.affiliate_combo.currentText()).strip() if affiliate: ans.append('affiliate:' + affiliate) diff --git a/src/calibre/gui2/store/search/adv_search_builder.ui b/src/calibre/gui2/store/search/adv_search_builder.ui index ab12dbbc00..02eb8f6aa1 100644 --- a/src/calibre/gui2/store/search/adv_search_builder.ui +++ b/src/calibre/gui2/store/search/adv_search_builder.ui @@ -226,7 +226,7 @@ - + @@ -244,7 +244,7 @@ - + Qt::Vertical @@ -283,14 +283,14 @@ - + Affiliate: - + @@ -309,6 +309,32 @@ + + + + Download: + + + + + + + + + + + + + true + + + + + false + + + + diff --git a/src/calibre/gui2/store/search/models.py b/src/calibre/gui2/store/search/models.py index e2c1a03e90..1fb0e5327b 100644 --- a/src/calibre/gui2/store/search/models.py +++ b/src/calibre/gui2/store/search/models.py @@ -33,7 +33,7 @@ class Matches(QAbstractItemModel): total_changed = pyqtSignal(int) - HEADERS = [_('Cover'), _('Title'), _('Price'), _('DRM'), _('Store'), ''] + HEADERS = [_('Cover'), _('Title'), _('Price'), _('DRM'), _('Store'), _('Download'), _('Affiliate')] HTML_COLS = (1, 4) def __init__(self, cover_thread_count=2, detail_thread_count=4): @@ -47,6 +47,8 @@ class Matches(QAbstractItemModel): Qt.SmoothTransformation) self.DONATE_ICON = QPixmap(I('donate.png')).scaledToHeight(16, Qt.SmoothTransformation) + self.DOWNLOAD_ICON = QPixmap(I('arrow-down.png')).scaledToHeight(16, + Qt.SmoothTransformation) # All matches. Used to determine the order to display # self.matches because the SearchFilter returns @@ -181,9 +183,11 @@ class Matches(QAbstractItemModel): elif result.drm == SearchResult.DRM_UNKNOWN: return QVariant(self.DRM_UNKNOWN_ICON) if col == 5: + if result.downloads: + return QVariant(self.DOWNLOAD_ICON) + if col == 6: if result.affiliate: return QVariant(self.DONATE_ICON) - return NONE elif role == Qt.ToolTipRole: if col == 1: return QVariant('

%s

' % result.title) @@ -199,6 +203,9 @@ class Matches(QAbstractItemModel): elif col == 4: return QVariant('

%s

' % result.formats) elif col == 5: + if result.downloads: + return QVariant('

' + _('The following formats can be downloaded directly: %s.') % ', '.join(result.downloads.keys()) + '

') + elif col == 6: if result.affiliate: return QVariant('

' + _('Buying from this store supports the calibre developer: %s.') % result.plugin_author + '

') elif role == Qt.SizeHintRole: @@ -221,6 +228,11 @@ class Matches(QAbstractItemModel): elif col == 4: text = result.store_name elif col == 5: + if result.downloads: + text = 'a' + else: + text = 'b' + elif col == 6: if result.affiliate: text = 'a' else: @@ -257,6 +269,8 @@ class SearchFilter(SearchQueryParser): 'author', 'authors', 'cover', + 'download', + 'downloads', 'drm', 'format', 'formats', @@ -282,6 +296,8 @@ class SearchFilter(SearchQueryParser): location = location.lower().strip() if location == 'authors': location = 'author' + elif location == 'downloads': + location = 'download' elif location == 'formats': location = 'format' @@ -308,12 +324,13 @@ class SearchFilter(SearchQueryParser): 'author': lambda x: x.author.lower(), 'cover': attrgetter('cover_url'), 'drm': attrgetter('drm'), + 'download': attrgetter('downloads'), 'format': attrgetter('formats'), 'price': lambda x: comparable_price(x.price), 'store': lambda x: x.store_name.lower(), 'title': lambda x: x.title.lower(), } - for x in ('author', 'format'): + for x in ('author', 'download', 'format'): q[x+'s'] = q[x] for sr in self.srs: for locvalue in locations: @@ -347,7 +364,7 @@ class SearchFilter(SearchQueryParser): matches.add(sr) continue # this is bool or treated as bool, so can't match below. - if locvalue in ('affiliate', 'drm'): + if locvalue in ('affiliate', 'drm', 'download', 'downloads'): continue try: ### Can't separate authors because comma is used for name sep and author sep diff --git a/src/calibre/gui2/store/search/results_view.py b/src/calibre/gui2/store/search/results_view.py index 91c067006e..3957e18ef6 100644 --- a/src/calibre/gui2/store/search/results_view.py +++ b/src/calibre/gui2/store/search/results_view.py @@ -6,13 +6,18 @@ __license__ = 'GPL 3' __copyright__ = '2011, John Schember ' __docformat__ = 'restructuredtext en' -from PyQt4.Qt import (QTreeView) +from functools import partial + +from PyQt4.Qt import (pyqtSignal, QMenu, QTreeView) from calibre.gui2.metadata.single_download import RichTextDelegate from calibre.gui2.store.search.models import Matches class ResultsView(QTreeView): + download_requested = pyqtSignal(object) + open_requested = pyqtSignal(object) + def __init__(self, *args): QTreeView.__init__(self,*args) @@ -24,3 +29,18 @@ class ResultsView(QTreeView): for i in self._model.HTML_COLS: self.setItemDelegateForColumn(i, self.rt_delegate) + def contextMenuEvent(self, event): + index = self.indexAt(event.pos()) + + if not index.isValid(): + return + + result = self.model().get_result(index) + + menu = QMenu() + da = menu.addAction(_('Download...'), partial(self.download_requested.emit, result)) + if not result.downloads: + da.setEnabled(False) + menu.addSeparator() + menu.addAction(_('Goto in store...'), partial(self.open_requested.emit, result)) + menu.exec_(event.globalPos()) diff --git a/src/calibre/gui2/store/search/search.py b/src/calibre/gui2/store/search/search.py index 7db4d1bbaf..fd20669f09 100644 --- a/src/calibre/gui2/store/search/search.py +++ b/src/calibre/gui2/store/search/search.py @@ -14,6 +14,7 @@ from PyQt4.Qt import (Qt, QDialog, QDialogButtonBox, QTimer, QCheckBox, QLabel, QComboBox) from calibre.gui2 import JSONConfig, info_dialog +from calibre.gui2.dialogs.choose_format import ChooseFormatDialog from calibre.gui2.progress_indicator import ProgressIndicator from calibre.gui2.store.config.chooser.chooser_widget import StoreChooserWidget from calibre.gui2.store.config.search.search_widget import StoreConfigWidget @@ -72,7 +73,9 @@ class SearchDialog(QDialog, Ui_Dialog): self.search.clicked.connect(self.do_search) self.checker.timeout.connect(self.get_results) self.progress_checker.timeout.connect(self.check_progress) - self.results_view.activated.connect(self.open_store) + self.results_view.activated.connect(self.result_item_activated) + self.results_view.download_requested.connect(self.download_book) + self.results_view.open_requested.connect(self.open_store) self.results_view.model().total_changed.connect(self.update_book_total) self.select_all_stores.clicked.connect(self.stores_select_all) self.select_invert_stores.clicked.connect(self.stores_select_invert) @@ -129,11 +132,15 @@ class SearchDialog(QDialog, Ui_Dialog): # Title / Author self.results_view.setColumnWidth(1,int(total*.40)) # Price - self.results_view.setColumnWidth(2,int(total*.20)) + self.results_view.setColumnWidth(2,int(total*.12)) # DRM self.results_view.setColumnWidth(3, int(total*.15)) # Store / Formats self.results_view.setColumnWidth(4, int(total*.25)) + # Download + self.results_view.setColumnWidth(5, 20) + # Affiliate + self.results_view.setColumnWidth(6, 20) def do_search(self): # Stop all running threads. @@ -183,7 +190,7 @@ class SearchDialog(QDialog, Ui_Dialog): query = re.sub(r'%s:"(?P[^\s"]+)"' % loc, '\g', query) query = query.replace('%s:' % loc, '') # Remove the prefix and search text. - for loc in ('cover', 'drm', 'format', 'formats', 'price', 'store'): + for loc in ('cover', 'download', 'downloads', 'drm', 'format', 'formats', 'price', 'store'): query = re.sub(r'%s:"[^"]"' % loc, '', query) query = re.sub(r'%s:[^\s]*' % loc, '', query) # Remove logic. @@ -330,8 +337,21 @@ class SearchDialog(QDialog, Ui_Dialog): def update_book_total(self, total): self.total.setText('%s' % total) - def open_store(self, index): + def result_item_activated(self, index): result = self.results_view.model().get_result(index) + + if result.downloads: + self.download_book(result) + else: + self.open_store(result) + + def download_book(self, result): + d = ChooseFormatDialog(self, _('Choose format to download to your library.'), result.downloads.keys()) + if d.exec_() == d.Accepted: + ext = d.format() + self.gui.download_ebook(result.downloads[ext]) + + def open_store(self, result): self.gui.istores[result.store_name].open(self, result.detail_item, self.open_external.isChecked()) def check_progress(self): diff --git a/src/calibre/gui2/store/search_result.py b/src/calibre/gui2/store/search_result.py index 7d6ac5acad..5cf71f011e 100644 --- a/src/calibre/gui2/store/search_result.py +++ b/src/calibre/gui2/store/search_result.py @@ -22,6 +22,9 @@ class SearchResult(object): self.detail_item = '' self.drm = None self.formats = '' + # key = format in upper case. + # value = url to download the file. + self.downloads = {} self.affiliate = False self.plugin_author = '' diff --git a/src/calibre/gui2/store/stores/__init__.py b/src/calibre/gui2/store/stores/__init__.py new file mode 100644 index 0000000000..db25b54b67 --- /dev/null +++ b/src/calibre/gui2/store/stores/__init__.py @@ -0,0 +1,3 @@ +''' +All store plugins are placed here. +''' \ No newline at end of file diff --git a/src/calibre/gui2/store/amazon_de_plugin.py b/src/calibre/gui2/store/stores/amazon_de_plugin.py similarity index 100% rename from src/calibre/gui2/store/amazon_de_plugin.py rename to src/calibre/gui2/store/stores/amazon_de_plugin.py diff --git a/src/calibre/gui2/store/amazon_plugin.py b/src/calibre/gui2/store/stores/amazon_plugin.py similarity index 100% rename from src/calibre/gui2/store/amazon_plugin.py rename to src/calibre/gui2/store/stores/amazon_plugin.py diff --git a/src/calibre/gui2/store/amazon_uk_plugin.py b/src/calibre/gui2/store/stores/amazon_uk_plugin.py similarity index 100% rename from src/calibre/gui2/store/amazon_uk_plugin.py rename to src/calibre/gui2/store/stores/amazon_uk_plugin.py diff --git a/src/calibre/gui2/store/stores/archive_org_plugin.py b/src/calibre/gui2/store/stores/archive_org_plugin.py new file mode 100644 index 0000000000..b2ea4e04f1 --- /dev/null +++ b/src/calibre/gui2/store/stores/archive_org_plugin.py @@ -0,0 +1,39 @@ +# -*- coding: utf-8 -*- + +from __future__ import (unicode_literals, division, absolute_import, print_function) + +__license__ = 'GPL 3' +__copyright__ = '2011, John Schember ' +__docformat__ = 'restructuredtext en' + + +from calibre.gui2.store.basic_config import BasicStoreConfig +from calibre.gui2.store.opensearch_store import OpenSearchStore +from calibre.gui2.store.search_result import SearchResult + +class ArchiveOrgStore(BasicStoreConfig, OpenSearchStore): + + open_search_url = 'http://bookserver.archive.org/catalog/opensearch.xml' + web_url = 'http://www.archive.org/details/texts' + + # http://bookserver.archive.org/catalog/ + + def search(self, query, max_results=10, timeout=60): + for s in OpenSearchStore.search(self, query, max_results, timeout): + s.detail_item = 'http://www.archive.org/details/' + s.detail_item.split(':')[-1] + s.price = '$0.00' + s.drm = SearchResult.DRM_UNLOCKED + yield s + + def get_details(self, search_result, timeout): + ''' + The opensearch feed only returns a subset of formats that are available. + We want to get a list of all formats that the user can get. + ''' + br = browser() + with closing(br.open(search_result.detail_item, timeout=timeout)) as nf: + idata = html.fromstring(nf.read()) + formats = ', '.join(idata.xpath('//p[@id="dl" and @class="content"]//a/text()')) + search_result.formats = formats.upper() + + return True diff --git a/src/calibre/gui2/store/baen_webscription_plugin.py b/src/calibre/gui2/store/stores/baen_webscription_plugin.py similarity index 100% rename from src/calibre/gui2/store/baen_webscription_plugin.py rename to src/calibre/gui2/store/stores/baen_webscription_plugin.py diff --git a/src/calibre/gui2/store/beam_ebooks_de_plugin.py b/src/calibre/gui2/store/stores/beam_ebooks_de_plugin.py similarity index 100% rename from src/calibre/gui2/store/beam_ebooks_de_plugin.py rename to src/calibre/gui2/store/stores/beam_ebooks_de_plugin.py diff --git a/src/calibre/gui2/store/bewrite_plugin.py b/src/calibre/gui2/store/stores/bewrite_plugin.py similarity index 100% rename from src/calibre/gui2/store/bewrite_plugin.py rename to src/calibre/gui2/store/stores/bewrite_plugin.py diff --git a/src/calibre/gui2/store/bn_plugin.py b/src/calibre/gui2/store/stores/bn_plugin.py similarity index 100% rename from src/calibre/gui2/store/bn_plugin.py rename to src/calibre/gui2/store/stores/bn_plugin.py diff --git a/src/calibre/gui2/store/diesel_ebooks_plugin.py b/src/calibre/gui2/store/stores/diesel_ebooks_plugin.py similarity index 100% rename from src/calibre/gui2/store/diesel_ebooks_plugin.py rename to src/calibre/gui2/store/stores/diesel_ebooks_plugin.py diff --git a/src/calibre/gui2/store/ebooks_com_plugin.py b/src/calibre/gui2/store/stores/ebooks_com_plugin.py similarity index 100% rename from src/calibre/gui2/store/ebooks_com_plugin.py rename to src/calibre/gui2/store/stores/ebooks_com_plugin.py diff --git a/src/calibre/gui2/store/ebookshoppe_uk_plugin.py b/src/calibre/gui2/store/stores/ebookshoppe_uk_plugin.py similarity index 100% rename from src/calibre/gui2/store/ebookshoppe_uk_plugin.py rename to src/calibre/gui2/store/stores/ebookshoppe_uk_plugin.py diff --git a/src/calibre/gui2/store/eharlequin_plugin.py b/src/calibre/gui2/store/stores/eharlequin_plugin.py similarity index 100% rename from src/calibre/gui2/store/eharlequin_plugin.py rename to src/calibre/gui2/store/stores/eharlequin_plugin.py diff --git a/src/calibre/gui2/store/stores/epubbud_plugin.py b/src/calibre/gui2/store/stores/epubbud_plugin.py new file mode 100644 index 0000000000..b4d642f62b --- /dev/null +++ b/src/calibre/gui2/store/stores/epubbud_plugin.py @@ -0,0 +1,27 @@ +# -*- coding: utf-8 -*- + +from __future__ import (unicode_literals, division, absolute_import, print_function) + +__license__ = 'GPL 3' +__copyright__ = '2011, John Schember ' +__docformat__ = 'restructuredtext en' + +from calibre.gui2.store.basic_config import BasicStoreConfig +from calibre.gui2.store.opensearch_store import OpenSearchStore +from calibre.gui2.store.search_result import SearchResult + +class EpubBudStore(BasicStoreConfig, OpenSearchStore): + + open_search_url = 'http://www.epubbud.com/feeds/opensearch.xml' + web_url = 'http://www.epubbud.com/' + + # http://www.epubbud.com/feeds/catalog.atom + + def search(self, query, max_results=10, timeout=60): + for s in OpenSearchStore.search(self, query, max_results, timeout): + s.price = '$0.00' + s.drm = SearchResult.DRM_UNLOCKED + s.formats = 'EPUB' + # Download links are broken for this store. + s.downloads = {} + yield s diff --git a/src/calibre/gui2/store/epubbuy_de_plugin.py b/src/calibre/gui2/store/stores/epubbuy_de_plugin.py similarity index 100% rename from src/calibre/gui2/store/epubbuy_de_plugin.py rename to src/calibre/gui2/store/stores/epubbuy_de_plugin.py diff --git a/src/calibre/gui2/store/stores/feedbooks_plugin.py b/src/calibre/gui2/store/stores/feedbooks_plugin.py new file mode 100644 index 0000000000..96d0a10dc7 --- /dev/null +++ b/src/calibre/gui2/store/stores/feedbooks_plugin.py @@ -0,0 +1,28 @@ +# -*- coding: utf-8 -*- + +from __future__ import (unicode_literals, division, absolute_import, print_function) + +__license__ = 'GPL 3' +__copyright__ = '2011, John Schember ' +__docformat__ = 'restructuredtext en' + +from calibre.gui2.store.basic_config import BasicStoreConfig +from calibre.gui2.store.opensearch_store import OpenSearchStore +from calibre.gui2.store.search_result import SearchResult + +class FeedbooksStore(BasicStoreConfig, OpenSearchStore): + + open_search_url = 'http://assets0.feedbooks.net/opensearch.xml?t=1253087147' + web_url = 'http://feedbooks.com/' + + # http://www.feedbooks.com/catalog + + def search(self, query, max_results=10, timeout=60): + for s in OpenSearchStore.search(self, query, max_results, timeout): + if s.downloads: + s.drm = SearchResult.DRM_UNLOCKED + s.price = '$0.00' + else: + s.drm = SearchResult.DRM_LOCKED + s.formats = 'EPUB' + yield s diff --git a/src/calibre/gui2/store/foyles_uk_plugin.py b/src/calibre/gui2/store/stores/foyles_uk_plugin.py similarity index 100% rename from src/calibre/gui2/store/foyles_uk_plugin.py rename to src/calibre/gui2/store/stores/foyles_uk_plugin.py diff --git a/src/calibre/gui2/store/gandalf_plugin.py b/src/calibre/gui2/store/stores/gandalf_plugin.py similarity index 100% rename from src/calibre/gui2/store/gandalf_plugin.py rename to src/calibre/gui2/store/stores/gandalf_plugin.py diff --git a/src/calibre/gui2/store/google_books_plugin.py b/src/calibre/gui2/store/stores/google_books_plugin.py similarity index 100% rename from src/calibre/gui2/store/google_books_plugin.py rename to src/calibre/gui2/store/stores/google_books_plugin.py diff --git a/src/calibre/gui2/store/gutenberg_plugin.py b/src/calibre/gui2/store/stores/gutenberg_plugin.py similarity index 100% rename from src/calibre/gui2/store/gutenberg_plugin.py rename to src/calibre/gui2/store/stores/gutenberg_plugin.py diff --git a/src/calibre/gui2/store/kobo_plugin.py b/src/calibre/gui2/store/stores/kobo_plugin.py similarity index 100% rename from src/calibre/gui2/store/kobo_plugin.py rename to src/calibre/gui2/store/stores/kobo_plugin.py diff --git a/src/calibre/gui2/store/legimi_plugin.py b/src/calibre/gui2/store/stores/legimi_plugin.py similarity index 100% rename from src/calibre/gui2/store/legimi_plugin.py rename to src/calibre/gui2/store/stores/legimi_plugin.py diff --git a/src/calibre/gui2/store/libri_de_plugin.py b/src/calibre/gui2/store/stores/libri_de_plugin.py similarity index 100% rename from src/calibre/gui2/store/libri_de_plugin.py rename to src/calibre/gui2/store/stores/libri_de_plugin.py diff --git a/src/calibre/gui2/store/manybooks_plugin.py b/src/calibre/gui2/store/stores/manybooks_plugin.py similarity index 95% rename from src/calibre/gui2/store/manybooks_plugin.py rename to src/calibre/gui2/store/stores/manybooks_plugin.py index efd8d21e68..829a97012f 100644 --- a/src/calibre/gui2/store/manybooks_plugin.py +++ b/src/calibre/gui2/store/stores/manybooks_plugin.py @@ -20,7 +20,7 @@ from calibre.gui2.store import StorePlugin from calibre.gui2.store.basic_config import BasicStoreConfig from calibre.gui2.store.search_result import SearchResult from calibre.gui2.store.web_store_dialog import WebStoreDialog - + class ManyBooksStore(BasicStoreConfig, StorePlugin): def open(self, parent=None, detail_item=None, external=False): @@ -29,7 +29,7 @@ class ManyBooksStore(BasicStoreConfig, StorePlugin): detail_url = None if detail_item: detail_url = url + detail_item - + if external or self.config.get('open_external', False): open_url(QUrl(url_slash_cleaner(detail_url if detail_url else url))) else: @@ -44,16 +44,16 @@ class ManyBooksStore(BasicStoreConfig, StorePlugin): # secondary titles. Google is also faster. # Using a google search so we can search on both fields at once. url = 'http://www.google.com/xhtml?q=site:manybooks.net+' + urllib.quote_plus(query) - + br = browser() - + counter = max_results with closing(br.open(url, timeout=timeout)) as f: doc = html.fromstring(f.read()) for data in doc.xpath('//div[@class="edewpi"]//div[@class="r ld"]'): if counter <= 0: break - + url = '' url_a = data.xpath('div[@class="jd"]/a') if url_a: @@ -65,13 +65,13 @@ class ManyBooksStore(BasicStoreConfig, StorePlugin): continue id = url.split('/')[-1] id = id.strip() - + url_a = html.fromstring(html.tostring(url_a)) heading = ''.join(url_a.xpath('//text()')) title, _, author = heading.rpartition('by ') author = author.split('-')[0] price = '$0.00' - + cover_url = '' mo = re.match('^\D+', id) if mo: @@ -79,10 +79,9 @@ class ManyBooksStore(BasicStoreConfig, StorePlugin): cover_name = cover_name.replace('etext', '') cover_id = id.split('.')[0] cover_url = 'http://www.manybooks.net/images/' + id[0] + '/' + cover_name + '/' + cover_id + '-thumb.jpg' - print(cover_url) counter -= 1 - + s = SearchResult() s.cover_url = cover_url s.title = title.strip() @@ -91,5 +90,5 @@ class ManyBooksStore(BasicStoreConfig, StorePlugin): s.detail_item = '/titles/' + id s.drm = SearchResult.DRM_UNLOCKED s.formts = 'EPUB, PDB (eReader, PalmDoc, zTXT, Plucker, iSilo), FB2, ZIP, AZW, MOBI, PRC, LIT, PKG, PDF, TXT, RB, RTF, LRF, TCR, JAR' - + yield s diff --git a/src/calibre/gui2/store/mobileread/__init__.py b/src/calibre/gui2/store/stores/mobileread/__init__.py similarity index 100% rename from src/calibre/gui2/store/mobileread/__init__.py rename to src/calibre/gui2/store/stores/mobileread/__init__.py diff --git a/src/calibre/gui2/store/mobileread/adv_search_builder.py b/src/calibre/gui2/store/stores/mobileread/adv_search_builder.py similarity index 97% rename from src/calibre/gui2/store/mobileread/adv_search_builder.py rename to src/calibre/gui2/store/stores/mobileread/adv_search_builder.py index 8c41f1924b..8bc7850b0f 100644 --- a/src/calibre/gui2/store/mobileread/adv_search_builder.py +++ b/src/calibre/gui2/store/stores/mobileread/adv_search_builder.py @@ -10,7 +10,7 @@ import re from PyQt4.Qt import (QDialog, QDialogButtonBox) -from calibre.gui2.store.mobileread.adv_search_builder_ui import Ui_Dialog +from calibre.gui2.store.stores.mobileread.adv_search_builder_ui import Ui_Dialog from calibre.library.caches import CONTAINS_MATCH, EQUALS_MATCH class AdvSearchBuilderDialog(QDialog, Ui_Dialog): diff --git a/src/calibre/gui2/store/mobileread/adv_search_builder.ui b/src/calibre/gui2/store/stores/mobileread/adv_search_builder.ui similarity index 100% rename from src/calibre/gui2/store/mobileread/adv_search_builder.ui rename to src/calibre/gui2/store/stores/mobileread/adv_search_builder.ui diff --git a/src/calibre/gui2/store/mobileread/cache_progress_dialog.py b/src/calibre/gui2/store/stores/mobileread/cache_progress_dialog.py similarity index 94% rename from src/calibre/gui2/store/mobileread/cache_progress_dialog.py rename to src/calibre/gui2/store/stores/mobileread/cache_progress_dialog.py index 71416d8680..de2cf50f84 100644 --- a/src/calibre/gui2/store/mobileread/cache_progress_dialog.py +++ b/src/calibre/gui2/store/stores/mobileread/cache_progress_dialog.py @@ -8,7 +8,7 @@ __docformat__ = 'restructuredtext en' from PyQt4.Qt import QDialog -from calibre.gui2.store.mobileread.cache_progress_dialog_ui import Ui_Dialog +from calibre.gui2.store.stores.mobileread.cache_progress_dialog_ui import Ui_Dialog class CacheProgressDialog(QDialog, Ui_Dialog): diff --git a/src/calibre/gui2/store/mobileread/cache_progress_dialog.ui b/src/calibre/gui2/store/stores/mobileread/cache_progress_dialog.ui similarity index 100% rename from src/calibre/gui2/store/mobileread/cache_progress_dialog.ui rename to src/calibre/gui2/store/stores/mobileread/cache_progress_dialog.ui diff --git a/src/calibre/gui2/store/mobileread/cache_update_thread.py b/src/calibre/gui2/store/stores/mobileread/cache_update_thread.py similarity index 100% rename from src/calibre/gui2/store/mobileread/cache_update_thread.py rename to src/calibre/gui2/store/stores/mobileread/cache_update_thread.py diff --git a/src/calibre/gui2/store/mobileread/mobileread_plugin.py b/src/calibre/gui2/store/stores/mobileread/mobileread_plugin.py similarity index 91% rename from src/calibre/gui2/store/mobileread/mobileread_plugin.py rename to src/calibre/gui2/store/stores/mobileread/mobileread_plugin.py index 4e11d62bbd..3ffb5c36a1 100644 --- a/src/calibre/gui2/store/mobileread/mobileread_plugin.py +++ b/src/calibre/gui2/store/stores/mobileread/mobileread_plugin.py @@ -15,10 +15,10 @@ from calibre.gui2.store import StorePlugin from calibre.gui2.store.basic_config import BasicStoreConfig from calibre.gui2.store.search_result import SearchResult from calibre.gui2.store.web_store_dialog import WebStoreDialog -from calibre.gui2.store.mobileread.models import SearchFilter -from calibre.gui2.store.mobileread.cache_progress_dialog import CacheProgressDialog -from calibre.gui2.store.mobileread.cache_update_thread import CacheUpdateThread -from calibre.gui2.store.mobileread.store_dialog import MobileReadStoreDialog +from calibre.gui2.store.stores.mobileread.models import SearchFilter +from calibre.gui2.store.stores.mobileread.cache_progress_dialog import CacheProgressDialog +from calibre.gui2.store.stores.mobileread.cache_update_thread import CacheUpdateThread +from calibre.gui2.store.stores.mobileread.store_dialog import MobileReadStoreDialog class MobileReadStore(BasicStoreConfig, StorePlugin): diff --git a/src/calibre/gui2/store/mobileread/models.py b/src/calibre/gui2/store/stores/mobileread/models.py similarity index 100% rename from src/calibre/gui2/store/mobileread/models.py rename to src/calibre/gui2/store/stores/mobileread/models.py diff --git a/src/calibre/gui2/store/mobileread/store_dialog.py b/src/calibre/gui2/store/stores/mobileread/store_dialog.py similarity index 93% rename from src/calibre/gui2/store/mobileread/store_dialog.py rename to src/calibre/gui2/store/stores/mobileread/store_dialog.py index 749f96a614..ff7fea090f 100644 --- a/src/calibre/gui2/store/mobileread/store_dialog.py +++ b/src/calibre/gui2/store/stores/mobileread/store_dialog.py @@ -9,9 +9,9 @@ __docformat__ = 'restructuredtext en' from PyQt4.Qt import (Qt, QDialog, QIcon, QComboBox) -from calibre.gui2.store.mobileread.adv_search_builder import AdvSearchBuilderDialog -from calibre.gui2.store.mobileread.models import BooksModel -from calibre.gui2.store.mobileread.store_dialog_ui import Ui_Dialog +from calibre.gui2.store.stores.mobileread.adv_search_builder import AdvSearchBuilderDialog +from calibre.gui2.store.stores.mobileread.models import BooksModel +from calibre.gui2.store.stores.mobileread.store_dialog_ui import Ui_Dialog class MobileReadStoreDialog(QDialog, Ui_Dialog): diff --git a/src/calibre/gui2/store/mobileread/store_dialog.ui b/src/calibre/gui2/store/stores/mobileread/store_dialog.ui similarity index 100% rename from src/calibre/gui2/store/mobileread/store_dialog.ui rename to src/calibre/gui2/store/stores/mobileread/store_dialog.ui diff --git a/src/calibre/gui2/store/nexto_plugin.py b/src/calibre/gui2/store/stores/nexto_plugin.py similarity index 100% rename from src/calibre/gui2/store/nexto_plugin.py rename to src/calibre/gui2/store/stores/nexto_plugin.py diff --git a/src/calibre/gui2/store/epubbud_plugin.py b/src/calibre/gui2/store/stores/open_books_plugin.py similarity index 58% rename from src/calibre/gui2/store/epubbud_plugin.py rename to src/calibre/gui2/store/stores/open_books_plugin.py index d6193f6ae0..2e6c438d9d 100644 --- a/src/calibre/gui2/store/epubbud_plugin.py +++ b/src/calibre/gui2/store/stores/open_books_plugin.py @@ -20,59 +20,56 @@ from calibre.gui2.store.basic_config import BasicStoreConfig from calibre.gui2.store.search_result import SearchResult from calibre.gui2.store.web_store_dialog import WebStoreDialog -class EpubBudStore(BasicStoreConfig, StorePlugin): +class OpenBooksStore(BasicStoreConfig, StorePlugin): def open(self, parent=None, detail_item=None, external=False): - url = 'http://epubbud.com/' + url = 'http://drmfree.calibre-ebook.com/' if external or self.config.get('open_external', False): open_url(QUrl(url_slash_cleaner(detail_item if detail_item else url))) else: - d = WebStoreDialog(self.gui, url, parent, detail_item) + d = WebStoreDialog(self.gui, self.url, parent, detail_item) d.setWindowTitle(self.name) d.set_tags(self.config.get('tags', '')) d.exec_() def search(self, query, max_results=10, timeout=60): - ''' - OPDS based search. - - We really should get the catelog from http://pragprog.com/catalog.opds - and look for the application/opensearchdescription+xml entry. - Then get the opensearch description to get the search url and - format. However, we are going to be lazy and hard code it. - ''' - url = 'http://www.epubbud.com/search.php?format=atom&q=' + urllib.quote_plus(query) - + url = 'http://drmfree.calibre-ebook.com/search/?q=' + urllib.quote_plus(query) + br = browser() - + counter = max_results with closing(br.open(url, timeout=timeout)) as f: - # Use html instead of etree as html allows us - # to ignore the namespace easily. doc = html.fromstring(f.read()) - for data in doc.xpath('//entry'): + for data in doc.xpath('//ul[@id="object_list"]//li'): if counter <= 0: break - - id = ''.join(data.xpath('.//id/text()')) + + id = ''.join(data.xpath('.//div[@class="links"]/a[1]/@href')) + id = id.strip() if not id: continue - cover_url = ''.join(data.xpath('.//link[@rel="http://opds-spec.org/thumbnail"]/@href')) - - title = u''.join(data.xpath('.//title/text()')) - author = u''.join(data.xpath('.//author/name/text()')) + cover_url = ''.join(data.xpath('.//div[@class="cover"]/img/@src')) + + price = ''.join(data.xpath('.//div[@class="price"]/text()')) + a, b, price = price.partition('Price:') + price = price.strip() + if not price: + continue + + title = ''.join(data.xpath('.//div/strong/text()')) + author = ''.join(data.xpath('.//div[@class="author"]//text()')) + author = author.partition('by')[-1] counter -= 1 - + s = SearchResult() s.cover_url = cover_url s.title = title.strip() s.author = author.strip() - s.price = '$0.00' + s.price = price.strip() s.detail_item = id.strip() s.drm = SearchResult.DRM_UNLOCKED - s.formats = 'EPUB' - + yield s diff --git a/src/calibre/gui2/store/open_library_plugin.py b/src/calibre/gui2/store/stores/open_library_plugin.py similarity index 100% rename from src/calibre/gui2/store/open_library_plugin.py rename to src/calibre/gui2/store/stores/open_library_plugin.py diff --git a/src/calibre/gui2/store/oreilly_plugin.py b/src/calibre/gui2/store/stores/oreilly_plugin.py similarity index 100% rename from src/calibre/gui2/store/oreilly_plugin.py rename to src/calibre/gui2/store/stores/oreilly_plugin.py diff --git a/src/calibre/gui2/store/stores/pragmatic_bookshelf_plugin.py b/src/calibre/gui2/store/stores/pragmatic_bookshelf_plugin.py new file mode 100644 index 0000000000..671186ba87 --- /dev/null +++ b/src/calibre/gui2/store/stores/pragmatic_bookshelf_plugin.py @@ -0,0 +1,24 @@ +# -*- coding: utf-8 -*- + +from __future__ import (unicode_literals, division, absolute_import, print_function) + +__license__ = 'GPL 3' +__copyright__ = '2011, John Schember ' +__docformat__ = 'restructuredtext en' + +from calibre.gui2.store.basic_config import BasicStoreConfig +from calibre.gui2.store.opensearch_store import OpenSearchStore +from calibre.gui2.store.search_result import SearchResult + +class PragmaticBookshelfStore(BasicStoreConfig, OpenSearchStore): + + open_search_url = 'http://pragprog.com/catalog/search-description' + web_url = 'http://pragprog.com/' + + # http://pragprog.com/catalog.opds + + def search(self, query, max_results=10, timeout=60): + for s in OpenSearchStore.search(self, query, max_results, timeout): + s.drm = SearchResult.DRM_UNLOCKED + s.formats = 'EPUB, PDF, MOBI' + yield s diff --git a/src/calibre/gui2/store/smashwords_plugin.py b/src/calibre/gui2/store/stores/smashwords_plugin.py similarity index 100% rename from src/calibre/gui2/store/smashwords_plugin.py rename to src/calibre/gui2/store/stores/smashwords_plugin.py diff --git a/src/calibre/gui2/store/virtualo_plugin.py b/src/calibre/gui2/store/stores/virtualo_plugin.py similarity index 100% rename from src/calibre/gui2/store/virtualo_plugin.py rename to src/calibre/gui2/store/stores/virtualo_plugin.py diff --git a/src/calibre/gui2/store/waterstones_uk_plugin.py b/src/calibre/gui2/store/stores/waterstones_uk_plugin.py similarity index 100% rename from src/calibre/gui2/store/waterstones_uk_plugin.py rename to src/calibre/gui2/store/stores/waterstones_uk_plugin.py diff --git a/src/calibre/gui2/store/weightless_books_plugin.py b/src/calibre/gui2/store/stores/weightless_books_plugin.py similarity index 100% rename from src/calibre/gui2/store/weightless_books_plugin.py rename to src/calibre/gui2/store/stores/weightless_books_plugin.py diff --git a/src/calibre/gui2/store/whsmith_uk_plugin.py b/src/calibre/gui2/store/stores/whsmith_uk_plugin.py similarity index 100% rename from src/calibre/gui2/store/whsmith_uk_plugin.py rename to src/calibre/gui2/store/stores/whsmith_uk_plugin.py diff --git a/src/calibre/gui2/store/wizards_tower_books_plugin.py b/src/calibre/gui2/store/stores/wizards_tower_books_plugin.py similarity index 100% rename from src/calibre/gui2/store/wizards_tower_books_plugin.py rename to src/calibre/gui2/store/stores/wizards_tower_books_plugin.py diff --git a/src/calibre/gui2/store/woblink_plugin.py b/src/calibre/gui2/store/stores/woblink_plugin.py similarity index 100% rename from src/calibre/gui2/store/woblink_plugin.py rename to src/calibre/gui2/store/stores/woblink_plugin.py diff --git a/src/calibre/gui2/store/zixo_plugin.py b/src/calibre/gui2/store/stores/zixo_plugin.py similarity index 100% rename from src/calibre/gui2/store/zixo_plugin.py rename to src/calibre/gui2/store/stores/zixo_plugin.py diff --git a/src/calibre/gui2/tag_browser/__init__.py b/src/calibre/gui2/tag_browser/__init__.py new file mode 100644 index 0000000000..cc6da1e995 --- /dev/null +++ b/src/calibre/gui2/tag_browser/__init__.py @@ -0,0 +1,11 @@ +#!/usr/bin/env python +# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai +from __future__ import (unicode_literals, division, absolute_import, + print_function) + +__license__ = 'GPL v3' +__copyright__ = '2011, Kovid Goyal ' +__docformat__ = 'restructuredtext en' + + + diff --git a/src/calibre/gui2/tag_browser/model.py b/src/calibre/gui2/tag_browser/model.py new file mode 100644 index 0000000000..0e599ed431 --- /dev/null +++ b/src/calibre/gui2/tag_browser/model.py @@ -0,0 +1,1294 @@ +#!/usr/bin/env python +# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai +from __future__ import (unicode_literals, division, absolute_import, + print_function) +from future_builtins import map + +__license__ = 'GPL v3' +__copyright__ = '2011, Kovid Goyal ' +__docformat__ = 'restructuredtext en' + +import traceback, cPickle, copy +from itertools import repeat + +from PyQt4.Qt import (QAbstractItemModel, QIcon, QVariant, QFont, Qt, + QMimeData, QModelIndex, pyqtSignal) + +from calibre.gui2 import NONE, gprefs, config, error_dialog +from calibre.library.database2 import Tag +from calibre.utils.config import tweaks +from calibre.utils.icu import sort_key, lower, strcmp +from calibre.library.field_metadata import TagsIcons, category_icon_map +from calibre.gui2.dialogs.confirm_delete import confirm +from calibre.utils.formatter import EvalFormatter +from calibre.utils.search_query_parser import saved_searches + +TAG_SEARCH_STATES = {'clear': 0, 'mark_plus': 1, 'mark_plusplus': 2, + 'mark_minus': 3, 'mark_minusminus': 4} + +_bf = None +def bf(): + global _bf + if _bf is None: + _bf = QFont() + _bf.setBold(True) + _bf = QVariant(_bf) + return _bf + +class TagTreeItem(object): # {{{ + + CATEGORY = 0 + TAG = 1 + ROOT = 2 + + def __init__(self, data=None, category_icon=None, icon_map=None, + parent=None, tooltip=None, category_key=None, temporary=False): + self.parent = parent + self.children = [] + self.id_set = set() + self.is_gst = False + self.boxed = False + self.icon_state_map = list(map(QVariant, icon_map)) + if self.parent is not None: + self.parent.append(self) + + if data is None: + self.type = self.ROOT + else: + self.type = self.TAG if category_icon is None else self.CATEGORY + + if self.type == self.CATEGORY: + self.name, self.icon = map(QVariant, (data, category_icon)) + self.py_name = data + self.category_key = category_key + self.temporary = temporary + self.tag = Tag(data, category=category_key, + is_editable=category_key not in ['news', 'search', 'identifiers'], + is_searchable=category_key not in ['search']) + + elif self.type == self.TAG: + self.icon_state_map[0] = QVariant(data.icon) + self.tag = data + + self.tooltip = (tooltip + ' ') if tooltip else '' + + def break_cycles(self): + del self.parent + del self.children + + def __str__(self): + if self.type == self.ROOT: + return 'ROOT' + if self.type == self.CATEGORY: + return 'CATEGORY:'+str(QVariant.toString( + self.name))+':%d'%len(getattr(self, + 'children', [])) + return 'TAG: %s'%self.tag.name + + def row(self): + if self.parent is not None: + return self.parent.children.index(self) + return 0 + + def append(self, child): + child.parent = self + self.children.append(child) + + def data(self, role): + if role == Qt.UserRole: + return self + if self.type == self.TAG: + return self.tag_data(role) + if self.type == self.CATEGORY: + return self.category_data(role) + return NONE + + def category_data(self, role): + if role == Qt.DisplayRole: + return QVariant(self.py_name + ' [%d]'%len(self.child_tags())) + if role == Qt.EditRole: + return QVariant(self.py_name) + if role == Qt.DecorationRole: + if self.tag.state: + return self.icon_state_map[self.tag.state] + return self.icon + if role == Qt.FontRole: + return bf() + if role == Qt.ToolTipRole and self.tooltip is not None: + return QVariant(self.tooltip) + return NONE + + def tag_data(self, role): + tag = self.tag + if tag.use_sort_as_name: + name = tag.sort + tt_author = True + else: + p = self + while p.parent.type != self.ROOT: + p = p.parent + if not tag.is_hierarchical: + name = tag.original_name + else: + name = tag.name + tt_author = False + if role == Qt.DisplayRole: + count = len(self.id_set) + count = count if count > 0 else tag.count + if count == 0: + return QVariant('%s'%(name)) + else: + return QVariant('[%d] %s'%(count, name)) + if role == Qt.EditRole: + return QVariant(tag.original_name) + if role == Qt.DecorationRole: + return self.icon_state_map[tag.state] + if role == Qt.ToolTipRole: + if tt_author: + if tag.tooltip is not None: + return QVariant('(%s) %s'%(tag.name, tag.tooltip)) + else: + return QVariant(tag.name) + if tag.tooltip: + return QVariant(self.tooltip + tag.tooltip) + else: + return QVariant(self.tooltip) + return NONE + + def toggle(self, set_to=None): + ''' + set_to: None => advance the state, otherwise a value from TAG_SEARCH_STATES + ''' + if set_to is None: + while True: + self.tag.state = (self.tag.state + 1)%5 + if self.tag.state == TAG_SEARCH_STATES['mark_plus'] or \ + self.tag.state == TAG_SEARCH_STATES['mark_minus']: + if self.tag.is_searchable: + break + elif self.tag.state == TAG_SEARCH_STATES['mark_plusplus'] or\ + self.tag.state == TAG_SEARCH_STATES['mark_minusminus']: + if self.tag.is_searchable and len(self.children) and \ + self.tag.is_hierarchical == '5state': + break + else: + break + else: + self.tag.state = set_to + + def all_children(self): + res = [] + def recurse(nodes, res): + for t in nodes: + res.append(t) + recurse(t.children, res) + recurse(self.children, res) + return res + + def child_tags(self): + res = [] + def recurse(nodes, res): + for t in nodes: + if t.type != TagTreeItem.CATEGORY: + res.append(t) + recurse(t.children, res) + recurse(self.children, res) + return res + # }}} + +class TagsModel(QAbstractItemModel): # {{{ + + search_item_renamed = pyqtSignal() + tag_item_renamed = pyqtSignal() + refresh_required = pyqtSignal() + restriction_error = pyqtSignal() + drag_drop_finished = pyqtSignal(object) + user_categories_edited = pyqtSignal(object, object) + + def __init__(self, parent): + QAbstractItemModel.__init__(self, parent) + self.node_map = {} + self.category_nodes = [] + iconmap = {} + for key in category_icon_map: + iconmap[key] = QIcon(I(category_icon_map[key])) + self.category_icon_map = TagsIcons(iconmap) + self.categories_with_ratings = ['authors', 'series', 'publisher', 'tags'] + self.icon_state_map = [None, QIcon(I('plus.png')), QIcon(I('plusplus.png')), + QIcon(I('minus.png')), QIcon(I('minusminus.png'))] + + self.hidden_categories = set() + self.search_restriction = None + self.filter_categories_by = None + self.collapse_model = 'disable' + self.row_map = [] + self.root_item = self.create_node(icon_map=self.icon_state_map) + self.db = None + self.reread_collapse_model({}, rebuild=False) + + def reread_collapse_model(self, state_map, rebuild=True): + if gprefs['tags_browser_collapse_at'] == 0: + self.collapse_model = 'disable' + else: + self.collapse_model = gprefs['tags_browser_partition_method'] + if rebuild: + self.rebuild_node_tree(state_map) + + def set_search_restriction(self, s): + self.search_restriction = s + self.rebuild_node_tree() + + def set_database(self, db): + self.beginResetModel() + self.search_restriction = None + hidden_cats = db.prefs.get('tag_browser_hidden_categories', None) + # migrate from config to db prefs + if hidden_cats is None: + hidden_cats = config['tag_browser_hidden_categories'] + self.hidden_categories = set() + # strip out any non-existence field keys + for cat in hidden_cats: + if cat in db.field_metadata: + self.hidden_categories.add(cat) + db.prefs.set('tag_browser_hidden_categories', list(self.hidden_categories)) + + self.db = db + self._run_rebuild() + self.endResetModel() + + def rebuild_node_tree(self, state_map={}): + self.beginResetModel() + self._run_rebuild(state_map=state_map) + self.endResetModel() + + def _run_rebuild(self, state_map={}): + for node in self.node_map.itervalues(): + node.break_cycles() + del node #Clear reference to node in the current frame + self.node_map.clear() + self.category_nodes = [] + self.root_item = self.create_node(icon_map=self.icon_state_map) + self._rebuild_node_tree(state_map=state_map) + + def _rebuild_node_tree(self, state_map): + # Note that _get_category_nodes can indirectly change the + # user_categories dict. + data = self._get_category_nodes(config['sort_tags_by']) + gst = self.db.prefs.get('grouped_search_terms', {}) + + last_category_node = None + category_node_map = {} + self.category_node_tree = {} + for i, key in enumerate(self.row_map): + if self.hidden_categories: + if key in self.hidden_categories: + continue + found = False + for cat in self.hidden_categories: + if cat.startswith('@') and key.startswith(cat + '.'): + found = True + if found: + continue + is_gst = False + if key.startswith('@') and key[1:] in gst: + tt = _(u'The grouped search term name is "{0}"').format(key[1:]) + is_gst = True + elif key == 'news': + tt = '' + else: + tt = _(u'The lookup/search name is "{0}"').format(key) + + if key.startswith('@'): + path_parts = [p for p in key.split('.')] + path = '' + last_category_node = self.root_item + tree_root = self.category_node_tree + for i,p in enumerate(path_parts): + path += p + if path not in category_node_map: + node = self.create_node(parent=last_category_node, + data=p[1:] if i == 0 else p, + category_icon=self.category_icon_map[key], + tooltip=tt if path == key else path, + category_key=path, + icon_map=self.icon_state_map) + last_category_node = node + category_node_map[path] = node + self.category_nodes.append(node) + node.can_be_edited = (not is_gst) and (i == (len(path_parts)-1)) + node.is_gst = is_gst + if not is_gst: + node.tag.is_hierarchical = '5state' + if not is_gst: + tree_root[p] = {} + tree_root = tree_root[p] + else: + last_category_node = category_node_map[path] + tree_root = tree_root[p] + path += '.' + else: + node = self.create_node(parent=self.root_item, + data=self.categories[key], + category_icon=self.category_icon_map[key], + tooltip=tt, category_key=key, + icon_map=self.icon_state_map) + node.is_gst = False + category_node_map[key] = node + last_category_node = node + self.category_nodes.append(node) + self._create_node_tree(data, state_map) + + def _create_node_tree(self, data, state_map): + sort_by = config['sort_tags_by'] + + eval_formatter = EvalFormatter() + + if data is None: + print ('_create_node_tree: no data!') + traceback.print_stack() + return + + collapse = gprefs['tags_browser_collapse_at'] + collapse_model = self.collapse_model + if collapse == 0: + collapse_model = 'disable' + elif collapse_model != 'disable': + if sort_by == 'name': + collapse_template = tweaks['categories_collapsed_name_template'] + elif sort_by == 'rating': + collapse_model = 'partition' + collapse_template = tweaks['categories_collapsed_rating_template'] + else: + collapse_model = 'partition' + collapse_template = tweaks['categories_collapsed_popularity_template'] + + def process_one_node(category, state_map): # {{{ + collapse_letter = None + category_node = category + key = category_node.category_key + if key not in data: + return + cat_len = len(data[key]) + if cat_len <= 0: + return + + category_child_map = {} + fm = self.db.field_metadata[key] + clear_rating = True if key not in self.categories_with_ratings and \ + not fm['is_custom'] and \ + not fm['kind'] == 'user' \ + else False + in_uc = fm['kind'] == 'user' + tt = key if in_uc else None + + if collapse_model == 'first letter': + # Build a list of 'equal' first letters by looking for + # overlapping ranges. If a range overlaps another, then the + # letters are assumed to be equivalent. ICU collating is complex + # beyond belief. This mechanism lets us determine the logical + # first character from ICU's standpoint. + chardict = {} + for idx,tag in enumerate(data[key]): + if not tag.sort: + c = ' ' + else: + c = icu_upper(tag.sort[0]) + if c not in chardict: + chardict[c] = [idx, idx] + else: + chardict[c][1] = idx + + # sort the ranges to facilitate detecting overlap + ranges = sorted([(v[0], v[1], c) for c,v in chardict.items()]) + + # Create a list of 'first letters' to use for each item in + # the category. The list is generated using the ranges. Overlaps + # are filled with the character that first occurs. + cl_list = list(repeat(None, len(data[key]))) + for t in ranges: + start = t[0] + c = t[2] + if cl_list[start] is None: + nc = c + else: + nc = cl_list[start] + for i in range(start, t[1]+1): + cl_list[i] = nc + + for idx,tag in enumerate(data[key]): + if clear_rating: + tag.avg_rating = None + tag.state = state_map.get((tag.name, tag.category), 0) + + if collapse_model != 'disable' and cat_len > collapse: + if collapse_model == 'partition': + if (idx % collapse) == 0: + d = {'first': tag} + if cat_len > idx + collapse: + d['last'] = data[key][idx+collapse-1] + else: + d['last'] = data[key][cat_len-1] + name = eval_formatter.safe_format(collapse_template, + d, 'TAG_VIEW', None) + sub_cat = self.create_node(parent=category, data = name, + tooltip = None, temporary=True, + category_icon = category_node.icon, + category_key=category_node.category_key, + icon_map=self.icon_state_map) + sub_cat.tag.is_searchable = False + else: # by 'first letter' + cl = cl_list[idx] + if cl != collapse_letter: + collapse_letter = cl + sub_cat = self.create_node(parent=category, + data = collapse_letter, + category_icon = category_node.icon, + tooltip = None, temporary=True, + category_key=category_node.category_key, + icon_map=self.icon_state_map) + node_parent = sub_cat + else: + node_parent = category + + # category display order is important here. The following works + # only of all the non-user categories are displayed before the + # user categories + components = [t.strip() for t in tag.original_name.split('.') + if t.strip()] + if len(components) == 0 or '.'.join(components) != tag.original_name: + components = [tag.original_name] + if (not tag.is_hierarchical) and (in_uc or + (fm['is_custom'] and fm['display'].get('is_names', False)) or + key in ['authors', 'publisher', 'news', 'formats', 'rating'] or + key not in self.db.prefs.get('categories_using_hierarchy', []) or + len(components) == 1): + n = self.create_node(parent=node_parent, data=tag, tooltip=tt, + icon_map=self.icon_state_map) + if tag.id_set is not None: + n.id_set |= tag.id_set + category_child_map[tag.name, tag.category] = n + else: + for i,comp in enumerate(components): + if i == 0: + child_map = category_child_map + else: + child_map = dict([((t.tag.name, t.tag.category), t) + for t in node_parent.children + if t.type != TagTreeItem.CATEGORY]) + if (comp,tag.category) in child_map: + node_parent = child_map[(comp,tag.category)] + node_parent.tag.is_hierarchical = \ + '5state' if tag.category != 'search' else '3state' + else: + if i < len(components)-1: + t = copy.copy(tag) + t.original_name = '.'.join(components[:i+1]) + if key != 'search': + # This 'manufactured' intermediate node can + # be searched, but cannot be edited. + t.is_editable = False + else: + t.is_searchable = t.is_editable = False + else: + t = tag + if not in_uc: + t.original_name = t.name + t.is_hierarchical = \ + '5state' if t.category != 'search' else '3state' + t.name = comp + node_parent = self.create_node(parent=node_parent, data=t, + tooltip=tt, icon_map=self.icon_state_map) + child_map[(comp,tag.category)] = node_parent + # This id_set must not be None + node_parent.id_set |= tag.id_set + return + # }}} + + for category in self.category_nodes: + process_one_node(category, state_map.get(category.py_name, {})) + + # Drag'n Drop {{{ + def mimeTypes(self): + return ["application/calibre+from_library", + 'application/calibre+from_tag_browser'] + + def mimeData(self, indexes): + data = [] + for idx in indexes: + if idx.isValid(): + # get some useful serializable data + node = self.get_node(idx) + path = self.path_for_index(idx) + if node.type == TagTreeItem.CATEGORY: + d = (node.type, node.py_name, node.category_key) + else: + t = node.tag + p = node + while p.type != TagTreeItem.CATEGORY: + p = p.parent + d = (node.type, p.category_key, p.is_gst, t.original_name, + t.category, path) + data.append(d) + else: + data.append(None) + raw = bytearray(cPickle.dumps(data, -1)) + ans = QMimeData() + ans.setData('application/calibre+from_tag_browser', raw) + return ans + + def dropMimeData(self, md, action, row, column, parent): + fmts = set([unicode(x) for x in md.formats()]) + if not fmts.intersection(set(self.mimeTypes())): + return False + if "application/calibre+from_library" in fmts: + if action != Qt.CopyAction: + return False + return self.do_drop_from_library(md, action, row, column, parent) + elif 'application/calibre+from_tag_browser' in fmts: + return self.do_drop_from_tag_browser(md, action, row, column, parent) + + def do_drop_from_tag_browser(self, md, action, row, column, parent): + if not parent.isValid(): + return False + dest = self.get_node(parent) + if dest.type != TagTreeItem.CATEGORY: + return False + if not md.hasFormat('application/calibre+from_tag_browser'): + return False + data = str(md.data('application/calibre+from_tag_browser')) + src = cPickle.loads(data) + for s in src: + if s[0] != TagTreeItem.TAG: + return False + return self.move_or_copy_item_to_user_category(src, dest, action) + + def move_or_copy_item_to_user_category(self, src, dest, action): + ''' + src is a list of tuples representing items to copy. The tuple is + (type, containing category key, category key is global search term, + full name, category key, path to node) + The type must be TagTreeItem.TAG + dest is the TagTreeItem node to receive the items + action is Qt.CopyAction or Qt.MoveAction + ''' + def process_source_node(user_cats, src_parent, src_parent_is_gst, + is_uc, dest_key, node): + ''' + Copy/move an item and all its children to the destination + ''' + copied = False + src_name = node.tag.original_name + src_cat = node.tag.category + # delete the item if the source is a user category and action is move + if is_uc and not src_parent_is_gst and src_parent in user_cats and \ + action == Qt.MoveAction: + new_cat = [] + for tup in user_cats[src_parent]: + if src_name == tup[0] and src_cat == tup[1]: + continue + new_cat.append(list(tup)) + user_cats[src_parent] = new_cat + else: + copied = True + + # Now add the item to the destination user category + add_it = True + if not is_uc and src_cat == 'news': + src_cat = 'tags' + for tup in user_cats[dest_key]: + if src_name == tup[0] and src_cat == tup[1]: + add_it = False + if add_it: + user_cats[dest_key].append([src_name, src_cat, 0]) + + for c in node.children: + copied = process_source_node(user_cats, src_parent, src_parent_is_gst, + is_uc, dest_key, c) + return copied + + user_cats = self.db.prefs.get('user_categories', {}) + path = None + for s in src: + src_parent, src_parent_is_gst = s[1:3] + path = s[5] + + if src_parent.startswith('@'): + is_uc = True + src_parent = src_parent[1:] + else: + is_uc = False + dest_key = dest.category_key[1:] + + if dest_key not in user_cats: + continue + + node = self.index_for_path(path) + if node: + process_source_node(user_cats, src_parent, src_parent_is_gst, + is_uc, dest_key, + self.get_node(node)) + + self.db.prefs.set('user_categories', user_cats) + self.refresh_required.emit() + + return True + + def do_drop_from_library(self, md, action, row, column, parent): + idx = parent + if idx.isValid(): + node = self.data(idx, Qt.UserRole) + if node.type == TagTreeItem.TAG: + fm = self.db.metadata_for_field(node.tag.category) + if node.tag.category in \ + ('tags', 'series', 'authors', 'rating', 'publisher') or \ + (fm['is_custom'] and ( + fm['datatype'] in ['text', 'rating', 'series', + 'enumeration'] or + (fm['datatype'] == 'composite' and + fm['display'].get('make_category', False)))): + mime = 'application/calibre+from_library' + ids = list(map(int, str(md.data(mime)).split())) + self.handle_drop(node, ids) + return True + elif node.type == TagTreeItem.CATEGORY: + fm_dest = self.db.metadata_for_field(node.category_key) + if fm_dest['kind'] == 'user': + fm_src = self.db.metadata_for_field(md.column_name) + if md.column_name in ['authors', 'publisher', 'series'] or \ + (fm_src['is_custom'] and ( + (fm_src['datatype'] in ['series', 'text', 'enumeration'] and + not fm_src['is_multiple']))or + (fm_src['datatype'] == 'composite' and + fm_src['display'].get('make_category', False))): + mime = 'application/calibre+from_library' + ids = list(map(int, str(md.data(mime)).split())) + self.handle_user_category_drop(node, ids, md.column_name) + return True + return False + + def handle_user_category_drop(self, on_node, ids, column): + categories = self.db.prefs.get('user_categories', {}) + category = categories.get(on_node.category_key[1:], None) + if category is None: + return + fm_src = self.db.metadata_for_field(column) + for id in ids: + label = fm_src['label'] + if not fm_src['is_custom']: + if label == 'authors': + items = self.db.get_authors_with_ids() + items = [(i[0], i[1].replace('|', ',')) for i in items] + value = self.db.authors(id, index_is_id=True) + value = [v.replace('|', ',') for v in value.split(',')] + elif label == 'publisher': + items = self.db.get_publishers_with_ids() + value = self.db.publisher(id, index_is_id=True) + elif label == 'series': + items = self.db.get_series_with_ids() + value = self.db.series(id, index_is_id=True) + else: + items = self.db.get_custom_items_with_ids(label=label) + if fm_src['datatype'] != 'composite': + value = self.db.get_custom(id, label=label, index_is_id=True) + else: + value = self.db.get_property(id, loc=fm_src['rec_index'], + index_is_id=True) + if value is None: + return + if not isinstance(value, list): + value = [value] + for val in value: + for (v, c, id) in category: + if v == val and c == column: + break + else: + category.append([val, column, 0]) + categories[on_node.category_key[1:]] = category + self.db.prefs.set('user_categories', categories) + self.refresh_required.emit() + + def handle_drop(self, on_node, ids): + #print 'Dropped ids:', ids, on_node.tag + key = on_node.tag.category + if (key == 'authors' and len(ids) >= 5): + if not confirm('

'+_('Changing the authors for several books can ' + 'take a while. Are you sure?') + +'

', 'tag_browser_drop_authors', self.parent()): + return + elif len(ids) > 15: + if not confirm('

'+_('Changing the metadata for that many books ' + 'can take a while. Are you sure?') + +'

', 'tag_browser_many_changes', self.parent()): + return + + fm = self.db.metadata_for_field(key) + is_multiple = fm['is_multiple'] + val = on_node.tag.original_name + for id in ids: + mi = self.db.get_metadata(id, index_is_id=True) + + # Prepare to ignore the author, unless it is changed. Title is + # always ignored -- see the call to set_metadata + set_authors = False + + # Author_sort cannot change explicitly. Changing the author might + # change it. + mi.author_sort = None # Never will change by itself. + + if key == 'authors': + mi.authors = [val] + set_authors=True + elif fm['datatype'] == 'rating': + mi.set(key, len(val) * 2) + elif fm['is_custom'] and fm['datatype'] == 'series': + mi.set(key, val, extra=1.0) + elif is_multiple: + new_val = mi.get(key, []) + if val in new_val: + # Fortunately, only one field can change, so the continue + # won't break anything + continue + new_val.append(val) + mi.set(key, new_val) + else: + mi.set(key, val) + self.db.set_metadata(id, mi, set_title=False, + set_authors=set_authors, commit=False) + self.db.commit() + self.drag_drop_finished.emit(ids) + # }}} + + def _get_category_nodes(self, sort): + ''' + Called by __init__. Do not directly call this method. + ''' + self.row_map = [] + self.categories = {} + + # Get the categories + if self.search_restriction: + try: + data = self.db.get_categories(sort=sort, + icon_map=self.category_icon_map, + ids=self.db.search('', return_matches=True)) + except: + data = self.db.get_categories(sort=sort, icon_map=self.category_icon_map) + self.restriction_error.emit() + else: + data = self.db.get_categories(sort=sort, icon_map=self.category_icon_map) + + # Reconstruct the user categories, putting them into metadata + self.db.field_metadata.remove_dynamic_categories() + tb_cats = self.db.field_metadata + for user_cat in sorted(self.db.prefs.get('user_categories', {}).keys(), + key=sort_key): + cat_name = '@' + user_cat # add the '@' to avoid name collision + while True: + try: + tb_cats.add_user_category(label=cat_name, name=user_cat) + dot = cat_name.rfind('.') + if dot < 0: + break + cat_name = cat_name[:dot] + except ValueError: + break + + for cat in sorted(self.db.prefs.get('grouped_search_terms', {}).keys(), + key=sort_key): + if (u'@' + cat) in data: + try: + tb_cats.add_user_category(label=u'@' + cat, name=cat) + except ValueError: + traceback.print_exc() + self.db.data.change_search_locations(self.db.field_metadata.get_search_terms()) + + if len(saved_searches().names()): + tb_cats.add_search_category(label='search', name=_('Searches')) + + if self.filter_categories_by: + for category in data.keys(): + data[category] = [t for t in data[category] + if lower(t.name).find(self.filter_categories_by) >= 0] + + tb_categories = self.db.field_metadata + for category in tb_categories: + if category in data: # The search category can come and go + self.row_map.append(category) + self.categories[category] = tb_categories[category]['name'] + return data + + def refresh(self, data=None): + ''' + Here to trap usages of refresh in the old architecture. Can eventually + be removed. + ''' + print ('TagsModel: refresh called!') + traceback.print_stack() + return False + + def create_node(self, *args, **kwargs): + node = TagTreeItem(*args, **kwargs) + self.node_map[id(node)] = node + return node + + def get_node(self, idx): + ans = self.node_map.get(idx.internalId(), self.root_item) + return ans + + def createIndex(self, row, column, internal_pointer=None): + idx = QAbstractItemModel.createIndex(self, row, column, + id(internal_pointer)) + return idx + + def index_for_category(self, name): + for row, category in enumerate(self.category_nodes): + if category.py_name == name: + return self.index(row, 0, QModelIndex()) + + def columnCount(self, parent): + return 1 + + def data(self, index, role): + if not index.isValid(): + return NONE + item = self.get_node(index) + return item.data(role) + + def setData(self, index, value, role=Qt.EditRole): + if not index.isValid(): + return False + # set up to reposition at the same item. We can do this except if + # working with the last item and that item is deleted, in which case + # we position at the parent label + val = unicode(value.toString()).strip() + if not val: + error_dialog(self.parent(), _('Item is blank'), + _('An item cannot be set to nothing. Delete it instead.')).exec_() + return False + item = self.get_node(index) + if item.type == TagTreeItem.CATEGORY and item.category_key.startswith('@'): + if val.find('.') >= 0: + error_dialog(self.parent(), _('Rename user category'), + _('You cannot use periods in the name when ' + 'renaming user categories'), show=True) + return False + + user_cats = self.db.prefs.get('user_categories', {}) + user_cat_keys_lower = [icu_lower(k) for k in user_cats] + ckey = item.category_key[1:] + ckey_lower = icu_lower(ckey) + dotpos = ckey.rfind('.') + if dotpos < 0: + nkey = val + else: + nkey = ckey[:dotpos+1] + val + nkey_lower = icu_lower(nkey) + for c in sorted(user_cats.keys(), key=sort_key): + if icu_lower(c).startswith(ckey_lower): + if len(c) == len(ckey): + if strcmp(ckey, nkey) != 0 and \ + nkey_lower in user_cat_keys_lower: + error_dialog(self.parent(), _('Rename user category'), + _('The name %s is already used')%nkey, show=True) + return False + user_cats[nkey] = user_cats[ckey] + del user_cats[ckey] + elif c[len(ckey)] == '.': + rest = c[len(ckey):] + if strcmp(ckey, nkey) != 0 and \ + icu_lower(nkey + rest) in user_cat_keys_lower: + error_dialog(self.parent(), _('Rename user category'), + _('The name %s is already used')%(nkey+rest), show=True) + return False + user_cats[nkey + rest] = user_cats[ckey + rest] + del user_cats[ckey + rest] + self.user_categories_edited.emit(user_cats, nkey) # Does a refresh + return True + + key = item.tag.category + name = item.tag.original_name + # make certain we know about the item's category + if key not in self.db.field_metadata: + return False + if key == 'authors': + if val.find('&') >= 0: + error_dialog(self.parent(), _('Invalid author name'), + _('Author names cannot contain & characters.')).exec_() + return False + if key == 'search': + if val in saved_searches().names(): + error_dialog(self.parent(), _('Duplicate search name'), + _('The saved search name %s is already used.')%val).exec_() + return False + saved_searches().rename(unicode(item.data(role).toString()), val) + item.tag.name = val + self.search_item_renamed.emit() # Does a refresh + else: + if key == 'series': + self.db.rename_series(item.tag.id, val) + elif key == 'publisher': + self.db.rename_publisher(item.tag.id, val) + elif key == 'tags': + self.db.rename_tag(item.tag.id, val) + elif key == 'authors': + self.db.rename_author(item.tag.id, val) + elif self.db.field_metadata[key]['is_custom']: + self.db.rename_custom_item(item.tag.id, val, + label=self.db.field_metadata[key]['label']) + self.tag_item_renamed.emit() + item.tag.name = val + self.rename_item_in_all_user_categories(name, key, val) + self.refresh_required.emit() + return True + + def rename_item_in_all_user_categories(self, item_name, item_category, new_name): + ''' + Search all user categories for items named item_name with category + item_category and rename them to new_name. The caller must arrange to + redisplay the tree as appropriate. + ''' + user_cats = self.db.prefs.get('user_categories', {}) + for k in user_cats.keys(): + new_contents = [] + for tup in user_cats[k]: + if tup[0] == item_name and tup[1] == item_category: + new_contents.append([new_name, item_category, 0]) + else: + new_contents.append(tup) + user_cats[k] = new_contents + self.db.prefs.set('user_categories', user_cats) + + def delete_item_from_all_user_categories(self, item_name, item_category): + ''' + Search all user categories for items named item_name with category + item_category and delete them. The caller must arrange to redisplay the + tree as appropriate. + ''' + user_cats = self.db.prefs.get('user_categories', {}) + for cat in user_cats.keys(): + self.delete_item_from_user_category(cat, item_name, item_category, + user_categories=user_cats) + self.db.prefs.set('user_categories', user_cats) + + def delete_item_from_user_category(self, category, item_name, item_category, + user_categories=None): + if user_categories is not None: + user_cats = user_categories + else: + user_cats = self.db.prefs.get('user_categories', {}) + new_contents = [] + for tup in user_cats[category]: + if tup[0] != item_name or tup[1] != item_category: + new_contents.append(tup) + user_cats[category] = new_contents + if user_categories is None: + self.db.prefs.set('user_categories', user_cats) + + def headerData(self, *args): + return NONE + + def flags(self, index, *args): + ans = Qt.ItemIsEnabled|Qt.ItemIsSelectable|Qt.ItemIsEditable + if index.isValid(): + node = self.data(index, Qt.UserRole) + if node.type == TagTreeItem.TAG: + if node.tag.is_editable: + ans |= Qt.ItemIsDragEnabled + fm = self.db.metadata_for_field(node.tag.category) + if node.tag.category in \ + ('tags', 'series', 'authors', 'rating', 'publisher') or \ + (fm['is_custom'] and \ + fm['datatype'] in ['text', 'rating', 'series', 'enumeration']): + ans |= Qt.ItemIsDropEnabled + else: + ans |= Qt.ItemIsDropEnabled + return ans + + def supportedDropActions(self): + return Qt.CopyAction|Qt.MoveAction + + def path_for_index(self, index): + ans = [] + while index.isValid(): + ans.append(index.row()) + index = self.parent(index) + ans.reverse() + return ans + + def index_for_path(self, path): + parent = QModelIndex() + for i in path: + parent = self.index(i, 0, parent) + if not parent.isValid(): + return QModelIndex() + return parent + + def index(self, row, column, parent): + if not self.hasIndex(row, column, parent): + return QModelIndex() + + if not parent.isValid(): + parent_item = self.root_item + else: + parent_item = self.get_node(parent) + + try: + child_item = parent_item.children[row] + except IndexError: + return QModelIndex() + + ans = self.createIndex(row, column, child_item) + return ans + + def parent(self, index): + if not index.isValid(): + return QModelIndex() + + child_item = self.get_node(index) + parent_item = getattr(child_item, 'parent', None) + + if parent_item is self.root_item or parent_item is None: + return QModelIndex() + + ans = self.createIndex(parent_item.row(), 0, parent_item) + return ans + + def rowCount(self, parent): + if parent.column() > 0: + return 0 + + if not parent.isValid(): + parent_item = self.root_item + else: + parent_item = self.get_node(parent) + + return len(parent_item.children) + + def reset_all_states(self, except_=None): + update_list = [] + def process_tag(tag_item): + tag = tag_item.tag + if tag is except_: + tag_index = self.createIndex(tag_item.row(), 0, tag_item) + self.dataChanged.emit(tag_index, tag_index) + elif tag.state != 0 or tag in update_list: + tag_index = self.createIndex(tag_item.row(), 0, tag_item) + tag.state = 0 + update_list.append(tag) + self.dataChanged.emit(tag_index, tag_index) + for t in tag_item.children: + process_tag(t) + + for t in self.root_item.children: + process_tag(t) + + def clear_state(self): + self.reset_all_states() + + def toggle(self, index, exclusive, set_to=None): + ''' + exclusive: clear all states before applying this one + set_to: None => advance the state, otherwise a value from TAG_SEARCH_STATES + ''' + if not index.isValid(): return False + item = self.get_node(index) + item.toggle(set_to=set_to) + if exclusive: + self.reset_all_states(except_=item.tag) + self.dataChanged.emit(index, index) + return True + + def tokens(self): + ans = [] + # Tags can be in the news and the tags categories. However, because of + # the desire to use two different icons (tags and news), the nodes are + # not shared, which can lead to the possibility of searching twice for + # the same tag. The tags_seen set helps us prevent that + tags_seen = set() + # Tag nodes are in their own category and possibly in user categories. + # They will be 'checked' in both places, but we want to put the node + # into the search string only once. The nodes_seen set helps us do that + nodes_seen = set() + + node_searches = {TAG_SEARCH_STATES['mark_plus'] : 'true', + TAG_SEARCH_STATES['mark_plusplus'] : '.true', + TAG_SEARCH_STATES['mark_minus'] : 'false', + TAG_SEARCH_STATES['mark_minusminus'] : '.false'} + + for node in self.category_nodes: + if node.tag.state: + if node.category_key == "news": + if node_searches[node.tag.state] == 'true': + ans.append('tags:"=' + _('News') + '"') + else: + ans.append('( not tags:"=' + _('News') + '")') + else: + ans.append('%s:%s'%(node.category_key, node_searches[node.tag.state])) + + key = node.category_key + for tag_item in node.all_children(): + if tag_item.type == TagTreeItem.CATEGORY: + if self.collapse_model == 'first letter' and \ + tag_item.temporary and not key.startswith('@') \ + and tag_item.tag.state: + k = 'author_sort' if key == 'authors' else key + letters_seen = {} + for subnode in tag_item.children: + letters_seen[subnode.tag.sort[0]] = True + charclass = ''.join(letters_seen) + if k == 'author_sort': + expr = r'%s:"~(^[%s])|(&\\s*[%s])"'%(k, charclass, charclass) + else: + expr = r'%s:"~^[%s]"'%(k, charclass) + if node_searches[tag_item.tag.state] == 'true': + ans.append(expr) + else: + ans.append('(not ' + expr + ')') + continue + tag = tag_item.tag + if tag.state != TAG_SEARCH_STATES['clear']: + if tag.state == TAG_SEARCH_STATES['mark_minus'] or \ + tag.state == TAG_SEARCH_STATES['mark_minusminus']: + prefix = ' not ' + else: + prefix = '' + category = tag.category if key != 'news' else 'tag' + add_colon = False + if self.db.field_metadata[tag.category]['is_csp']: + add_colon = True + + if tag.name and tag.name[0] == u'\u2605': # char is a star. Assume rating + ans.append('%s%s:%s'%(prefix, category, len(tag.name))) + else: + name = tag.original_name + use_prefix = tag.state in [TAG_SEARCH_STATES['mark_plusplus'], + TAG_SEARCH_STATES['mark_minusminus']] + if category == 'tags': + if name in tags_seen: + continue + tags_seen.add(name) + if tag in nodes_seen: + continue + nodes_seen.add(tag) + n = name.replace(r'"', r'\"') + if name.startswith('.'): + n = '.' + n + ans.append('%s%s:"=%s%s%s"'%(prefix, category, + '.' if use_prefix else '', n, + ':' if add_colon else '')) + return ans + + def find_item_node(self, key, txt, start_path, equals_match=False): + ''' + Search for an item (a node) in the tags browser list that matches both + the key (exact case-insensitive match) and txt (not equals_match => + case-insensitive contains match; equals_match => case_insensitive + equal match). Returns the path to the node. Note that paths are to a + location (second item, fourth item, 25 item), not to a node. If + start_path is None, the search starts with the topmost node. If the tree + is changed subsequent to calling this method, the path can easily refer + to a different node or no node at all. + ''' + if not txt: + return None + txt = lower(txt) if not equals_match else txt + self.path_found = None + if start_path is None: + start_path = [] + + def process_tag(depth, tag_index, tag_item, start_path): + path = self.path_for_index(tag_index) + if depth < len(start_path) and path[depth] <= start_path[depth]: + return False + tag = tag_item.tag + if tag is None: + return False + name = tag.original_name + if (equals_match and strcmp(name, txt) == 0) or \ + (not equals_match and lower(name).find(txt) >= 0): + self.path_found = path + return True + for i,c in enumerate(tag_item.children): + if process_tag(depth+1, self.createIndex(i, 0, c), c, start_path): + return True + return False + + def process_level(depth, category_index, start_path): + path = self.path_for_index(category_index) + if depth < len(start_path): + if path[depth] < start_path[depth]: + return False + if path[depth] > start_path[depth]: + start_path = path + my_key = self.get_node(category_index).category_key + for j in xrange(self.rowCount(category_index)): + tag_index = self.index(j, 0, category_index) + tag_item = self.get_node(tag_index) + if tag_item.type == TagTreeItem.CATEGORY: + if process_level(depth+1, tag_index, start_path): + return True + elif not key or strcmp(key, my_key) == 0: + if process_tag(depth+1, tag_index, tag_item, start_path): + return True + return False + + for i in xrange(self.rowCount(QModelIndex())): + if process_level(0, self.index(i, 0, QModelIndex()), start_path): + break + return self.path_found + + def find_category_node(self, key, parent=QModelIndex()): + ''' + Search for an category node (a top-level node) in the tags browser list + that matches the key (exact case-insensitive match). Returns the path to + the node. Paths are as in find_item_node. + ''' + if not key: + return None + + for i in xrange(self.rowCount(parent)): + idx = self.index(i, 0, parent) + node = self.get_node(idx) + if node.type == TagTreeItem.CATEGORY: + ckey = node.category_key + if strcmp(ckey, key) == 0: + return self.path_for_index(idx) + if len(node.children): + v = self.find_category_node(key, idx) + if v is not None: + return v + return None + + def set_boxed(self, idx): + tag_item = self.get_node(idx) + tag_item.boxed = True + self.dataChanged.emit(idx, idx) + + def clear_boxed(self): + ''' + Clear all boxes around items. + ''' + def process_tag(tag_index, tag_item): + if tag_item.boxed: + tag_item.boxed = False + self.dataChanged.emit(tag_index, tag_index) + for i,c in enumerate(tag_item.children): + process_tag(self.index(i, 0, tag_index), c) + + def process_level(category_index): + for j in xrange(self.rowCount(category_index)): + tag_index = self.index(j, 0, category_index) + tag_item = self.get_node(tag_index) + if tag_item.boxed: + tag_item.boxed = False + self.dataChanged.emit(tag_index, tag_index) + if tag_item.type == TagTreeItem.CATEGORY: + process_level(tag_index) + else: + process_tag(tag_index, tag_item) + + for i in xrange(self.rowCount(QModelIndex())): + process_level(self.index(i, 0, QModelIndex())) + + # }}} + diff --git a/src/calibre/gui2/tag_browser/ui.py b/src/calibre/gui2/tag_browser/ui.py new file mode 100644 index 0000000000..f7f724b118 --- /dev/null +++ b/src/calibre/gui2/tag_browser/ui.py @@ -0,0 +1,458 @@ +#!/usr/bin/env python +# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai +from __future__ import (unicode_literals, division, absolute_import, + print_function) + +__license__ = 'GPL v3' +__copyright__ = '2011, Kovid Goyal ' +__docformat__ = 'restructuredtext en' + +from functools import partial + +from PyQt4.Qt import (Qt, QIcon, QWidget, QHBoxLayout, QVBoxLayout, QShortcut, + QKeySequence, QToolButton, QString, QLabel, QFrame, QTimer, QComboBox, + QMenu, QPushButton) + +from calibre.gui2 import error_dialog, question_dialog +from calibre.gui2.widgets import HistoryLineEdit +from calibre.library.field_metadata import category_icon_map +from calibre.utils.icu import sort_key +from calibre.gui2.tag_browser.view import TagsView +from calibre.ebooks.metadata import title_sort +from calibre.gui2.dialogs.tag_categories import TagCategories +from calibre.gui2.dialogs.tag_list_editor import TagListEditor +from calibre.gui2.dialogs.edit_authors_dialog import EditAuthorsDialog + +class TagBrowserMixin(object): # {{{ + + def __init__(self, db): + self.library_view.model().count_changed_signal.connect(self.tags_view.recount) + self.tags_view.set_database(db, self.tag_match, self.sort_by) + self.tags_view.tags_marked.connect(self.search.set_search_string) + self.tags_view.tag_list_edit.connect(self.do_tags_list_edit) + self.tags_view.edit_user_category.connect(self.do_edit_user_categories) + self.tags_view.delete_user_category.connect(self.do_delete_user_category) + self.tags_view.del_item_from_user_cat.connect(self.do_del_item_from_user_cat) + self.tags_view.add_subcategory.connect(self.do_add_subcategory) + self.tags_view.add_item_to_user_cat.connect(self.do_add_item_to_user_cat) + self.tags_view.saved_search_edit.connect(self.do_saved_search_edit) + self.tags_view.rebuild_saved_searches.connect(self.do_rebuild_saved_searches) + self.tags_view.author_sort_edit.connect(self.do_author_sort_edit) + self.tags_view.tag_item_renamed.connect(self.do_tag_item_renamed) + self.tags_view.search_item_renamed.connect(self.saved_searches_changed) + self.tags_view.drag_drop_finished.connect(self.drag_drop_finished) + self.tags_view.restriction_error.connect(self.do_restriction_error, + type=Qt.QueuedConnection) + + for text, func, args, cat_name in ( + (_('Manage Authors'), + self.do_author_sort_edit, (self, None), 'authors'), + (_('Manage Series'), + self.do_tags_list_edit, (None, 'series'), 'series'), + (_('Manage Publishers'), + self.do_tags_list_edit, (None, 'publisher'), 'publisher'), + (_('Manage Tags'), + self.do_tags_list_edit, (None, 'tags'), 'tags'), + (_('Manage User Categories'), + self.do_edit_user_categories, (None,), 'user:'), + (_('Manage Saved Searches'), + self.do_saved_search_edit, (None,), 'search') + ): + self.manage_items_button.menu().addAction( + QIcon(I(category_icon_map[cat_name])), + text, partial(func, *args)) + + def do_restriction_error(self): + error_dialog(self.tags_view, _('Invalid search restriction'), + _('The current search restriction is invalid'), show=True) + + def do_add_subcategory(self, on_category_key, new_category_name=None): + ''' + Add a subcategory to the category 'on_category'. If new_category_name is + None, then a default name is shown and the user is offered the + opportunity to edit the name. + ''' + db = self.library_view.model().db + user_cats = db.prefs.get('user_categories', {}) + + # Ensure that the temporary name we will use is not already there + i = 0 + if new_category_name is not None: + new_name = new_category_name.replace('.', '') + else: + new_name = _('New Category').replace('.', '') + n = new_name + while True: + new_cat = on_category_key[1:] + '.' + n + if new_cat not in user_cats: + break + i += 1 + n = new_name + unicode(i) + # Add the new category + user_cats[new_cat] = [] + db.prefs.set('user_categories', user_cats) + self.tags_view.set_new_model() + m = self.tags_view.model() + idx = m.index_for_path(m.find_category_node('@' + new_cat)) + m.show_item_at_index(idx) + # Open the editor on the new item to rename it + if new_category_name is None: + self.tags_view.edit(idx) + + def do_edit_user_categories(self, on_category=None): + ''' + Open the user categories editor. + ''' + db = self.library_view.model().db + d = TagCategories(self, db, on_category) + if d.exec_() == d.Accepted: + db.prefs.set('user_categories', d.categories) + db.field_metadata.remove_user_categories() + for k in d.categories: + db.field_metadata.add_user_category('@' + k, k) + db.data.change_search_locations(db.field_metadata.get_search_terms()) + self.tags_view.set_new_model() + + def do_delete_user_category(self, category_name): + ''' + Delete the user category named category_name. Any leading '@' is removed + ''' + if category_name.startswith('@'): + category_name = category_name[1:] + db = self.library_view.model().db + user_cats = db.prefs.get('user_categories', {}) + cat_keys = sorted(user_cats.keys(), key=sort_key) + has_children = False + found = False + for k in cat_keys: + if k == category_name: + found = True + has_children = len(user_cats[k]) + elif k.startswith(category_name + '.'): + has_children = True + if not found: + return error_dialog(self.tags_view, _('Delete user category'), + _('%s is not a user category')%category_name, show=True) + if has_children: + if not question_dialog(self.tags_view, _('Delete user category'), + _('%s contains items. Do you really ' + 'want to delete it?')%category_name): + return + for k in cat_keys: + if k == category_name: + del user_cats[k] + elif k.startswith(category_name + '.'): + del user_cats[k] + db.prefs.set('user_categories', user_cats) + self.tags_view.set_new_model() + + def do_del_item_from_user_cat(self, user_cat, item_name, item_category): + ''' + Delete the item (item_name, item_category) from the user category with + key user_cat. Any leading '@' characters are removed + ''' + if user_cat.startswith('@'): + user_cat = user_cat[1:] + db = self.library_view.model().db + user_cats = db.prefs.get('user_categories', {}) + if user_cat not in user_cats: + error_dialog(self.tags_view, _('Remove category'), + _('User category %s does not exist')%user_cat, + show=True) + return + self.tags_view.model().delete_item_from_user_category(user_cat, + item_name, item_category) + self.tags_view.recount() + + def do_add_item_to_user_cat(self, dest_category, src_name, src_category): + ''' + Add the item src_name in src_category to the user category + dest_category. Any leading '@' is removed + ''' + db = self.library_view.model().db + user_cats = db.prefs.get('user_categories', {}) + + if dest_category and dest_category.startswith('@'): + dest_category = dest_category[1:] + + if dest_category not in user_cats: + return error_dialog(self.tags_view, _('Add to user category'), + _('A user category %s does not exist')%dest_category, show=True) + + # Now add the item to the destination user category + add_it = True + if src_category == 'news': + src_category = 'tags' + for tup in user_cats[dest_category]: + if src_name == tup[0] and src_category == tup[1]: + add_it = False + if add_it: + user_cats[dest_category].append([src_name, src_category, 0]) + db.prefs.set('user_categories', user_cats) + self.tags_view.recount() + + def do_tags_list_edit(self, tag, category): + ''' + Open the 'manage_X' dialog where X == category. If tag is not None, the + dialog will position the editor on that item. + ''' + db=self.library_view.model().db + if category == 'tags': + result = db.get_tags_with_ids() + key = sort_key + elif category == 'series': + result = db.get_series_with_ids() + key = lambda x:sort_key(title_sort(x)) + elif category == 'publisher': + result = db.get_publishers_with_ids() + key = sort_key + else: # should be a custom field + cc_label = None + if category in db.field_metadata: + cc_label = db.field_metadata[category]['label'] + result = db.get_custom_items_with_ids(label=cc_label) + else: + result = [] + key = sort_key + + d = TagListEditor(self, tag_to_match=tag, data=result, key=key) + d.exec_() + if d.result() == d.Accepted: + to_rename = d.to_rename # dict of new text to old id + to_delete = d.to_delete # list of ids + orig_name = d.original_names # dict of id: name + + rename_func = None + if category == 'tags': + rename_func = db.rename_tag + delete_func = db.delete_tag_using_id + elif category == 'series': + rename_func = db.rename_series + delete_func = db.delete_series_using_id + elif category == 'publisher': + rename_func = db.rename_publisher + delete_func = db.delete_publisher_using_id + else: + rename_func = partial(db.rename_custom_item, label=cc_label) + delete_func = partial(db.delete_custom_item_using_id, label=cc_label) + m = self.tags_view.model() + if rename_func: + for item in to_delete: + delete_func(item) + m.delete_item_from_all_user_categories(orig_name[item], category) + for old_id in to_rename: + rename_func(old_id, new_name=unicode(to_rename[old_id])) + m.rename_item_in_all_user_categories(orig_name[old_id], + category, unicode(to_rename[old_id])) + + # Clean up the library view + self.do_tag_item_renamed() + self.tags_view.recount() + + def do_tag_item_renamed(self): + # Clean up library view and search + # get information to redo the selection + rows = [r.row() for r in \ + self.library_view.selectionModel().selectedRows()] + m = self.library_view.model() + ids = [m.id(r) for r in rows] + + m.refresh(reset=False) + m.research() + self.library_view.select_rows(ids) + # refreshing the tags view happens at the emit()/call() site + + def do_author_sort_edit(self, parent, id, select_sort=True): + ''' + Open the manage authors dialog + ''' + db = self.library_view.model().db + editor = EditAuthorsDialog(parent, db, id, select_sort) + d = editor.exec_() + if d: + for (id, old_author, new_author, new_sort) in editor.result: + if old_author != new_author: + # The id might change if the new author already exists + id = db.rename_author(id, new_author) + db.set_sort_field_for_author(id, unicode(new_sort), + commit=False, notify=False) + db.commit() + self.library_view.model().refresh() + self.tags_view.recount() + + def drag_drop_finished(self, ids): + self.library_view.model().refresh_ids(ids) + +# }}} + +class TagBrowserWidget(QWidget): # {{{ + + def __init__(self, parent): + QWidget.__init__(self, parent) + self.parent = parent + self._layout = QVBoxLayout() + self.setLayout(self._layout) + self._layout.setContentsMargins(0,0,0,0) + + # Set up the find box & button + search_layout = QHBoxLayout() + self._layout.addLayout(search_layout) + self.item_search = HistoryLineEdit(parent) + try: + self.item_search.lineEdit().setPlaceholderText( + _('Find item in tag browser')) + except: + pass # Using Qt < 4.7 + self.item_search.setToolTip(_( + 'Search for items. This is a "contains" search; items containing the\n' + 'text anywhere in the name will be found. You can limit the search\n' + 'to particular categories using syntax similar to search. For example,\n' + 'tags:foo will find foo in any tag, but not in authors etc. Entering\n' + '*foo will filter all categories at once, showing only those items\n' + 'containing the text "foo"')) + search_layout.addWidget(self.item_search) + # Not sure if the shortcut should be translatable ... + sc = QShortcut(QKeySequence(_('ALT+f')), parent) + sc.activated.connect(self.set_focus_to_find_box) + + self.search_button = QToolButton() + self.search_button.setText(_('F&ind')) + self.search_button.setToolTip(_('Find the first/next matching item')) + search_layout.addWidget(self.search_button) + + self.expand_button = QToolButton() + self.expand_button.setText('-') + self.expand_button.setToolTip(_('Collapse all categories')) + search_layout.addWidget(self.expand_button) + search_layout.setStretch(0, 10) + search_layout.setStretch(1, 1) + search_layout.setStretch(2, 1) + + self.current_find_position = None + self.search_button.clicked.connect(self.find) + self.item_search.initialize('tag_browser_search') + self.item_search.lineEdit().returnPressed.connect(self.do_find) + self.item_search.lineEdit().textEdited.connect(self.find_text_changed) + self.item_search.activated[QString].connect(self.do_find) + self.item_search.completer().setCaseSensitivity(Qt.CaseSensitive) + + parent.tags_view = TagsView(parent) + self.tags_view = parent.tags_view + self.expand_button.clicked.connect(self.tags_view.collapseAll) + self._layout.addWidget(parent.tags_view) + + # Now the floating 'not found' box + l = QLabel(self.tags_view) + self.not_found_label = l + l.setFrameStyle(QFrame.StyledPanel) + l.setAutoFillBackground(True) + l.setText('

'+_('No More Matches.

Click Find again to go to first match')) + l.setAlignment(Qt.AlignVCenter) + l.setWordWrap(True) + l.resize(l.sizeHint()) + l.move(10,20) + l.setVisible(False) + self.not_found_label_timer = QTimer() + self.not_found_label_timer.setSingleShot(True) + self.not_found_label_timer.timeout.connect(self.not_found_label_timer_event, + type=Qt.QueuedConnection) + + parent.sort_by = QComboBox(parent) + # Must be in the same order as db2.CATEGORY_SORTS + for x in (_('Sort by name'), _('Sort by popularity'), + _('Sort by average rating')): + parent.sort_by.addItem(x) + parent.sort_by.setToolTip( + _('Set the sort order for entries in the Tag Browser')) + parent.sort_by.setStatusTip(parent.sort_by.toolTip()) + parent.sort_by.setCurrentIndex(0) + self._layout.addWidget(parent.sort_by) + + # Must be in the same order as db2.MATCH_TYPE + parent.tag_match = QComboBox(parent) + for x in (_('Match any'), _('Match all')): + parent.tag_match.addItem(x) + parent.tag_match.setCurrentIndex(0) + self._layout.addWidget(parent.tag_match) + parent.tag_match.setToolTip( + _('When selecting multiple entries in the Tag Browser ' + 'match any or all of them')) + parent.tag_match.setStatusTip(parent.tag_match.toolTip()) + + + l = parent.manage_items_button = QPushButton(self) + l.setStyleSheet('QPushButton {text-align: left; }') + l.setText(_('Manage authors, tags, etc')) + l.setToolTip(_('All of these category_managers are available by right-clicking ' + 'on items in the tag browser above')) + l.m = QMenu() + l.setMenu(l.m) + self._layout.addWidget(l) + + # self.leak_test_timer = QTimer(self) + # self.leak_test_timer.timeout.connect(self.test_for_leak) + # self.leak_test_timer.start(5000) + + def set_pane_is_visible(self, to_what): + self.tags_view.set_pane_is_visible(to_what) + + def find_text_changed(self, str): + self.current_find_position = None + + def set_focus_to_find_box(self): + self.item_search.setFocus() + self.item_search.lineEdit().selectAll() + + def do_find(self, str=None): + self.current_find_position = None + self.find() + + def find(self): + model = self.tags_view.model() + model.clear_boxed() + txt = unicode(self.item_search.currentText()).strip() + + if txt.startswith('*'): + self.tags_view.set_new_model(filter_categories_by=txt[1:]) + self.current_find_position = None + return + if model.get_filter_categories_by(): + self.tags_view.set_new_model(filter_categories_by=None) + self.current_find_position = None + model = self.tags_view.model() + + if not txt: + return + + self.item_search.lineEdit().blockSignals(True) + self.search_button.setFocus(True) + self.item_search.lineEdit().blockSignals(False) + + key = None + colon = txt.rfind(':') if len(txt) > 2 else 0 + if colon > 0: + key = self.parent.library_view.model().db.\ + field_metadata.search_term_to_field_key(txt[:colon]) + txt = txt[colon+1:] + + self.current_find_position = \ + model.find_item_node(key, txt, self.current_find_position) + if self.current_find_position: + model.show_item_at_path(self.current_find_position, box=True) + elif self.item_search.text(): + self.not_found_label.setVisible(True) + if self.tags_view.verticalScrollBar().isVisible(): + sbw = self.tags_view.verticalScrollBar().width() + else: + sbw = 0 + width = self.width() - 8 - sbw + height = self.not_found_label.heightForWidth(width) + 20 + self.not_found_label.resize(width, height) + self.not_found_label.move(4, 10) + self.not_found_label_timer.start(2000) + + def not_found_label_timer_event(self): + self.not_found_label.setVisible(False) + +# }}} + diff --git a/src/calibre/gui2/tag_browser/view.py b/src/calibre/gui2/tag_browser/view.py new file mode 100644 index 0000000000..4ff1227a6a --- /dev/null +++ b/src/calibre/gui2/tag_browser/view.py @@ -0,0 +1,588 @@ +#!/usr/bin/env python +# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai +from __future__ import (unicode_literals, division, absolute_import, + print_function) + +__license__ = 'GPL v3' +__copyright__ = '2011, Kovid Goyal ' +__docformat__ = 'restructuredtext en' + +import cPickle +from functools import partial +from itertools import izip + +from PyQt4.Qt import (QItemDelegate, Qt, QTreeView, pyqtSignal, QSize, QIcon, + QApplication, QMenu, QPoint, QModelIndex) + +from calibre.gui2.tag_browser.model import (TagTreeItem, TAG_SEARCH_STATES, + TagsModel) +from calibre.gui2 import config, gprefs +from calibre.utils.search_query_parser import saved_searches +from calibre.utils.icu import sort_key + +class TagDelegate(QItemDelegate): # {{{ + + def paint(self, painter, option, index): + item = index.data(Qt.UserRole).toPyObject() + if item.type != TagTreeItem.TAG: + QItemDelegate.paint(self, painter, option, index) + return + r = option.rect + model = self.parent().model() + icon = model.data(index, Qt.DecorationRole).toPyObject() + painter.save() + if item.tag.state != 0 or not config['show_avg_rating'] or \ + item.tag.avg_rating is None: + icon.paint(painter, r, Qt.AlignLeft) + else: + painter.setOpacity(0.3) + icon.paint(painter, r, Qt.AlignLeft) + painter.setOpacity(1) + rating = item.tag.avg_rating + painter.setClipRect(r.left(), r.bottom()-int(r.height()*(rating/5.0)), + r.width(), r.height()) + icon.paint(painter, r, Qt.AlignLeft) + painter.setClipRect(r) + + # Paint the text + if item.boxed: + painter.drawRoundedRect(r.adjusted(1,1,-1,-1), 5, 5) + r.setLeft(r.left()+r.height()+3) + painter.drawText(r, Qt.AlignLeft|Qt.AlignVCenter, + model.data(index, Qt.DisplayRole).toString()) + painter.restore() + + # }}} + +class TagsView(QTreeView): # {{{ + + refresh_required = pyqtSignal() + tags_marked = pyqtSignal(object) + edit_user_category = pyqtSignal(object) + delete_user_category = pyqtSignal(object) + del_item_from_user_cat = pyqtSignal(object, object, object) + add_item_to_user_cat = pyqtSignal(object, object, object) + add_subcategory = pyqtSignal(object) + tag_list_edit = pyqtSignal(object, object) + saved_search_edit = pyqtSignal(object) + rebuild_saved_searches = pyqtSignal() + author_sort_edit = pyqtSignal(object, object) + tag_item_renamed = pyqtSignal() + search_item_renamed = pyqtSignal() + drag_drop_finished = pyqtSignal(object) + restriction_error = pyqtSignal() + show_at_path = pyqtSignal() + + def __init__(self, parent=None): + QTreeView.__init__(self, parent=None) + self.tag_match = None + self.disable_recounting = False + self.setUniformRowHeights(True) + self.setCursor(Qt.PointingHandCursor) + self.setIconSize(QSize(30, 30)) + self.setTabKeyNavigation(True) + self.setAlternatingRowColors(True) + self.setAnimated(True) + self.setHeaderHidden(True) + self.setItemDelegate(TagDelegate(self)) + self.made_connections = False + self.setAcceptDrops(True) + self.setDragEnabled(True) + self.setDragDropMode(self.DragDrop) + self.setDropIndicatorShown(True) + self.setAutoExpandDelay(500) + self.pane_is_visible = False + self.search_icon = QIcon(I('search.png')) + self.user_category_icon = QIcon(I('tb_folder.png')) + self.delete_icon = QIcon(I('list_remove.png')) + self.rename_icon = QIcon(I('edit-undo.png')) + self.show_at_path.connect(self.show_item_at_path, + type=Qt.QueuedConnection) + + self._model = TagsModel(self) + self._model.search_item_renamed.connect(self.search_item_renamed) + self._model.refresh_required.connect(self.refresh_required, + type=Qt.QueuedConnection) + self._model.tag_item_renamed.connect(self.tag_item_renamed) + self._model.restriction_error.connect(self.restriction_error) + self._model.user_categories_edited.connect(self.user_categories_edited, + type=Qt.QueuedConnection) + self._model.drag_drop_finished.connect(self.drag_drop_finished) + + @property + def hidden_categories(self): + return self._model.hidden_categories + + @property + def db(self): + return self._model.db + + @property + def collapse_model(self): + return self._model.collapse_model + + def set_pane_is_visible(self, to_what): + pv = self.pane_is_visible + self.pane_is_visible = to_what + if to_what and not pv: + self.recount() + + def get_state(self): + state_map = {} + expanded_categories = [] + for row, category in enumerate(self._model.category_nodes): + if self.isExpanded(self._model.index(row, 0, QModelIndex())): + expanded_categories.append(category.py_name) + states = [c.tag.state for c in category.child_tags()] + names = [(c.tag.name, c.tag.category) for c in category.child_tags()] + state_map[category.py_name] = dict(izip(names, states)) + return expanded_categories, state_map + + def reread_collapse_parameters(self): + self._model.reread_collapse_parameters(self.get_state()[1]) + + def set_database(self, db, tag_match, sort_by): + self._model.set_database(db) + + self.pane_is_visible = True # because TagsModel.set_database did a recount + self.sort_by = sort_by + self.tag_match = tag_match + self.setModel(self._model) + self.setContextMenuPolicy(Qt.CustomContextMenu) + pop = config['sort_tags_by'] + self.sort_by.setCurrentIndex(self.db.CATEGORY_SORTS.index(pop)) + try: + match_pop = self.db.MATCH_TYPE.index(config['match_tags_type']) + except ValueError: + match_pop = 0 + self.tag_match.setCurrentIndex(match_pop) + if not self.made_connections: + self.clicked.connect(self.toggle) + self.customContextMenuRequested.connect(self.show_context_menu) + self.refresh_required.connect(self.recount, type=Qt.QueuedConnection) + self.sort_by.currentIndexChanged.connect(self.sort_changed) + self.tag_match.currentIndexChanged.connect(self.match_changed) + self.made_connections = True + self.refresh_signal_processed = True + db.add_listener(self.database_changed) + self.expanded.connect(self.item_expanded) + + def database_changed(self, event, ids): + if self.refresh_signal_processed: + self.refresh_signal_processed = False + self.refresh_required.emit() + + def user_categories_edited(self, user_cats, nkey): + state_map = self.get_state()[1] + self.db.prefs.set('user_categories', user_cats) + self._model.rebuild_node_tree(state_map=state_map) + self.show_at_path.emit('@'+nkey) + + @property + def match_all(self): + return self.tag_match and self.tag_match.currentIndex() > 0 + + def sort_changed(self, pop): + config.set('sort_tags_by', self.db.CATEGORY_SORTS[pop]) + self.recount() + + def match_changed(self, pop): + try: + config.set('match_tags_type', self.db.MATCH_TYPE[pop]) + except: + pass + + def set_search_restriction(self, s): + s = s if s else None + self._model.set_search_restriction(s) + + def mouseReleaseEvent(self, event): + # Swallow everything except leftButton so context menus work correctly + if event.button() == Qt.LeftButton: + QTreeView.mouseReleaseEvent(self, event) + + def mouseDoubleClickEvent(self, event): + # swallow these to avoid toggling and editing at the same time + pass + + @property + def search_string(self): + tokens = self._model.tokens() + joiner = ' and ' if self.match_all else ' or ' + return joiner.join(tokens) + + def toggle(self, index): + self._toggle(index, None) + + def _toggle(self, index, set_to): + ''' + set_to: if None, advance the state. Otherwise must be one of the values + in TAG_SEARCH_STATES + ''' + modifiers = int(QApplication.keyboardModifiers()) + exclusive = modifiers not in (Qt.CTRL, Qt.SHIFT) + if self._model.toggle(index, exclusive, set_to=set_to): + self.tags_marked.emit(self.search_string) + + def conditional_clear(self, search_string): + if search_string != self.search_string: + self.clear() + + def context_menu_handler(self, action=None, category=None, + key=None, index=None, search_state=None): + if not action: + return + try: + if action == 'edit_item': + self.edit(index) + return + if action == 'open_editor': + self.tag_list_edit.emit(category, key) + return + if action == 'manage_categories': + self.edit_user_category.emit(category) + return + if action == 'search': + self._toggle(index, set_to=search_state) + return + if action == 'add_to_category': + tag = index.tag + if len(index.children) > 0: + for c in index.children: + self.add_item_to_user_cat.emit(category, c.tag.original_name, + c.tag.category) + self.add_item_to_user_cat.emit(category, tag.original_name, + tag.category) + return + if action == 'add_subcategory': + self.add_subcategory.emit(key) + return + if action == 'search_category': + self._toggle(index, set_to=search_state) + return + if action == 'delete_user_category': + self.delete_user_category.emit(key) + return + if action == 'delete_search': + saved_searches().delete(key) + self.rebuild_saved_searches.emit() + return + if action == 'delete_item_from_user_category': + tag = index.tag + if len(index.children) > 0: + for c in index.children: + self.del_item_from_user_cat.emit(key, c.tag.original_name, + c.tag.category) + self.del_item_from_user_cat.emit(key, tag.original_name, tag.category) + return + if action == 'manage_searches': + self.saved_search_edit.emit(category) + return + if action == 'edit_author_sort': + self.author_sort_edit.emit(self, index) + return + + reset_filter_categories = True + if action == 'hide': + self.hidden_categories.add(category) + elif action == 'show': + self.hidden_categories.discard(category) + elif action == 'categorization': + changed = self.collapse_model != category + self._model.collapse_model = category + if changed: + reset_filter_categories = False + gprefs['tags_browser_partition_method'] = category + elif action == 'defaults': + self.hidden_categories.clear() + self.db.prefs.set('tag_browser_hidden_categories', list(self.hidden_categories)) + if reset_filter_categories: + self._model.filter_categories_by = None + self._model.rebuild_node_tree() + except: + return + + def show_context_menu(self, point): + def display_name( tag): + if tag.category == 'search': + n = tag.name + if len(n) > 45: + n = n[:45] + '...' + return "'" + n + "'" + return tag.name + + index = self.indexAt(point) + self.context_menu = QMenu(self) + + if index.isValid(): + item = index.data(Qt.UserRole).toPyObject() + tag = None + + if item.type == TagTreeItem.TAG: + tag_item = item + tag = item.tag + while item.type != TagTreeItem.CATEGORY: + item = item.parent + + if item.type == TagTreeItem.CATEGORY: + if not item.category_key.startswith('@'): + while item.parent != self._model.root_item: + item = item.parent + category = unicode(item.name.toString()) + key = item.category_key + # Verify that we are working with a field that we know something about + if key not in self.db.field_metadata: + return True + + # Did the user click on a leaf node? + if tag: + # If the user right-clicked on an editable item, then offer + # the possibility of renaming that item. + if tag.is_editable: + # Add the 'rename' items + self.context_menu.addAction(self.rename_icon, + _('Rename %s')%display_name(tag), + partial(self.context_menu_handler, action='edit_item', + index=index)) + if key == 'authors': + self.context_menu.addAction(_('Edit sort for %s')%display_name(tag), + partial(self.context_menu_handler, + action='edit_author_sort', index=tag.id)) + + # is_editable is also overloaded to mean 'can be added + # to a user category' + m = self.context_menu.addMenu(self.user_category_icon, + _('Add %s to user category')%display_name(tag)) + nt = self.model().category_node_tree + def add_node_tree(tree_dict, m, path): + p = path[:] + for k in sorted(tree_dict.keys(), key=sort_key): + p.append(k) + n = k[1:] if k.startswith('@') else k + m.addAction(self.user_category_icon, n, + partial(self.context_menu_handler, + 'add_to_category', + category='.'.join(p), index=tag_item)) + if len(tree_dict[k]): + tm = m.addMenu(self.user_category_icon, + _('Children of %s')%n) + add_node_tree(tree_dict[k], tm, p) + p.pop() + add_node_tree(nt, m, []) + elif key == 'search': + self.context_menu.addAction(self.rename_icon, + _('Rename %s')%display_name(tag), + partial(self.context_menu_handler, action='edit_item', + index=index)) + self.context_menu.addAction(self.delete_icon, + _('Delete search %s')%display_name(tag), + partial(self.context_menu_handler, + action='delete_search', key=tag.name)) + if key.startswith('@') and not item.is_gst: + self.context_menu.addAction(self.user_category_icon, + _('Remove %s from category %s')% + (display_name(tag), item.py_name), + partial(self.context_menu_handler, + action='delete_item_from_user_category', + key = key, index = tag_item)) + # Add the search for value items. All leaf nodes are searchable + self.context_menu.addAction(self.search_icon, + _('Search for %s')%display_name(tag), + partial(self.context_menu_handler, action='search', + search_state=TAG_SEARCH_STATES['mark_plus'], + index=index)) + self.context_menu.addAction(self.search_icon, + _('Search for everything but %s')%display_name(tag), + partial(self.context_menu_handler, action='search', + search_state=TAG_SEARCH_STATES['mark_minus'], + index=index)) + self.context_menu.addSeparator() + elif key.startswith('@') and not item.is_gst: + if item.can_be_edited: + self.context_menu.addAction(self.rename_icon, + _('Rename %s')%item.py_name, + partial(self.context_menu_handler, action='edit_item', + index=index)) + self.context_menu.addAction(self.user_category_icon, + _('Add sub-category to %s')%item.py_name, + partial(self.context_menu_handler, + action='add_subcategory', key=key)) + self.context_menu.addAction(self.delete_icon, + _('Delete user category %s')%item.py_name, + partial(self.context_menu_handler, + action='delete_user_category', key=key)) + self.context_menu.addSeparator() + # Hide/Show/Restore categories + self.context_menu.addAction(_('Hide category %s') % category, + partial(self.context_menu_handler, action='hide', + category=key)) + if self.hidden_categories: + m = self.context_menu.addMenu(_('Show category')) + for col in sorted(self.hidden_categories, + key=lambda x: sort_key(self.db.field_metadata[x]['name'])): + m.addAction(self.db.field_metadata[col]['name'], + partial(self.context_menu_handler, action='show', category=col)) + + # search by category. Some categories are not searchable, such + # as search and news + if item.tag.is_searchable: + self.context_menu.addAction(self.search_icon, + _('Search for books in category %s')%category, + partial(self.context_menu_handler, + action='search_category', + index=self._model.createIndex(item.row(), 0, item), + search_state=TAG_SEARCH_STATES['mark_plus'])) + self.context_menu.addAction(self.search_icon, + _('Search for books not in category %s')%category, + partial(self.context_menu_handler, + action='search_category', + index=self._model.createIndex(item.row(), 0, item), + search_state=TAG_SEARCH_STATES['mark_minus'])) + # Offer specific editors for tags/series/publishers/saved searches + self.context_menu.addSeparator() + if key in ['tags', 'publisher', 'series'] or \ + self.db.field_metadata[key]['is_custom']: + self.context_menu.addAction(_('Manage %s')%category, + partial(self.context_menu_handler, action='open_editor', + category=tag.original_name if tag else None, + key=key)) + elif key == 'authors': + self.context_menu.addAction(_('Manage %s')%category, + partial(self.context_menu_handler, action='edit_author_sort')) + elif key == 'search': + self.context_menu.addAction(_('Manage Saved Searches'), + partial(self.context_menu_handler, action='manage_searches', + category=tag.name if tag else None)) + + # Always show the user categories editor + self.context_menu.addSeparator() + if key.startswith('@') and \ + key[1:] in self.db.prefs.get('user_categories', {}).keys(): + self.context_menu.addAction(_('Manage User Categories'), + partial(self.context_menu_handler, action='manage_categories', + category=key[1:])) + else: + self.context_menu.addAction(_('Manage User Categories'), + partial(self.context_menu_handler, action='manage_categories', + category=None)) + + if self.hidden_categories: + if not self.context_menu.isEmpty(): + self.context_menu.addSeparator() + self.context_menu.addAction(_('Show all categories'), + partial(self.context_menu_handler, action='defaults')) + + m = self.context_menu.addMenu(_('Change sub-categorization scheme')) + da = m.addAction('Disable', + partial(self.context_menu_handler, action='categorization', category='disable')) + fla = m.addAction('By first letter', + partial(self.context_menu_handler, action='categorization', category='first letter')) + pa = m.addAction('Partition', + partial(self.context_menu_handler, action='categorization', category='partition')) + if self.collapse_model == 'disable': + da.setCheckable(True) + da.setChecked(True) + elif self.collapse_model == 'first letter': + fla.setCheckable(True) + fla.setChecked(True) + else: + pa.setCheckable(True) + pa.setChecked(True) + + if not self.context_menu.isEmpty(): + self.context_menu.popup(self.mapToGlobal(point)) + return True + + def dragMoveEvent(self, event): + QTreeView.dragMoveEvent(self, event) + self.setDropIndicatorShown(False) + index = self.indexAt(event.pos()) + if not index.isValid(): + return + src_is_tb = event.mimeData().hasFormat('application/calibre+from_tag_browser') + item = index.data(Qt.UserRole).toPyObject() + flags = self._model.flags(index) + if item.type == TagTreeItem.TAG and flags & Qt.ItemIsDropEnabled: + self.setDropIndicatorShown(not src_is_tb) + return + if item.type == TagTreeItem.CATEGORY and not item.is_gst: + fm_dest = self.db.metadata_for_field(item.category_key) + if fm_dest['kind'] == 'user': + if src_is_tb: + if event.dropAction() == Qt.MoveAction: + data = str(event.mimeData().data('application/calibre+from_tag_browser')) + src = cPickle.loads(data) + for s in src: + if s[0] == TagTreeItem.TAG and \ + (not s[1].startswith('@') or s[2]): + return + self.setDropIndicatorShown(True) + return + md = event.mimeData() + if hasattr(md, 'column_name'): + fm_src = self.db.metadata_for_field(md.column_name) + if md.column_name in ['authors', 'publisher', 'series'] or \ + (fm_src['is_custom'] and ( + (fm_src['datatype'] in ['series', 'text', 'enumeration'] and + not fm_src['is_multiple']) or + (fm_src['datatype'] == 'composite' and + fm_src['display'].get('make_category', False)))): + self.setDropIndicatorShown(True) + + def clear(self): + if self.model(): + self.model().clear_state() + + def is_visible(self, idx): + item = idx.data(Qt.UserRole).toPyObject() + if getattr(item, 'type', None) == TagTreeItem.TAG: + idx = idx.parent() + return self.isExpanded(idx) + + def recount(self, *args): + ''' + Rebuild the category tree, expand any categories that were expanded, + reset the search states, and reselect the current node. + ''' + if self.disable_recounting or not self.pane_is_visible: + return + self.refresh_signal_processed = True + ci = self.currentIndex() + if not ci.isValid(): + ci = self.indexAt(QPoint(10, 10)) + path = self.model().path_for_index(ci) if self.is_visible(ci) else None + expanded_categories, state_map = self.get_state() + self._model.rebuild_node_tree(state_map=state_map) + for category in expanded_categories: + self.expand(self._model.index_for_category(category)) + self.show_item_at_path(path) + + def show_item_at_path(self, path, box=False, + position=QTreeView.PositionAtCenter): + ''' + Scroll the browser and open categories to show the item referenced by + path. If possible, the item is placed in the center. If box=True, a + box is drawn around the item. + ''' + if path: + self.show_item_at_index(self._model.index_for_path(path), box=box, + position=position) + + def show_item_at_index(self, idx, box=False, + position=QTreeView.PositionAtCenter): + if idx.isValid(): + self.setCurrentIndex(idx) + self.scrollTo(idx, position) + self.setCurrentIndex(idx) + if box: + self._model.set_boxed(idx) + + def item_expanded(self, idx): + ''' + Called by the expanded signal + ''' + self.setCurrentIndex(idx) + + # }}} + + diff --git a/src/calibre/gui2/tag_view.py b/src/calibre/gui2/tag_view.py deleted file mode 100644 index 730c91d37f..0000000000 --- a/src/calibre/gui2/tag_view.py +++ /dev/null @@ -1,2301 +0,0 @@ -#!/usr/bin/env python -__license__ = 'GPL v3' -__copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net' -__docformat__ = 'restructuredtext en' - -''' -Browsing book collection by tags. -''' - -import traceback, copy, cPickle - -from itertools import izip, repeat -from functools import partial - -from PyQt4.Qt import (Qt, QTreeView, QApplication, pyqtSignal, QFont, QSize, - QIcon, QPoint, QVBoxLayout, QHBoxLayout, QComboBox, QTimer, - QAbstractItemModel, QVariant, QModelIndex, QMenu, QFrame, - QWidget, QItemDelegate, QString, QLabel, QPushButton, - QShortcut, QKeySequence, SIGNAL, QMimeData, QToolButton) - -from calibre.ebooks.metadata import title_sort -from calibre.gui2 import config, NONE, gprefs -from calibre.library.field_metadata import TagsIcons, category_icon_map -from calibre.library.database2 import Tag -from calibre.utils.config import tweaks -from calibre.utils.icu import sort_key, lower, strcmp -from calibre.utils.search_query_parser import saved_searches -from calibre.utils.formatter import eval_formatter -from calibre.gui2 import error_dialog, question_dialog -from calibre.gui2.dialogs.confirm_delete import confirm -from calibre.gui2.dialogs.tag_categories import TagCategories -from calibre.gui2.dialogs.tag_list_editor import TagListEditor -from calibre.gui2.dialogs.edit_authors_dialog import EditAuthorsDialog -from calibre.gui2.widgets import HistoryLineEdit - -class TagDelegate(QItemDelegate): # {{{ - - def paint(self, painter, option, index): - item = index.internalPointer() - if item.type != TagTreeItem.TAG: - QItemDelegate.paint(self, painter, option, index) - return - r = option.rect - model = self.parent().model() - icon = model.data(index, Qt.DecorationRole).toPyObject() - painter.save() - if item.tag.state != 0 or not config['show_avg_rating'] or \ - item.tag.avg_rating is None: - icon.paint(painter, r, Qt.AlignLeft) - else: - painter.setOpacity(0.3) - icon.paint(painter, r, Qt.AlignLeft) - painter.setOpacity(1) - rating = item.tag.avg_rating - painter.setClipRect(r.left(), r.bottom()-int(r.height()*(rating/5.0)), - r.width(), r.height()) - icon.paint(painter, r, Qt.AlignLeft) - painter.setClipRect(r) - - # Paint the text - if item.boxed: - painter.drawRoundedRect(r.adjusted(1,1,-1,-1), 5, 5) - r.setLeft(r.left()+r.height()+3) - painter.drawText(r, Qt.AlignLeft|Qt.AlignVCenter, - model.data(index, Qt.DisplayRole).toString()) - painter.restore() - - # }}} - -TAG_SEARCH_STATES = {'clear': 0, 'mark_plus': 1, 'mark_plusplus': 2, - 'mark_minus': 3, 'mark_minusminus': 4} - -class TagsView(QTreeView): # {{{ - - refresh_required = pyqtSignal() - tags_marked = pyqtSignal(object) - edit_user_category = pyqtSignal(object) - delete_user_category = pyqtSignal(object) - del_item_from_user_cat = pyqtSignal(object, object, object) - add_item_to_user_cat = pyqtSignal(object, object, object) - add_subcategory = pyqtSignal(object) - tag_list_edit = pyqtSignal(object, object) - saved_search_edit = pyqtSignal(object) - rebuild_saved_searches = pyqtSignal() - author_sort_edit = pyqtSignal(object, object) - tag_item_renamed = pyqtSignal() - search_item_renamed = pyqtSignal() - drag_drop_finished = pyqtSignal(object) - restriction_error = pyqtSignal() - - def __init__(self, parent=None): - QTreeView.__init__(self, parent=None) - self.tag_match = None - self.disable_recounting = False - self.setUniformRowHeights(True) - self.setCursor(Qt.PointingHandCursor) - self.setIconSize(QSize(30, 30)) - self.setTabKeyNavigation(True) - self.setAlternatingRowColors(True) - self.setAnimated(True) - self.setHeaderHidden(True) - self.setItemDelegate(TagDelegate(self)) - self.made_connections = False - self.setAcceptDrops(True) - self.setDragEnabled(True) - self.setDragDropMode(self.DragDrop) - self.setDropIndicatorShown(True) - self.setAutoExpandDelay(500) - self.pane_is_visible = False - if gprefs['tags_browser_collapse_at'] == 0: - self.collapse_model = 'disable' - else: - self.collapse_model = gprefs['tags_browser_partition_method'] - self.search_icon = QIcon(I('search.png')) - self.user_category_icon = QIcon(I('tb_folder.png')) - self.delete_icon = QIcon(I('list_remove.png')) - self.rename_icon = QIcon(I('edit-undo.png')) - - def set_pane_is_visible(self, to_what): - pv = self.pane_is_visible - self.pane_is_visible = to_what - if to_what and not pv: - self.recount() - - def reread_collapse_parameters(self): - if gprefs['tags_browser_collapse_at'] == 0: - self.collapse_model = 'disable' - else: - self.collapse_model = gprefs['tags_browser_partition_method'] - self.set_new_model(self._model.get_filter_categories_by()) - - def set_database(self, db, tag_match, sort_by): - hidden_cats = db.prefs.get('tag_browser_hidden_categories', None) - self.hidden_categories = [] - # migrate from config to db prefs - if hidden_cats is None: - hidden_cats = config['tag_browser_hidden_categories'] - # strip out any non-existence field keys - for cat in hidden_cats: - if cat in db.field_metadata: - self.hidden_categories.append(cat) - db.prefs.set('tag_browser_hidden_categories', list(self.hidden_categories)) - self.hidden_categories = set(self.hidden_categories) - - old = getattr(self, '_model', None) - if old is not None: - old.break_cycles() - self._model = TagsModel(db, parent=self, - hidden_categories=self.hidden_categories, - search_restriction=None, - drag_drop_finished=self.drag_drop_finished, - collapse_model=self.collapse_model, - state_map={}) - self.pane_is_visible = True # because TagsModel.init did a recount - self.sort_by = sort_by - self.tag_match = tag_match - self.db = db - self.search_restriction = None - self.setModel(self._model) - self.setContextMenuPolicy(Qt.CustomContextMenu) - pop = config['sort_tags_by'] - self.sort_by.setCurrentIndex(self.db.CATEGORY_SORTS.index(pop)) - try: - match_pop = self.db.MATCH_TYPE.index(config['match_tags_type']) - except ValueError: - match_pop = 0 - self.tag_match.setCurrentIndex(match_pop) - if not self.made_connections: - self.clicked.connect(self.toggle) - self.customContextMenuRequested.connect(self.show_context_menu) - self.refresh_required.connect(self.recount, type=Qt.QueuedConnection) - self.sort_by.currentIndexChanged.connect(self.sort_changed) - self.tag_match.currentIndexChanged.connect(self.match_changed) - self.made_connections = True - self.refresh_signal_processed = True - db.add_listener(self.database_changed) - self.expanded.connect(self.item_expanded) - - def database_changed(self, event, ids): - if self.refresh_signal_processed: - self.refresh_signal_processed = False - self.refresh_required.emit() - - @property - def match_all(self): - return self.tag_match and self.tag_match.currentIndex() > 0 - - def sort_changed(self, pop): - config.set('sort_tags_by', self.db.CATEGORY_SORTS[pop]) - self.recount() - - def match_changed(self, pop): - try: - config.set('match_tags_type', self.db.MATCH_TYPE[pop]) - except: - pass - - def set_search_restriction(self, s): - if s: - self.search_restriction = s - else: - self.search_restriction = None - self.set_new_model() - - def mouseReleaseEvent(self, event): - # Swallow everything except leftButton so context menus work correctly - if event.button() == Qt.LeftButton: - QTreeView.mouseReleaseEvent(self, event) - - def mouseDoubleClickEvent(self, event): - # swallow these to avoid toggling and editing at the same time - pass - - @property - def search_string(self): - tokens = self._model.tokens() - joiner = ' and ' if self.match_all else ' or ' - return joiner.join(tokens) - - def toggle(self, index): - self._toggle(index, None) - - def _toggle(self, index, set_to): - ''' - set_to: if None, advance the state. Otherwise must be one of the values - in TAG_SEARCH_STATES - ''' - modifiers = int(QApplication.keyboardModifiers()) - exclusive = modifiers not in (Qt.CTRL, Qt.SHIFT) - if self._model.toggle(index, exclusive, set_to=set_to): - self.tags_marked.emit(self.search_string) - - def conditional_clear(self, search_string): - if search_string != self.search_string: - self.clear() - - def context_menu_handler(self, action=None, category=None, - key=None, index=None, search_state=None): - if not action: - return - try: - if action == 'edit_item': - self.edit(index) - return - if action == 'open_editor': - self.tag_list_edit.emit(category, key) - return - if action == 'manage_categories': - self.edit_user_category.emit(category) - return - if action == 'search': - self._toggle(index, set_to=search_state) - return - if action == 'add_to_category': - tag = index.tag - if len(index.children) > 0: - for c in index.children: - self.add_item_to_user_cat.emit(category, c.tag.original_name, - c.tag.category) - self.add_item_to_user_cat.emit(category, tag.original_name, - tag.category) - return - if action == 'add_subcategory': - self.add_subcategory.emit(key) - return - if action == 'search_category': - self._toggle(index, set_to=search_state) - return - if action == 'delete_user_category': - self.delete_user_category.emit(key) - return - if action == 'delete_search': - saved_searches().delete(key) - self.rebuild_saved_searches.emit() - return - if action == 'delete_item_from_user_category': - tag = index.tag - if len(index.children) > 0: - for c in index.children: - self.del_item_from_user_cat.emit(key, c.tag.original_name, - c.tag.category) - self.del_item_from_user_cat.emit(key, tag.original_name, tag.category) - return - if action == 'manage_searches': - self.saved_search_edit.emit(category) - return - if action == 'edit_author_sort': - self.author_sort_edit.emit(self, index) - return - - if action == 'hide': - self.hidden_categories.add(category) - elif action == 'show': - self.hidden_categories.discard(category) - elif action == 'categorization': - changed = self.collapse_model != category - self.collapse_model = category - if changed: - self.set_new_model(self._model.get_filter_categories_by()) - gprefs['tags_browser_partition_method'] = category - elif action == 'defaults': - self.hidden_categories.clear() - self.db.prefs.set('tag_browser_hidden_categories', list(self.hidden_categories)) - self.set_new_model() - except: - return - - def show_context_menu(self, point): - def display_name( tag): - if tag.category == 'search': - n = tag.name - if len(n) > 45: - n = n[:45] + '...' - return "'" + n + "'" - return tag.name - - index = self.indexAt(point) - self.context_menu = QMenu(self) - - if index.isValid(): - item = index.internalPointer() - tag = None - - if item.type == TagTreeItem.TAG: - tag_item = item - tag = item.tag - while item.type != TagTreeItem.CATEGORY: - item = item.parent - - if item.type == TagTreeItem.CATEGORY: - if not item.category_key.startswith('@'): - while item.parent != self._model.root_item: - item = item.parent - category = unicode(item.name.toString()) - key = item.category_key - # Verify that we are working with a field that we know something about - if key not in self.db.field_metadata: - return True - - # Did the user click on a leaf node? - if tag: - # If the user right-clicked on an editable item, then offer - # the possibility of renaming that item. - if tag.is_editable: - # Add the 'rename' items - self.context_menu.addAction(self.rename_icon, - _('Rename %s')%display_name(tag), - partial(self.context_menu_handler, action='edit_item', - index=index)) - if key == 'authors': - self.context_menu.addAction(_('Edit sort for %s')%display_name(tag), - partial(self.context_menu_handler, - action='edit_author_sort', index=tag.id)) - - # is_editable is also overloaded to mean 'can be added - # to a user category' - m = self.context_menu.addMenu(self.user_category_icon, - _('Add %s to user category')%display_name(tag)) - nt = self.model().category_node_tree - def add_node_tree(tree_dict, m, path): - p = path[:] - for k in sorted(tree_dict.keys(), key=sort_key): - p.append(k) - n = k[1:] if k.startswith('@') else k - m.addAction(self.user_category_icon, n, - partial(self.context_menu_handler, - 'add_to_category', - category='.'.join(p), index=tag_item)) - if len(tree_dict[k]): - tm = m.addMenu(self.user_category_icon, - _('Children of %s')%n) - add_node_tree(tree_dict[k], tm, p) - p.pop() - add_node_tree(nt, m, []) - elif key == 'search': - self.context_menu.addAction(self.rename_icon, - _('Rename %s')%display_name(tag), - partial(self.context_menu_handler, action='edit_item', - index=index)) - self.context_menu.addAction(self.delete_icon, - _('Delete search %s')%display_name(tag), - partial(self.context_menu_handler, - action='delete_search', key=tag.name)) - if key.startswith('@') and not item.is_gst: - self.context_menu.addAction(self.user_category_icon, - _('Remove %s from category %s')% - (display_name(tag), item.py_name), - partial(self.context_menu_handler, - action='delete_item_from_user_category', - key = key, index = tag_item)) - # Add the search for value items. All leaf nodes are searchable - self.context_menu.addAction(self.search_icon, - _('Search for %s')%display_name(tag), - partial(self.context_menu_handler, action='search', - search_state=TAG_SEARCH_STATES['mark_plus'], - index=index)) - self.context_menu.addAction(self.search_icon, - _('Search for everything but %s')%display_name(tag), - partial(self.context_menu_handler, action='search', - search_state=TAG_SEARCH_STATES['mark_minus'], - index=index)) - self.context_menu.addSeparator() - elif key.startswith('@') and not item.is_gst: - if item.can_be_edited: - self.context_menu.addAction(self.rename_icon, - _('Rename %s')%item.py_name, - partial(self.context_menu_handler, action='edit_item', - index=index)) - self.context_menu.addAction(self.user_category_icon, - _('Add sub-category to %s')%item.py_name, - partial(self.context_menu_handler, - action='add_subcategory', key=key)) - self.context_menu.addAction(self.delete_icon, - _('Delete user category %s')%item.py_name, - partial(self.context_menu_handler, - action='delete_user_category', key=key)) - self.context_menu.addSeparator() - # Hide/Show/Restore categories - self.context_menu.addAction(_('Hide category %s') % category, - partial(self.context_menu_handler, action='hide', - category=key)) - if self.hidden_categories: - m = self.context_menu.addMenu(_('Show category')) - for col in sorted(self.hidden_categories, - key=lambda x: sort_key(self.db.field_metadata[x]['name'])): - m.addAction(self.db.field_metadata[col]['name'], - partial(self.context_menu_handler, action='show', category=col)) - - # search by category. Some categories are not searchable, such - # as search and news - if item.tag.is_searchable: - self.context_menu.addAction(self.search_icon, - _('Search for books in category %s')%category, - partial(self.context_menu_handler, - action='search_category', - index=self._model.createIndex(item.row(), 0, item), - search_state=TAG_SEARCH_STATES['mark_plus'])) - self.context_menu.addAction(self.search_icon, - _('Search for books not in category %s')%category, - partial(self.context_menu_handler, - action='search_category', - index=self._model.createIndex(item.row(), 0, item), - search_state=TAG_SEARCH_STATES['mark_minus'])) - # Offer specific editors for tags/series/publishers/saved searches - self.context_menu.addSeparator() - if key in ['tags', 'publisher', 'series'] or \ - self.db.field_metadata[key]['is_custom']: - self.context_menu.addAction(_('Manage %s')%category, - partial(self.context_menu_handler, action='open_editor', - category=tag.original_name if tag else None, - key=key)) - elif key == 'authors': - self.context_menu.addAction(_('Manage %s')%category, - partial(self.context_menu_handler, action='edit_author_sort')) - elif key == 'search': - self.context_menu.addAction(_('Manage Saved Searches'), - partial(self.context_menu_handler, action='manage_searches', - category=tag.name if tag else None)) - - # Always show the user categories editor - self.context_menu.addSeparator() - if key.startswith('@') and \ - key[1:] in self.db.prefs.get('user_categories', {}).keys(): - self.context_menu.addAction(_('Manage User Categories'), - partial(self.context_menu_handler, action='manage_categories', - category=key[1:])) - else: - self.context_menu.addAction(_('Manage User Categories'), - partial(self.context_menu_handler, action='manage_categories', - category=None)) - - if self.hidden_categories: - if not self.context_menu.isEmpty(): - self.context_menu.addSeparator() - self.context_menu.addAction(_('Show all categories'), - partial(self.context_menu_handler, action='defaults')) - - m = self.context_menu.addMenu(_('Change sub-categorization scheme')) - da = m.addAction('Disable', - partial(self.context_menu_handler, action='categorization', category='disable')) - fla = m.addAction('By first letter', - partial(self.context_menu_handler, action='categorization', category='first letter')) - pa = m.addAction('Partition', - partial(self.context_menu_handler, action='categorization', category='partition')) - if self.collapse_model == 'disable': - da.setCheckable(True) - da.setChecked(True) - elif self.collapse_model == 'first letter': - fla.setCheckable(True) - fla.setChecked(True) - else: - pa.setCheckable(True) - pa.setChecked(True) - - if not self.context_menu.isEmpty(): - self.context_menu.popup(self.mapToGlobal(point)) - return True - - def dragMoveEvent(self, event): - QTreeView.dragMoveEvent(self, event) - self.setDropIndicatorShown(False) - index = self.indexAt(event.pos()) - if not index.isValid(): - return - src_is_tb = event.mimeData().hasFormat('application/calibre+from_tag_browser') - item = index.internalPointer() - flags = self._model.flags(index) - if item.type == TagTreeItem.TAG and flags & Qt.ItemIsDropEnabled: - self.setDropIndicatorShown(not src_is_tb) - return - if item.type == TagTreeItem.CATEGORY and not item.is_gst: - fm_dest = self.db.metadata_for_field(item.category_key) - if fm_dest['kind'] == 'user': - if src_is_tb: - if event.dropAction() == Qt.MoveAction: - data = str(event.mimeData().data('application/calibre+from_tag_browser')) - src = cPickle.loads(data) - for s in src: - if s[0] == TagTreeItem.TAG and \ - (not s[1].startswith('@') or s[2]): - return - self.setDropIndicatorShown(True) - return - md = event.mimeData() - if hasattr(md, 'column_name'): - fm_src = self.db.metadata_for_field(md.column_name) - if md.column_name in ['authors', 'publisher', 'series'] or \ - (fm_src['is_custom'] and ( - (fm_src['datatype'] in ['series', 'text', 'enumeration'] and - not fm_src['is_multiple']) or - (fm_src['datatype'] == 'composite' and - fm_src['display'].get('make_category', False)))): - self.setDropIndicatorShown(True) - - def clear(self): - if self.model(): - self.model().clear_state() - - def is_visible(self, idx): - item = idx.internalPointer() - if getattr(item, 'type', None) == TagTreeItem.TAG: - idx = idx.parent() - return self.isExpanded(idx) - - def recount(self, *args): - ''' - Rebuild the category tree, expand any categories that were expanded, - reset the search states, and reselect the current node. - ''' - if self.disable_recounting or not self.pane_is_visible: - return - self.refresh_signal_processed = True - ci = self.currentIndex() - if not ci.isValid(): - ci = self.indexAt(QPoint(10, 10)) - path = self.model().path_for_index(ci) if self.is_visible(ci) else None - expanded_categories, state_map = self.model().get_state() - self.set_new_model(state_map=state_map) - for category in expanded_categories: - self.expand(self.model().index_for_category(category)) - self._model.show_item_at_path(path) - - def item_expanded(self, idx): - ''' - Called by the expanded signal - ''' - self.setCurrentIndex(idx) - - def set_new_model(self, filter_categories_by=None, state_map={}): - ''' - There are cases where we need to rebuild the category tree without - attempting to reposition the current node. - ''' - try: - old = getattr(self, '_model', None) - if old is not None: - old.break_cycles() - self._model = TagsModel(self.db, parent=self, - hidden_categories=self.hidden_categories, - search_restriction=self.search_restriction, - drag_drop_finished=self.drag_drop_finished, - filter_categories_by=filter_categories_by, - collapse_model=self.collapse_model, - state_map=state_map) - self.setModel(self._model) - except: - # The DB must be gone. Set the model to None and hope that someone - # will call set_database later. I don't know if this in fact works. - # But perhaps a Bad Thing Happened, so print the exception - traceback.print_exc() - self._model = None - self.setModel(None) - # }}} - -class TagTreeItem(object): # {{{ - - CATEGORY = 0 - TAG = 1 - ROOT = 2 - - def __init__(self, data=None, category_icon=None, icon_map=None, - parent=None, tooltip=None, category_key=None, temporary=False): - self.parent = parent - self.children = [] - self.id_set = set() - self.is_gst = False - self.boxed = False - self.icon_state_map = list(map(QVariant, icon_map)) - if self.parent is not None: - self.parent.append(self) - if data is None: - self.type = self.ROOT - else: - self.type = self.TAG if category_icon is None else self.CATEGORY - if self.type == self.CATEGORY: - self.name, self.icon = map(QVariant, (data, category_icon)) - self.py_name = data - self.bold_font = QFont() - self.bold_font.setBold(True) - self.bold_font = QVariant(self.bold_font) - self.category_key = category_key - self.temporary = temporary - self.tag = Tag(data, category=category_key, - is_editable=category_key not in ['news', 'search', 'identifiers'], - is_searchable=category_key not in ['search']) - - elif self.type == self.TAG: - self.icon_state_map[0] = QVariant(data.icon) - self.tag = data - if tooltip: - self.tooltip = tooltip + ' ' - else: - self.tooltip = '' - - def break_cycles(self): - for x in self.children: - try: - x.break_cycles() - except: - pass - self.parent = self.icon_state_map = self.bold_font = self.tag = \ - self.icon = self.children = self.tooltip = \ - self.py_name = self.id_set = self.category_key = None - - def __str__(self): - if self.type == self.ROOT: - return 'ROOT' - if self.type == self.CATEGORY: - return 'CATEGORY:'+str(QVariant.toString(self.name))+':%d'%len(self.children) - return 'TAG:'+self.tag.name - - def row(self): - if self.parent is not None: - return self.parent.children.index(self) - return 0 - - def append(self, child): - child.parent = self - self.children.append(child) - - def data(self, role): - if role == Qt.UserRole: - return self - if self.type == self.TAG: - return self.tag_data(role) - if self.type == self.CATEGORY: - return self.category_data(role) - return NONE - - def category_data(self, role): - if role == Qt.DisplayRole: - return QVariant(self.py_name + ' [%d]'%len(self.child_tags())) - if role == Qt.EditRole: - return QVariant(self.py_name) - if role == Qt.DecorationRole: - if self.tag.state: - return self.icon_state_map[self.tag.state] - return self.icon - if role == Qt.FontRole: - return self.bold_font - if role == Qt.ToolTipRole and self.tooltip is not None: - return QVariant(self.tooltip) - return NONE - - def tag_data(self, role): - tag = self.tag - if tag.use_sort_as_name: - name = tag.sort - tt_author = True - else: - p = self - while p.parent.type != self.ROOT: - p = p.parent - if not tag.is_hierarchical: - name = tag.original_name - else: - name = tag.name - tt_author = False - if role == Qt.DisplayRole: - count = len(self.id_set) - count = count if count > 0 else tag.count - if count == 0: - return QVariant('%s'%(name)) - else: - return QVariant('[%d] %s'%(count, name)) - if role == Qt.EditRole: - return QVariant(tag.original_name) - if role == Qt.DecorationRole: - return self.icon_state_map[tag.state] - if role == Qt.ToolTipRole: - if tt_author: - if tag.tooltip is not None: - return QVariant('(%s) %s'%(tag.name, tag.tooltip)) - else: - return QVariant(tag.name) - if tag.tooltip: - return QVariant(self.tooltip + tag.tooltip) - else: - return QVariant(self.tooltip) - return NONE - - def toggle(self, set_to=None): - ''' - set_to: None => advance the state, otherwise a value from TAG_SEARCH_STATES - ''' - if set_to is None: - while True: - self.tag.state = (self.tag.state + 1)%5 - if self.tag.state == TAG_SEARCH_STATES['mark_plus'] or \ - self.tag.state == TAG_SEARCH_STATES['mark_minus']: - if self.tag.is_searchable: - break - elif self.tag.state == TAG_SEARCH_STATES['mark_plusplus'] or\ - self.tag.state == TAG_SEARCH_STATES['mark_minusminus']: - if self.tag.is_searchable and len(self.children) and \ - self.tag.is_hierarchical == '5state': - break - else: - break - else: - self.tag.state = set_to - - def all_children(self): - res = [] - def recurse(nodes, res): - for t in nodes: - res.append(t) - recurse(t.children, res) - recurse(self.children, res) - return res - - def child_tags(self): - res = [] - def recurse(nodes, res): - for t in nodes: - if t.type != TagTreeItem.CATEGORY: - res.append(t) - recurse(t.children, res) - recurse(self.children, res) - return res - # }}} - -class TagsModel(QAbstractItemModel): # {{{ - - def __init__(self, db, parent, hidden_categories=None, - search_restriction=None, drag_drop_finished=None, - filter_categories_by=None, collapse_model='disable', - state_map={}): - QAbstractItemModel.__init__(self, parent) - - # must do this here because 'QPixmap: Must construct a QApplication - # before a QPaintDevice'. The ':' at the end avoids polluting either of - # the other namespaces (alpha, '#', or '@') - iconmap = {} - for key in category_icon_map: - iconmap[key] = QIcon(I(category_icon_map[key])) - self.category_icon_map = TagsIcons(iconmap) - - self.categories_with_ratings = ['authors', 'series', 'publisher', 'tags'] - self.drag_drop_finished = drag_drop_finished - - self.icon_state_map = [None, QIcon(I('plus.png')), QIcon(I('plusplus.png')), - QIcon(I('minus.png')), QIcon(I('minusminus.png'))] - self.db = db - self.tags_view = parent - self.hidden_categories = hidden_categories - self.search_restriction = search_restriction - self.row_map = [] - self.filter_categories_by = filter_categories_by - self.collapse_model = collapse_model - - # Note that _get_category_nodes can indirectly change the - # user_categories dict. - - data = self._get_category_nodes(config['sort_tags_by']) - gst = db.prefs.get('grouped_search_terms', {}) - self.root_item = TagTreeItem(icon_map=self.icon_state_map) - self.category_nodes = [] - - last_category_node = None - category_node_map = {} - self.category_node_tree = {} - for i, key in enumerate(self.row_map): - if self.hidden_categories: - if key in self.hidden_categories: - continue - found = False - for cat in self.hidden_categories: - if cat.startswith('@') and key.startswith(cat + '.'): - found = True - if found: - continue - is_gst = False - if key.startswith('@') and key[1:] in gst: - tt = _(u'The grouped search term name is "{0}"').format(key[1:]) - is_gst = True - elif key == 'news': - tt = '' - else: - tt = _(u'The lookup/search name is "{0}"').format(key) - - if key.startswith('@'): - path_parts = [p for p in key.split('.')] - path = '' - last_category_node = self.root_item - tree_root = self.category_node_tree - for i,p in enumerate(path_parts): - path += p - if path not in category_node_map: - node = TagTreeItem(parent=last_category_node, - data=p[1:] if i == 0 else p, - category_icon=self.category_icon_map[key], - tooltip=tt if path == key else path, - category_key=path, - icon_map=self.icon_state_map) - last_category_node = node - category_node_map[path] = node - self.category_nodes.append(node) - node.can_be_edited = (not is_gst) and (i == (len(path_parts)-1)) - node.is_gst = is_gst - if not is_gst: - node.tag.is_hierarchical = '5state' - if not is_gst: - tree_root[p] = {} - tree_root = tree_root[p] - else: - last_category_node = category_node_map[path] - tree_root = tree_root[p] - path += '.' - else: - node = TagTreeItem(parent=self.root_item, - data=self.categories[key], - category_icon=self.category_icon_map[key], - tooltip=tt, category_key=key, - icon_map=self.icon_state_map) - node.is_gst = False - category_node_map[key] = node - last_category_node = node - self.category_nodes.append(node) - self._create_node_tree(data, state_map) - - def break_cycles(self): - self.root_item.break_cycles() - self.db = self.root_item = None - - def mimeTypes(self): - return ["application/calibre+from_library", - 'application/calibre+from_tag_browser'] - - def mimeData(self, indexes): - data = [] - for idx in indexes: - if idx.isValid(): - # get some useful serializable data - node = idx.internalPointer() - path = self.path_for_index(idx) - if node.type == TagTreeItem.CATEGORY: - d = (node.type, node.py_name, node.category_key) - else: - t = node.tag - p = node - while p.type != TagTreeItem.CATEGORY: - p = p.parent - d = (node.type, p.category_key, p.is_gst, t.original_name, - t.category, path) - data.append(d) - else: - data.append(None) - raw = bytearray(cPickle.dumps(data, -1)) - ans = QMimeData() - ans.setData('application/calibre+from_tag_browser', raw) - return ans - - def dropMimeData(self, md, action, row, column, parent): - fmts = set([unicode(x) for x in md.formats()]) - if not fmts.intersection(set(self.mimeTypes())): - return False - if "application/calibre+from_library" in fmts: - if action != Qt.CopyAction: - return False - return self.do_drop_from_library(md, action, row, column, parent) - elif 'application/calibre+from_tag_browser' in fmts: - return self.do_drop_from_tag_browser(md, action, row, column, parent) - - def do_drop_from_tag_browser(self, md, action, row, column, parent): - if not parent.isValid(): - return False - dest = parent.internalPointer() - if dest.type != TagTreeItem.CATEGORY: - return False - if not md.hasFormat('application/calibre+from_tag_browser'): - return False - data = str(md.data('application/calibre+from_tag_browser')) - src = cPickle.loads(data) - for s in src: - if s[0] != TagTreeItem.TAG: - return False - return self.move_or_copy_item_to_user_category(src, dest, action) - - def move_or_copy_item_to_user_category(self, src, dest, action): - ''' - src is a list of tuples representing items to copy. The tuple is - (type, containing category key, category key is global search term, - full name, category key, path to node) - The type must be TagTreeItem.TAG - dest is the TagTreeItem node to receive the items - action is Qt.CopyAction or Qt.MoveAction - ''' - def process_source_node(user_cats, src_parent, src_parent_is_gst, - is_uc, dest_key, node): - ''' - Copy/move an item and all its children to the destination - ''' - copied = False - src_name = node.tag.original_name - src_cat = node.tag.category - # delete the item if the source is a user category and action is move - if is_uc and not src_parent_is_gst and src_parent in user_cats and \ - action == Qt.MoveAction: - new_cat = [] - for tup in user_cats[src_parent]: - if src_name == tup[0] and src_cat == tup[1]: - continue - new_cat.append(list(tup)) - user_cats[src_parent] = new_cat - else: - copied = True - - # Now add the item to the destination user category - add_it = True - if not is_uc and src_cat == 'news': - src_cat = 'tags' - for tup in user_cats[dest_key]: - if src_name == tup[0] and src_cat == tup[1]: - add_it = False - if add_it: - user_cats[dest_key].append([src_name, src_cat, 0]) - - for c in node.children: - copied = process_source_node(user_cats, src_parent, src_parent_is_gst, - is_uc, dest_key, c) - return copied - - user_cats = self.db.prefs.get('user_categories', {}) - parent_node = None - copied = False - path = None - for s in src: - src_parent, src_parent_is_gst = s[1:3] - path = s[5] - parent_node = src_parent - - if src_parent.startswith('@'): - is_uc = True - src_parent = src_parent[1:] - else: - is_uc = False - dest_key = dest.category_key[1:] - - if dest_key not in user_cats: - continue - - node = self.index_for_path(path) - if node: - copied = process_source_node(user_cats, src_parent, src_parent_is_gst, - is_uc, dest_key, node.internalPointer()) - - self.db.prefs.set('user_categories', user_cats) - self.tags_view.recount() - - # Scroll to the item copied. If it was moved, scroll to the parent - if parent_node is not None: - self.clear_boxed() - m = self.tags_view.model() - if not copied: - p = path[-1] - if p == 0: - path = m.find_category_node(parent_node) - else: - path[-1] = p - 1 - idx = m.index_for_path(path) - self.tags_view.setExpanded(idx, True) - if idx.internalPointer().type == TagTreeItem.TAG: - m.show_item_at_index(idx, box=True) - else: - m.show_item_at_index(idx) - return True - - def do_drop_from_library(self, md, action, row, column, parent): - idx = parent - if idx.isValid(): - self.tags_view.setCurrentIndex(idx) - node = self.data(idx, Qt.UserRole) - if node.type == TagTreeItem.TAG: - fm = self.db.metadata_for_field(node.tag.category) - if node.tag.category in \ - ('tags', 'series', 'authors', 'rating', 'publisher') or \ - (fm['is_custom'] and ( - fm['datatype'] in ['text', 'rating', 'series', - 'enumeration'] or - (fm['datatype'] == 'composite' and - fm['display'].get('make_category', False)))): - mime = 'application/calibre+from_library' - ids = list(map(int, str(md.data(mime)).split())) - self.handle_drop(node, ids) - return True - elif node.type == TagTreeItem.CATEGORY: - fm_dest = self.db.metadata_for_field(node.category_key) - if fm_dest['kind'] == 'user': - fm_src = self.db.metadata_for_field(md.column_name) - if md.column_name in ['authors', 'publisher', 'series'] or \ - (fm_src['is_custom'] and ( - (fm_src['datatype'] in ['series', 'text', 'enumeration'] and - not fm_src['is_multiple']))or - (fm_src['datatype'] == 'composite' and - fm_src['display'].get('make_category', False))): - mime = 'application/calibre+from_library' - ids = list(map(int, str(md.data(mime)).split())) - self.handle_user_category_drop(node, ids, md.column_name) - return True - return False - - def handle_user_category_drop(self, on_node, ids, column): - categories = self.db.prefs.get('user_categories', {}) - category = categories.get(on_node.category_key[1:], None) - if category is None: - return - fm_src = self.db.metadata_for_field(column) - for id in ids: - label = fm_src['label'] - if not fm_src['is_custom']: - if label == 'authors': - items = self.db.get_authors_with_ids() - items = [(i[0], i[1].replace('|', ',')) for i in items] - value = self.db.authors(id, index_is_id=True) - value = [v.replace('|', ',') for v in value.split(',')] - elif label == 'publisher': - items = self.db.get_publishers_with_ids() - value = self.db.publisher(id, index_is_id=True) - elif label == 'series': - items = self.db.get_series_with_ids() - value = self.db.series(id, index_is_id=True) - else: - items = self.db.get_custom_items_with_ids(label=label) - if fm_src['datatype'] != 'composite': - value = self.db.get_custom(id, label=label, index_is_id=True) - else: - value = self.db.get_property(id, loc=fm_src['rec_index'], - index_is_id=True) - if value is None: - return - if not isinstance(value, list): - value = [value] - for val in value: - for (v, c, id) in category: - if v == val and c == column: - break - else: - category.append([val, column, 0]) - categories[on_node.category_key[1:]] = category - self.db.prefs.set('user_categories', categories) - self.tags_view.recount() - - def handle_drop(self, on_node, ids): - #print 'Dropped ids:', ids, on_node.tag - key = on_node.tag.category - if (key == 'authors' and len(ids) >= 5): - if not confirm('

'+_('Changing the authors for several books can ' - 'take a while. Are you sure?') - +'

', 'tag_browser_drop_authors', self.tags_view): - return - elif len(ids) > 15: - if not confirm('

'+_('Changing the metadata for that many books ' - 'can take a while. Are you sure?') - +'

', 'tag_browser_many_changes', self.tags_view): - return - - fm = self.db.metadata_for_field(key) - is_multiple = fm['is_multiple'] - val = on_node.tag.original_name - for id in ids: - mi = self.db.get_metadata(id, index_is_id=True) - - # Prepare to ignore the author, unless it is changed. Title is - # always ignored -- see the call to set_metadata - set_authors = False - - # Author_sort cannot change explicitly. Changing the author might - # change it. - mi.author_sort = None # Never will change by itself. - - if key == 'authors': - mi.authors = [val] - set_authors=True - elif fm['datatype'] == 'rating': - mi.set(key, len(val) * 2) - elif fm['is_custom'] and fm['datatype'] == 'series': - mi.set(key, val, extra=1.0) - elif is_multiple: - new_val = mi.get(key, []) - if val in new_val: - # Fortunately, only one field can change, so the continue - # won't break anything - continue - new_val.append(val) - mi.set(key, new_val) - else: - mi.set(key, val) - self.db.set_metadata(id, mi, set_title=False, - set_authors=set_authors, commit=False) - self.db.commit() - self.drag_drop_finished.emit(ids) - - def set_search_restriction(self, s): - self.search_restriction = s - - def _get_category_nodes(self, sort): - ''' - Called by __init__. Do not directly call this method. - ''' - self.row_map = [] - self.categories = {} - - # Get the categories - if self.search_restriction: - try: - data = self.db.get_categories(sort=sort, - icon_map=self.category_icon_map, - ids=self.db.search('', return_matches=True)) - except: - data = self.db.get_categories(sort=sort, icon_map=self.category_icon_map) - self.tags_view.restriction_error.emit() - else: - data = self.db.get_categories(sort=sort, icon_map=self.category_icon_map) - - # Reconstruct the user categories, putting them into metadata - self.db.field_metadata.remove_dynamic_categories() - tb_cats = self.db.field_metadata - for user_cat in sorted(self.db.prefs.get('user_categories', {}).keys(), - key=sort_key): - cat_name = '@' + user_cat # add the '@' to avoid name collision - while True: - try: - tb_cats.add_user_category(label=cat_name, name=user_cat) - dot = cat_name.rfind('.') - if dot < 0: - break - cat_name = cat_name[:dot] - except ValueError: - break - - for cat in sorted(self.db.prefs.get('grouped_search_terms', {}).keys(), - key=sort_key): - if (u'@' + cat) in data: - try: - tb_cats.add_user_category(label=u'@' + cat, name=cat) - except ValueError: - traceback.print_exc() - self.db.data.change_search_locations(self.db.field_metadata.get_search_terms()) - - if len(saved_searches().names()): - tb_cats.add_search_category(label='search', name=_('Searches')) - - if self.filter_categories_by: - for category in data.keys(): - data[category] = [t for t in data[category] - if lower(t.name).find(self.filter_categories_by) >= 0] - - tb_categories = self.db.field_metadata - for category in tb_categories: - if category in data: # The search category can come and go - self.row_map.append(category) - self.categories[category] = tb_categories[category]['name'] - return data - - def refresh(self, data=None): - ''' - Here to trap usages of refresh in the old architecture. Can eventually - be removed. - ''' - print 'TagsModel: refresh called!' - traceback.print_stack() - return False - - def _create_node_tree(self, data, state_map): - ''' - Called by __init__. Do not directly call this method. - ''' - sort_by = config['sort_tags_by'] - - if data is None: - print '_create_node_tree: no data!' - traceback.print_stack() - return - - collapse = gprefs['tags_browser_collapse_at'] - collapse_model = self.collapse_model - if collapse == 0: - collapse_model = 'disable' - elif collapse_model != 'disable': - if sort_by == 'name': - collapse_template = tweaks['categories_collapsed_name_template'] - elif sort_by == 'rating': - collapse_model = 'partition' - collapse_template = tweaks['categories_collapsed_rating_template'] - else: - collapse_model = 'partition' - collapse_template = tweaks['categories_collapsed_popularity_template'] - - def process_one_node(category, state_map): # {{{ - collapse_letter = None - category_index = self.createIndex(category.row(), 0, category) - category_node = category_index.internalPointer() - key = category_node.category_key - if key not in data: - return - cat_len = len(data[key]) - if cat_len <= 0: - return - - category_child_map = {} - fm = self.db.field_metadata[key] - clear_rating = True if key not in self.categories_with_ratings and \ - not fm['is_custom'] and \ - not fm['kind'] == 'user' \ - else False - in_uc = fm['kind'] == 'user' - tt = key if in_uc else None - - if collapse_model == 'first letter': - # Build a list of 'equal' first letters by looking for - # overlapping ranges. If a range overlaps another, then the - # letters are assumed to be equivalent. ICU collating is complex - # beyond belief. This mechanism lets us determine the logical - # first character from ICU's standpoint. - chardict = {} - for idx,tag in enumerate(data[key]): - if not tag.sort: - c = ' ' - else: - c = icu_upper(tag.sort[0]) - if c not in chardict: - chardict[c] = [idx, idx] - else: - chardict[c][1] = idx - - # sort the ranges to facilitate detecting overlap - ranges = sorted([(v[0], v[1], c) for c,v in chardict.items()]) - - # Create a list of 'first letters' to use for each item in - # the category. The list is generated using the ranges. Overlaps - # are filled with the character that first occurs. - cl_list = list(repeat(None, len(data[key]))) - for t in ranges: - start = t[0] - c = t[2] - if cl_list[start] is None: - nc = c - else: - nc = cl_list[start] - for i in range(start, t[1]+1): - cl_list[i] = nc - - for idx,tag in enumerate(data[key]): - if clear_rating: - tag.avg_rating = None - tag.state = state_map.get((tag.name, tag.category), 0) - - if collapse_model != 'disable' and cat_len > collapse: - if collapse_model == 'partition': - if (idx % collapse) == 0: - d = {'first': tag} - if cat_len > idx + collapse: - d['last'] = data[key][idx+collapse-1] - else: - d['last'] = data[key][cat_len-1] - name = eval_formatter.safe_format(collapse_template, - d, 'TAG_VIEW', None) - self.beginInsertRows(category_index, 999998, 999999) #len(data[key])-1) - sub_cat = TagTreeItem(parent=category, data = name, - tooltip = None, temporary=True, - category_icon = category_node.icon, - category_key=category_node.category_key, - icon_map=self.icon_state_map) - sub_cat.tag.is_searchable = False - self.endInsertRows() - else: # by 'first letter' - cl = cl_list[idx] - if cl != collapse_letter: - collapse_letter = cl - sub_cat = TagTreeItem(parent=category, - data = collapse_letter, - category_icon = category_node.icon, - tooltip = None, temporary=True, - category_key=category_node.category_key, - icon_map=self.icon_state_map) - node_parent = sub_cat - else: - node_parent = category - - # category display order is important here. The following works - # only of all the non-user categories are displayed before the - # user categories - components = [t.strip() for t in tag.original_name.split('.') - if t.strip()] - if len(components) == 0 or '.'.join(components) != tag.original_name: - components = [tag.original_name] - if (not tag.is_hierarchical) and (in_uc or - (fm['is_custom'] and fm['display'].get('is_names', False)) or - key in ['authors', 'publisher', 'news', 'formats', 'rating'] or - key not in self.db.prefs.get('categories_using_hierarchy', []) or - len(components) == 1): - self.beginInsertRows(category_index, 999998, 999999) - n = TagTreeItem(parent=node_parent, data=tag, tooltip=tt, - icon_map=self.icon_state_map) - if tag.id_set is not None: - n.id_set |= tag.id_set - category_child_map[tag.name, tag.category] = n - self.endInsertRows() - else: - for i,comp in enumerate(components): - if i == 0: - child_map = category_child_map - else: - child_map = dict([((t.tag.name, t.tag.category), t) - for t in node_parent.children - if t.type != TagTreeItem.CATEGORY]) - if (comp,tag.category) in child_map: - node_parent = child_map[(comp,tag.category)] - node_parent.tag.is_hierarchical = \ - '5state' if tag.category != 'search' else '3state' - else: - if i < len(components)-1: - t = copy.copy(tag) - t.original_name = '.'.join(components[:i+1]) - if key != 'search': - # This 'manufactured' intermediate node can - # be searched, but cannot be edited. - t.is_editable = False - else: - t.is_searchable = t.is_editable = False - else: - t = tag - if not in_uc: - t.original_name = t.name - t.is_hierarchical = \ - '5state' if t.category != 'search' else '3state' - t.name = comp - self.beginInsertRows(category_index, 999998, 999999) - node_parent = TagTreeItem(parent=node_parent, data=t, - tooltip=tt, icon_map=self.icon_state_map) - child_map[(comp,tag.category)] = node_parent - self.endInsertRows() - # This id_set must not be None - node_parent.id_set |= tag.id_set - return - # }}} - - for category in self.category_nodes: - process_one_node(category, state_map.get(category.py_name, {})) - - def get_state(self): - state_map = {} - expanded_categories = [] - for row, category in enumerate(self.category_nodes): - if self.tags_view.isExpanded(self.index(row, 0, QModelIndex())): - expanded_categories.append(category.py_name) - states = [c.tag.state for c in category.child_tags()] - names = [(c.tag.name, c.tag.category) for c in category.child_tags()] - state_map[category.py_name] = dict(izip(names, states)) - return expanded_categories, state_map - - def index_for_category(self, name): - for row, category in enumerate(self.category_nodes): - if category.py_name == name: - return self.index(row, 0, QModelIndex()) - - def columnCount(self, parent): - return 1 - - def data(self, index, role): - if not index.isValid(): - return NONE - item = index.internalPointer() - return item.data(role) - - def setData(self, index, value, role=Qt.EditRole): - if not index.isValid(): - return NONE - # set up to reposition at the same item. We can do this except if - # working with the last item and that item is deleted, in which case - # we position at the parent label - path = index.model().path_for_index(index) - val = unicode(value.toString()).strip() - if not val: - error_dialog(self.tags_view, _('Item is blank'), - _('An item cannot be set to nothing. Delete it instead.')).exec_() - return False - item = index.internalPointer() - if item.type == TagTreeItem.CATEGORY and item.category_key.startswith('@'): - if val.find('.') >= 0: - error_dialog(self.tags_view, _('Rename user category'), - _('You cannot use periods in the name when ' - 'renaming user categories'), show=True) - return False - - user_cats = self.db.prefs.get('user_categories', {}) - user_cat_keys_lower = [icu_lower(k) for k in user_cats] - ckey = item.category_key[1:] - ckey_lower = icu_lower(ckey) - dotpos = ckey.rfind('.') - if dotpos < 0: - nkey = val - else: - nkey = ckey[:dotpos+1] + val - nkey_lower = icu_lower(nkey) - for c in sorted(user_cats.keys(), key=sort_key): - if icu_lower(c).startswith(ckey_lower): - if len(c) == len(ckey): - if strcmp(ckey, nkey) != 0 and \ - nkey_lower in user_cat_keys_lower: - error_dialog(self.tags_view, _('Rename user category'), - _('The name %s is already used')%nkey, show=True) - return False - user_cats[nkey] = user_cats[ckey] - del user_cats[ckey] - elif c[len(ckey)] == '.': - rest = c[len(ckey):] - if strcmp(ckey, nkey) != 0 and \ - icu_lower(nkey + rest) in user_cat_keys_lower: - error_dialog(self.tags_view, _('Rename user category'), - _('The name %s is already used')%(nkey+rest), show=True) - return False - user_cats[nkey + rest] = user_cats[ckey + rest] - del user_cats[ckey + rest] - self.db.prefs.set('user_categories', user_cats) - self.tags_view.set_new_model() - # must not use 'self' below because the model has changed! - p = self.tags_view.model().find_category_node('@' + nkey) - self.tags_view.model().show_item_at_path(p) - return True - - key = item.tag.category - name = item.tag.original_name - # make certain we know about the item's category - if key not in self.db.field_metadata: - return False - if key == 'authors': - if val.find('&') >= 0: - error_dialog(self.tags_view, _('Invalid author name'), - _('Author names cannot contain & characters.')).exec_() - return False - if key == 'search': - if val in saved_searches().names(): - error_dialog(self.tags_view, _('Duplicate search name'), - _('The saved search name %s is already used.')%val).exec_() - return False - saved_searches().rename(unicode(item.data(role).toString()), val) - item.tag.name = val - self.tags_view.search_item_renamed.emit() # Does a refresh - else: - if key == 'series': - self.db.rename_series(item.tag.id, val) - elif key == 'publisher': - self.db.rename_publisher(item.tag.id, val) - elif key == 'tags': - self.db.rename_tag(item.tag.id, val) - elif key == 'authors': - self.db.rename_author(item.tag.id, val) - elif self.db.field_metadata[key]['is_custom']: - self.db.rename_custom_item(item.tag.id, val, - label=self.db.field_metadata[key]['label']) - self.tags_view.tag_item_renamed.emit() - item.tag.name = val - self.rename_item_in_all_user_categories(name, key, val) - self.refresh_required.emit() - self.show_item_at_path(path) - return True - - def rename_item_in_all_user_categories(self, item_name, item_category, new_name): - ''' - Search all user categories for items named item_name with category - item_category and rename them to new_name. The caller must arrange to - redisplay the tree as appropriate (recount or set_new_model) - ''' - user_cats = self.db.prefs.get('user_categories', {}) - for k in user_cats.keys(): - new_contents = [] - for tup in user_cats[k]: - if tup[0] == item_name and tup[1] == item_category: - new_contents.append([new_name, item_category, 0]) - else: - new_contents.append(tup) - user_cats[k] = new_contents - self.db.prefs.set('user_categories', user_cats) - - def delete_item_from_all_user_categories(self, item_name, item_category): - ''' - Search all user categories for items named item_name with category - item_category and delete them. The caller must arrange to redisplay the - tree as appropriate (recount or set_new_model) - ''' - user_cats = self.db.prefs.get('user_categories', {}) - for cat in user_cats.keys(): - self.delete_item_from_user_category(cat, item_name, item_category, - user_categories=user_cats) - self.db.prefs.set('user_categories', user_cats) - - def delete_item_from_user_category(self, category, item_name, item_category, - user_categories=None): - if user_categories is not None: - user_cats = user_categories - else: - user_cats = self.db.prefs.get('user_categories', {}) - new_contents = [] - for tup in user_cats[category]: - if tup[0] != item_name or tup[1] != item_category: - new_contents.append(tup) - user_cats[category] = new_contents - if user_categories is None: - self.db.prefs.set('user_categories', user_cats) - - def headerData(self, *args): - return NONE - - def flags(self, index, *args): - ans = Qt.ItemIsEnabled|Qt.ItemIsSelectable|Qt.ItemIsEditable - if index.isValid(): - node = self.data(index, Qt.UserRole) - if node.type == TagTreeItem.TAG: - if node.tag.is_editable: - ans |= Qt.ItemIsDragEnabled - fm = self.db.metadata_for_field(node.tag.category) - if node.tag.category in \ - ('tags', 'series', 'authors', 'rating', 'publisher') or \ - (fm['is_custom'] and \ - fm['datatype'] in ['text', 'rating', 'series', 'enumeration']): - ans |= Qt.ItemIsDropEnabled - else: - ans |= Qt.ItemIsDropEnabled - return ans - - def supportedDropActions(self): - return Qt.CopyAction|Qt.MoveAction - - def path_for_index(self, index): - ans = [] - while index.isValid(): - ans.append(index.row()) - index = self.parent(index) - ans.reverse() - return ans - - def index_for_path(self, path): - parent = QModelIndex() - for i in path: - parent = self.index(i, 0, parent) - if not parent.isValid(): - return QModelIndex() - return parent - - def index(self, row, column, parent): - if not self.hasIndex(row, column, parent): - return QModelIndex() - - if not parent.isValid(): - parent_item = self.root_item - else: - parent_item = parent.internalPointer() - - try: - child_item = parent_item.children[row] - except IndexError: - return QModelIndex() - - ans = self.createIndex(row, column, child_item) - return ans - - def parent(self, index): - if not index.isValid(): - return QModelIndex() - - child_item = index.internalPointer() - parent_item = getattr(child_item, 'parent', None) - - if parent_item is self.root_item or parent_item is None: - return QModelIndex() - - ans = self.createIndex(parent_item.row(), 0, parent_item) - return ans - - def rowCount(self, parent): - if parent.column() > 0: - return 0 - - if not parent.isValid(): - parent_item = self.root_item - else: - parent_item = parent.internalPointer() - - return len(parent_item.children) - - def reset_all_states(self, except_=None): - update_list = [] - def process_tag(tag_item): - tag = tag_item.tag - if tag is except_: - tag_index = self.createIndex(tag_item.row(), 0, tag_item) - self.dataChanged.emit(tag_index, tag_index) - elif tag.state != 0 or tag in update_list: - tag_index = self.createIndex(tag_item.row(), 0, tag_item) - tag.state = 0 - update_list.append(tag) - self.dataChanged.emit(tag_index, tag_index) - for t in tag_item.children: - process_tag(t) - - for t in self.root_item.children: - process_tag(t) - - def clear_state(self): - self.reset_all_states() - - def toggle(self, index, exclusive, set_to=None): - ''' - exclusive: clear all states before applying this one - set_to: None => advance the state, otherwise a value from TAG_SEARCH_STATES - ''' - if not index.isValid(): return False - item = index.internalPointer() - item.toggle(set_to=set_to) - if exclusive: - self.reset_all_states(except_=item.tag) - self.dataChanged.emit(index, index) - return True - - def tokens(self): - ans = [] - # Tags can be in the news and the tags categories. However, because of - # the desire to use two different icons (tags and news), the nodes are - # not shared, which can lead to the possibility of searching twice for - # the same tag. The tags_seen set helps us prevent that - tags_seen = set() - # Tag nodes are in their own category and possibly in user categories. - # They will be 'checked' in both places, but we want to put the node - # into the search string only once. The nodes_seen set helps us do that - nodes_seen = set() - - node_searches = {TAG_SEARCH_STATES['mark_plus'] : 'true', - TAG_SEARCH_STATES['mark_plusplus'] : '.true', - TAG_SEARCH_STATES['mark_minus'] : 'false', - TAG_SEARCH_STATES['mark_minusminus'] : '.false'} - - for node in self.category_nodes: - if node.tag.state: - if node.category_key == "news": - if node_searches[node.tag.state] == 'true': - ans.append('tags:"=' + _('News') + '"') - else: - ans.append('( not tags:"=' + _('News') + '")') - else: - ans.append('%s:%s'%(node.category_key, node_searches[node.tag.state])) - - key = node.category_key - for tag_item in node.all_children(): - if tag_item.type == TagTreeItem.CATEGORY: - if self.collapse_model == 'first letter' and \ - tag_item.temporary and not key.startswith('@') \ - and tag_item.tag.state: - if node_searches[tag_item.tag.state] == 'true': - ans.append('%s:~^%s'%(key, tag_item.py_name)) - else: - ans.append('(not %s:~^%s )'%(key, tag_item.py_name)) - continue - tag = tag_item.tag - if tag.state != TAG_SEARCH_STATES['clear']: - if tag.state == TAG_SEARCH_STATES['mark_minus'] or \ - tag.state == TAG_SEARCH_STATES['mark_minusminus']: - prefix = ' not ' - else: - prefix = '' - category = tag.category if key != 'news' else 'tag' - add_colon = False - if self.db.field_metadata[tag.category]['is_csp']: - add_colon = True - - if tag.name and tag.name[0] == u'\u2605': # char is a star. Assume rating - ans.append('%s%s:%s'%(prefix, category, len(tag.name))) - else: - name = tag.original_name - use_prefix = tag.state in [TAG_SEARCH_STATES['mark_plusplus'], - TAG_SEARCH_STATES['mark_minusminus']] - if category == 'tags': - if name in tags_seen: - continue - tags_seen.add(name) - if tag in nodes_seen: - continue - nodes_seen.add(tag) - n = name.replace(r'"', r'\"') - if name.startswith('.'): - n = '.' + n - ans.append('%s%s:"=%s%s%s"'%(prefix, category, - '.' if use_prefix else '', n, - ':' if add_colon else '')) - return ans - - def find_item_node(self, key, txt, start_path, equals_match=False): - ''' - Search for an item (a node) in the tags browser list that matches both - the key (exact case-insensitive match) and txt (not equals_match => - case-insensitive contains match; equals_match => case_insensitive - equal match). Returns the path to the node. Note that paths are to a - location (second item, fourth item, 25 item), not to a node. If - start_path is None, the search starts with the topmost node. If the tree - is changed subsequent to calling this method, the path can easily refer - to a different node or no node at all. - ''' - if not txt: - return None - txt = lower(txt) if not equals_match else txt - self.path_found = None - if start_path is None: - start_path = [] - - def process_tag(depth, tag_index, tag_item, start_path): - path = self.path_for_index(tag_index) - if depth < len(start_path) and path[depth] <= start_path[depth]: - return False - tag = tag_item.tag - if tag is None: - return False - name = tag.original_name - if (equals_match and strcmp(name, txt) == 0) or \ - (not equals_match and lower(name).find(txt) >= 0): - self.path_found = path - return True - for i,c in enumerate(tag_item.children): - if process_tag(depth+1, self.createIndex(i, 0, c), c, start_path): - return True - return False - - def process_level(depth, category_index, start_path): - path = self.path_for_index(category_index) - if depth < len(start_path): - if path[depth] < start_path[depth]: - return False - if path[depth] > start_path[depth]: - start_path = path - my_key = category_index.internalPointer().category_key - for j in xrange(self.rowCount(category_index)): - tag_index = self.index(j, 0, category_index) - tag_item = tag_index.internalPointer() - if tag_item.type == TagTreeItem.CATEGORY: - if process_level(depth+1, tag_index, start_path): - return True - elif not key or strcmp(key, my_key) == 0: - if process_tag(depth+1, tag_index, tag_item, start_path): - return True - return False - - for i in xrange(self.rowCount(QModelIndex())): - if process_level(0, self.index(i, 0, QModelIndex()), start_path): - break - return self.path_found - - def find_category_node(self, key, parent=QModelIndex()): - ''' - Search for an category node (a top-level node) in the tags browser list - that matches the key (exact case-insensitive match). Returns the path to - the node. Paths are as in find_item_node. - ''' - if not key: - return None - - for i in xrange(self.rowCount(parent)): - idx = self.index(i, 0, parent) - node = idx.internalPointer() - if node.type == TagTreeItem.CATEGORY: - ckey = node.category_key - if strcmp(ckey, key) == 0: - return self.path_for_index(idx) - if len(node.children): - v = self.find_category_node(key, idx) - if v is not None: - return v - return None - - def show_item_at_path(self, path, box=False, - position=QTreeView.PositionAtCenter): - ''' - Scroll the browser and open categories to show the item referenced by - path. If possible, the item is placed in the center. If box=True, a - box is drawn around the item. - ''' - if path: - self.show_item_at_index(self.index_for_path(path), box=box, - position=position) - - def show_item_at_index(self, idx, box=False, - position=QTreeView.PositionAtCenter): - if idx.isValid(): - self.tags_view.setCurrentIndex(idx) - self.tags_view.scrollTo(idx, position) - if box: - tag_item = idx.internalPointer() - tag_item.boxed = True - self.dataChanged.emit(idx, idx) - - def clear_boxed(self): - ''' - Clear all boxes around items. - ''' - def process_tag(tag_index, tag_item): - if tag_item.boxed: - tag_item.boxed = False - self.dataChanged.emit(tag_index, tag_index) - for i,c in enumerate(tag_item.children): - process_tag(self.index(i, 0, tag_index), c) - - def process_level(category_index): - for j in xrange(self.rowCount(category_index)): - tag_index = self.index(j, 0, category_index) - tag_item = tag_index.internalPointer() - if tag_item.boxed: - tag_item.boxed = False - self.dataChanged.emit(tag_index, tag_index) - if tag_item.type == TagTreeItem.CATEGORY: - process_level(tag_index) - else: - process_tag(tag_index, tag_item) - - for i in xrange(self.rowCount(QModelIndex())): - process_level(self.index(i, 0, QModelIndex())) - - def get_filter_categories_by(self): - return self.filter_categories_by - - # }}} - -class TagBrowserMixin(object): # {{{ - - def __init__(self, db): - self.library_view.model().count_changed_signal.connect(self.tags_view.recount) - self.tags_view.set_database(db, self.tag_match, self.sort_by) - self.tags_view.tags_marked.connect(self.search.set_search_string) - self.tags_view.tag_list_edit.connect(self.do_tags_list_edit) - self.tags_view.edit_user_category.connect(self.do_edit_user_categories) - self.tags_view.delete_user_category.connect(self.do_delete_user_category) - self.tags_view.del_item_from_user_cat.connect(self.do_del_item_from_user_cat) - self.tags_view.add_subcategory.connect(self.do_add_subcategory) - self.tags_view.add_item_to_user_cat.connect(self.do_add_item_to_user_cat) - self.tags_view.saved_search_edit.connect(self.do_saved_search_edit) - self.tags_view.rebuild_saved_searches.connect(self.do_rebuild_saved_searches) - self.tags_view.author_sort_edit.connect(self.do_author_sort_edit) - self.tags_view.tag_item_renamed.connect(self.do_tag_item_renamed) - self.tags_view.search_item_renamed.connect(self.saved_searches_changed) - self.tags_view.drag_drop_finished.connect(self.drag_drop_finished) - self.tags_view.restriction_error.connect(self.do_restriction_error, - type=Qt.QueuedConnection) - - for text, func, args, cat_name in ( - (_('Manage Authors'), - self.do_author_sort_edit, (self, None), 'authors'), - (_('Manage Series'), - self.do_tags_list_edit, (None, 'series'), 'series'), - (_('Manage Publishers'), - self.do_tags_list_edit, (None, 'publisher'), 'publisher'), - (_('Manage Tags'), - self.do_tags_list_edit, (None, 'tags'), 'tags'), - (_('Manage User Categories'), - self.do_edit_user_categories, (None,), 'user:'), - (_('Manage Saved Searches'), - self.do_saved_search_edit, (None,), 'search') - ): - self.manage_items_button.menu().addAction( - QIcon(I(category_icon_map[cat_name])), - text, partial(func, *args)) - - def do_restriction_error(self): - error_dialog(self.tags_view, _('Invalid search restriction'), - _('The current search restriction is invalid'), show=True) - - def do_add_subcategory(self, on_category_key, new_category_name=None): - ''' - Add a subcategory to the category 'on_category'. If new_category_name is - None, then a default name is shown and the user is offered the - opportunity to edit the name. - ''' - db = self.library_view.model().db - user_cats = db.prefs.get('user_categories', {}) - - # Ensure that the temporary name we will use is not already there - i = 0 - if new_category_name is not None: - new_name = new_category_name.replace('.', '') - else: - new_name = _('New Category').replace('.', '') - n = new_name - while True: - new_cat = on_category_key[1:] + '.' + n - if new_cat not in user_cats: - break - i += 1 - n = new_name + unicode(i) - # Add the new category - user_cats[new_cat] = [] - db.prefs.set('user_categories', user_cats) - self.tags_view.set_new_model() - m = self.tags_view.model() - idx = m.index_for_path(m.find_category_node('@' + new_cat)) - m.show_item_at_index(idx) - # Open the editor on the new item to rename it - if new_category_name is None: - self.tags_view.edit(idx) - - def do_edit_user_categories(self, on_category=None): - ''' - Open the user categories editor. - ''' - db = self.library_view.model().db - d = TagCategories(self, db, on_category) - if d.exec_() == d.Accepted: - db.prefs.set('user_categories', d.categories) - db.field_metadata.remove_user_categories() - for k in d.categories: - db.field_metadata.add_user_category('@' + k, k) - db.data.change_search_locations(db.field_metadata.get_search_terms()) - self.tags_view.set_new_model() - - def do_delete_user_category(self, category_name): - ''' - Delete the user category named category_name. Any leading '@' is removed - ''' - if category_name.startswith('@'): - category_name = category_name[1:] - db = self.library_view.model().db - user_cats = db.prefs.get('user_categories', {}) - cat_keys = sorted(user_cats.keys(), key=sort_key) - has_children = False - found = False - for k in cat_keys: - if k == category_name: - found = True - has_children = len(user_cats[k]) - elif k.startswith(category_name + '.'): - has_children = True - if not found: - return error_dialog(self.tags_view, _('Delete user category'), - _('%s is not a user category')%category_name, show=True) - if has_children: - if not question_dialog(self.tags_view, _('Delete user category'), - _('%s contains items. Do you really ' - 'want to delete it?')%category_name): - return - for k in cat_keys: - if k == category_name: - del user_cats[k] - elif k.startswith(category_name + '.'): - del user_cats[k] - db.prefs.set('user_categories', user_cats) - self.tags_view.set_new_model() - - def do_del_item_from_user_cat(self, user_cat, item_name, item_category): - ''' - Delete the item (item_name, item_category) from the user category with - key user_cat. Any leading '@' characters are removed - ''' - if user_cat.startswith('@'): - user_cat = user_cat[1:] - db = self.library_view.model().db - user_cats = db.prefs.get('user_categories', {}) - if user_cat not in user_cats: - error_dialog(self.tags_view, _('Remove category'), - _('User category %s does not exist')%user_cat, - show=True) - return - self.tags_view.model().delete_item_from_user_category(user_cat, - item_name, item_category) - self.tags_view.recount() - - def do_add_item_to_user_cat(self, dest_category, src_name, src_category): - ''' - Add the item src_name in src_category to the user category - dest_category. Any leading '@' is removed - ''' - db = self.library_view.model().db - user_cats = db.prefs.get('user_categories', {}) - - if dest_category and dest_category.startswith('@'): - dest_category = dest_category[1:] - - if dest_category not in user_cats: - return error_dialog(self.tags_view, _('Add to user category'), - _('A user category %s does not exist')%dest_category, show=True) - - # Now add the item to the destination user category - add_it = True - if src_category == 'news': - src_category = 'tags' - for tup in user_cats[dest_category]: - if src_name == tup[0] and src_category == tup[1]: - add_it = False - if add_it: - user_cats[dest_category].append([src_name, src_category, 0]) - db.prefs.set('user_categories', user_cats) - self.tags_view.recount() - - def do_tags_list_edit(self, tag, category): - ''' - Open the 'manage_X' dialog where X == category. If tag is not None, the - dialog will position the editor on that item. - ''' - db=self.library_view.model().db - if category == 'tags': - result = db.get_tags_with_ids() - key = sort_key - elif category == 'series': - result = db.get_series_with_ids() - key = lambda x:sort_key(title_sort(x)) - elif category == 'publisher': - result = db.get_publishers_with_ids() - key = sort_key - else: # should be a custom field - cc_label = None - if category in db.field_metadata: - cc_label = db.field_metadata[category]['label'] - result = db.get_custom_items_with_ids(label=cc_label) - else: - result = [] - key = sort_key - - d = TagListEditor(self, tag_to_match=tag, data=result, key=key) - d.exec_() - if d.result() == d.Accepted: - to_rename = d.to_rename # dict of new text to old id - to_delete = d.to_delete # list of ids - orig_name = d.original_names # dict of id: name - - rename_func = None - if category == 'tags': - rename_func = db.rename_tag - delete_func = db.delete_tag_using_id - elif category == 'series': - rename_func = db.rename_series - delete_func = db.delete_series_using_id - elif category == 'publisher': - rename_func = db.rename_publisher - delete_func = db.delete_publisher_using_id - else: - rename_func = partial(db.rename_custom_item, label=cc_label) - delete_func = partial(db.delete_custom_item_using_id, label=cc_label) - m = self.tags_view.model() - if rename_func: - for item in to_delete: - delete_func(item) - m.delete_item_from_all_user_categories(orig_name[item], category) - for old_id in to_rename: - rename_func(old_id, new_name=unicode(to_rename[old_id])) - m.rename_item_in_all_user_categories(orig_name[old_id], - category, unicode(to_rename[old_id])) - - # Clean up the library view - self.do_tag_item_renamed() - self.tags_view.recount() - - def do_tag_item_renamed(self): - # Clean up library view and search - # get information to redo the selection - rows = [r.row() for r in \ - self.library_view.selectionModel().selectedRows()] - m = self.library_view.model() - ids = [m.id(r) for r in rows] - - m.refresh(reset=False) - m.research() - self.library_view.select_rows(ids) - # refreshing the tags view happens at the emit()/call() site - - def do_author_sort_edit(self, parent, id, select_sort=True): - ''' - Open the manage authors dialog - ''' - db = self.library_view.model().db - editor = EditAuthorsDialog(parent, db, id, select_sort) - d = editor.exec_() - if d: - for (id, old_author, new_author, new_sort) in editor.result: - if old_author != new_author: - # The id might change if the new author already exists - id = db.rename_author(id, new_author) - db.set_sort_field_for_author(id, unicode(new_sort), - commit=False, notify=False) - db.commit() - self.library_view.model().refresh() - self.tags_view.recount() - - def drag_drop_finished(self, ids): - self.library_view.model().refresh_ids(ids) - -# }}} - -class TagBrowserWidget(QWidget): # {{{ - - def __init__(self, parent): - QWidget.__init__(self, parent) - self.parent = parent - self._layout = QVBoxLayout() - self.setLayout(self._layout) - self._layout.setContentsMargins(0,0,0,0) - - # Set up the find box & button - search_layout = QHBoxLayout() - self._layout.addLayout(search_layout) - self.item_search = HistoryLineEdit(parent) - try: - self.item_search.lineEdit().setPlaceholderText( - _('Find item in tag browser')) - except: - pass # Using Qt < 4.7 - self.item_search.setToolTip(_( - 'Search for items. This is a "contains" search; items containing the\n' - 'text anywhere in the name will be found. You can limit the search\n' - 'to particular categories using syntax similar to search. For example,\n' - 'tags:foo will find foo in any tag, but not in authors etc. Entering\n' - '*foo will filter all categories at once, showing only those items\n' - 'containing the text "foo"')) - search_layout.addWidget(self.item_search) - # Not sure if the shortcut should be translatable ... - sc = QShortcut(QKeySequence(_('ALT+f')), parent) - sc.connect(sc, SIGNAL('activated()'), self.set_focus_to_find_box) - - self.search_button = QToolButton() - self.search_button.setText(_('F&ind')) - self.search_button.setToolTip(_('Find the first/next matching item')) - search_layout.addWidget(self.search_button) - - self.expand_button = QToolButton() - self.expand_button.setText('-') - self.expand_button.setToolTip(_('Collapse all categories')) - search_layout.addWidget(self.expand_button) - search_layout.setStretch(0, 10) - search_layout.setStretch(1, 1) - search_layout.setStretch(2, 1) - - self.current_find_position = None - self.search_button.clicked.connect(self.find) - self.item_search.initialize('tag_browser_search') - self.item_search.lineEdit().returnPressed.connect(self.do_find) - self.item_search.lineEdit().textEdited.connect(self.find_text_changed) - self.item_search.activated[QString].connect(self.do_find) - self.item_search.completer().setCaseSensitivity(Qt.CaseSensitive) - - parent.tags_view = TagsView(parent) - self.tags_view = parent.tags_view - self.expand_button.clicked.connect(self.tags_view.collapseAll) - self._layout.addWidget(parent.tags_view) - - # Now the floating 'not found' box - l = QLabel(self.tags_view) - self.not_found_label = l - l.setFrameStyle(QFrame.StyledPanel) - l.setAutoFillBackground(True) - l.setText('

'+_('No More Matches.

Click Find again to go to first match')) - l.setAlignment(Qt.AlignVCenter) - l.setWordWrap(True) - l.resize(l.sizeHint()) - l.move(10,20) - l.setVisible(False) - self.not_found_label_timer = QTimer() - self.not_found_label_timer.setSingleShot(True) - self.not_found_label_timer.timeout.connect(self.not_found_label_timer_event, - type=Qt.QueuedConnection) - - parent.sort_by = QComboBox(parent) - # Must be in the same order as db2.CATEGORY_SORTS - for x in (_('Sort by name'), _('Sort by popularity'), - _('Sort by average rating')): - parent.sort_by.addItem(x) - parent.sort_by.setToolTip( - _('Set the sort order for entries in the Tag Browser')) - parent.sort_by.setStatusTip(parent.sort_by.toolTip()) - parent.sort_by.setCurrentIndex(0) - self._layout.addWidget(parent.sort_by) - - # Must be in the same order as db2.MATCH_TYPE - parent.tag_match = QComboBox(parent) - for x in (_('Match any'), _('Match all')): - parent.tag_match.addItem(x) - parent.tag_match.setCurrentIndex(0) - self._layout.addWidget(parent.tag_match) - parent.tag_match.setToolTip( - _('When selecting multiple entries in the Tag Browser ' - 'match any or all of them')) - parent.tag_match.setStatusTip(parent.tag_match.toolTip()) - - - l = parent.manage_items_button = QPushButton(self) - l.setStyleSheet('QPushButton {text-align: left; }') - l.setText(_('Manage authors, tags, etc')) - l.setToolTip(_('All of these category_managers are available by right-clicking ' - 'on items in the tag browser above')) - l.m = QMenu() - l.setMenu(l.m) - self._layout.addWidget(l) - - # self.leak_test_timer = QTimer(self) - # self.leak_test_timer.timeout.connect(self.test_for_leak) - # self.leak_test_timer.start(5000) - - def set_pane_is_visible(self, to_what): - self.tags_view.set_pane_is_visible(to_what) - - def find_text_changed(self, str): - self.current_find_position = None - - def set_focus_to_find_box(self): - self.item_search.setFocus() - self.item_search.lineEdit().selectAll() - - def do_find(self, str=None): - self.current_find_position = None - self.find() - - def find(self): - model = self.tags_view.model() - model.clear_boxed() - txt = unicode(self.item_search.currentText()).strip() - - if txt.startswith('*'): - self.tags_view.set_new_model(filter_categories_by=txt[1:]) - self.current_find_position = None - return - if model.get_filter_categories_by(): - self.tags_view.set_new_model(filter_categories_by=None) - self.current_find_position = None - model = self.tags_view.model() - - if not txt: - return - - self.item_search.lineEdit().blockSignals(True) - self.search_button.setFocus(True) - self.item_search.lineEdit().blockSignals(False) - - key = None - colon = txt.rfind(':') if len(txt) > 2 else 0 - if colon > 0: - key = self.parent.library_view.model().db.\ - field_metadata.search_term_to_field_key(txt[:colon]) - txt = txt[colon+1:] - - self.current_find_position = \ - model.find_item_node(key, txt, self.current_find_position) - if self.current_find_position: - model.show_item_at_path(self.current_find_position, box=True) - elif self.item_search.text(): - self.not_found_label.setVisible(True) - if self.tags_view.verticalScrollBar().isVisible(): - sbw = self.tags_view.verticalScrollBar().width() - else: - sbw = 0 - width = self.width() - 8 - sbw - height = self.not_found_label.heightForWidth(width) + 20 - self.not_found_label.resize(width, height) - self.not_found_label.move(4, 10) - self.not_found_label_timer.start(2000) - - def not_found_label_timer_event(self): - self.not_found_label.setVisible(False) - - def test_for_leak(self): - from calibre.utils.mem import memory - import gc - before = memory() - self.tags_view.recount() - for i in xrange(3): gc.collect() - print 'Used memory:', memory(before)/(1024.), 'KB' - -# }}} - diff --git a/src/calibre/gui2/ui.py b/src/calibre/gui2/ui.py index cf9f6ee610..5007d343ce 100644 --- a/src/calibre/gui2/ui.py +++ b/src/calibre/gui2/ui.py @@ -39,7 +39,7 @@ from calibre.gui2.jobs import JobManager, JobsDialog, JobsButton from calibre.gui2.init import LibraryViewMixin, LayoutMixin from calibre.gui2.search_box import SearchBoxMixin, SavedSearchBoxMixin from calibre.gui2.search_restriction_mixin import SearchRestrictionMixin -from calibre.gui2.tag_view import TagBrowserMixin +from calibre.gui2.tag_browser.ui import TagBrowserMixin class Listener(Thread): # {{{ diff --git a/src/calibre/gui2/update.py b/src/calibre/gui2/update.py index c9a908bdf6..9d4d4def75 100644 --- a/src/calibre/gui2/update.py +++ b/src/calibre/gui2/update.py @@ -72,9 +72,7 @@ class UpdateNotification(QDialog): self.label = QLabel(('

'+ _('%s has been updated to version %s. ' 'See the new features.') + '

'+_('Update only if one of the ' - 'new features or bug fixes is important to you. ' - 'If the current version works well for you, do not update.'))%( + '">new features.'))%( __appname__, calibre_version)) self.label.setOpenExternalLinks(True) self.label.setWordWrap(True) diff --git a/src/calibre/library/database2.py b/src/calibre/library/database2.py index a3211b3817..8b4ad47284 100644 --- a/src/calibre/library/database2.py +++ b/src/calibre/library/database2.py @@ -2012,7 +2012,7 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns): val = mi.get(key, None) if force_changes or val is not None: doit(self.set_custom, id, val=val, extra=mi.get_extra(key), - label=user_mi[key]['label'], commit=False) + label=user_mi[key]['label'], commit=False, notify=False) if commit: self.conn.commit() if notify: diff --git a/src/calibre/utils/formatter.py b/src/calibre/utils/formatter.py index ebf47db854..3a93c2b650 100644 --- a/src/calibre/utils/formatter.py +++ b/src/calibre/utils/formatter.py @@ -347,5 +347,6 @@ class EvalFormatter(TemplateFormatter): key = key.lower() return kwargs.get(key, _('No such variable ') + key) +# DEPRECATED. This is not thread safe. Do not use. eval_formatter = EvalFormatter() diff --git a/src/calibre/utils/formatter_functions.py b/src/calibre/utils/formatter_functions.py index 63264e6379..c6f4bd1b0e 100644 --- a/src/calibre/utils/formatter_functions.py +++ b/src/calibre/utils/formatter_functions.py @@ -202,9 +202,9 @@ class BuiltinEval(BuiltinFormatterFunction): 'results from local variables.') def evaluate(self, formatter, kwargs, mi, locals, template): - from formatter import eval_formatter + from formatter import EvalFormatter template = template.replace('[[', '{').replace(']]', '}') - return eval_formatter.safe_format(template, locals, 'EVAL', None) + return EvalFormatter().safe_format(template, locals, 'EVAL', None) class BuiltinAssign(BuiltinFormatterFunction): name = 'assign' @@ -785,7 +785,7 @@ class BuiltinDaysBetween(BuiltinFormatterFunction): except: return '' i = d1 - d2 - return str(i.days) + return str('%d.%d'%(i.days, i.seconds/8640)) builtin_add = BuiltinAdd() diff --git a/src/calibre/utils/ipc/proxy.py b/src/calibre/utils/ipc/proxy.py index b60f9eb755..3b037a42a6 100644 --- a/src/calibre/utils/ipc/proxy.py +++ b/src/calibre/utils/ipc/proxy.py @@ -7,15 +7,82 @@ __license__ = 'GPL v3' __copyright__ = '2011, Kovid Goyal ' __docformat__ = 'restructuredtext en' -import os +import os, cPickle, struct from threading import Thread +from Queue import Queue, Empty from multiprocessing.connection import arbitrary_address, Listener +from functools import partial -from calibre.constants import iswindows +from calibre import as_unicode, prints +from calibre.constants import iswindows, DEBUG + +def _encode(msg): + raw = cPickle.dumps(msg, -1) + size = len(raw) + header = struct.pack('!Q', size) + return header + raw + +def _decode(raw): + sz = struct.calcsize('!Q') + if len(raw) < sz: + return 'invalid', None + header, = struct.unpack('!Q', raw[:sz]) + if len(raw) != sz + header or header == 0: + return 'invalid', None + return cPickle.loads(raw[sz:]) + + +class Writer(Thread): + + TIMEOUT = 60 #seconds + + def __init__(self, conn): + Thread.__init__(self) + self.daemon = True + self.dataq, self.resultq = Queue(), Queue() + self.conn = conn + self.start() + self.data_written = False + + def close(self): + self.dataq.put(None) + + def flush(self): + pass + + def write(self, raw_data): + self.dataq.put(raw_data) + + try: + ex = self.resultq.get(True, self.TIMEOUT) + except Empty: + raise IOError('Writing to socket timed out') + else: + if ex is not None: + raise IOError('Writing to socket failed with error: %s' % ex) + + def run(self): + while True: + x = self.dataq.get() + if x is None: + break + try: + self.data_written = True + self.conn.send_bytes(x) + except Exception as e: + self.resultq.put(as_unicode(e)) + else: + self.resultq.put(None) + + def __enter__(self): + return self + + def __exit__(self, exc_type, exc_val, exc_tb): + self.close() class Server(Thread): - def __init__(self): + def __init__(self, dispatcher): Thread.__init__(self) self.daemon = True @@ -27,6 +94,13 @@ class Server(Thread): authkey=self.auth_key, backlog=4) self.keep_going = True + self.dispatcher = dispatcher + + @property + def connection_information(self): + if not self.is_alive(): + self.start() + return (self.address, self.auth_key) def stop(self): self.keep_going = False @@ -43,4 +117,49 @@ class Server(Thread): except: pass + def handle_client(self, conn): + t = Thread(target=partial(self._handle_client, conn)) + t.daemon = True + t.start() + + def _handle_client(self, conn): + while True: + try: + func_name, args, kwargs = conn.recv() + except EOFError: + try: + conn.close() + except: + pass + return + else: + try: + self.call_func(func_name, args, kwargs, conn) + except: + try: + conn.close() + except: + pass + prints('Proxy function: %s with args: %r and' + ' kwargs: %r failed') + if DEBUG: + import traceback + traceback.print_exc() + break + + def call_func(self, func_name, args, kwargs, conn): + with Writer(conn) as f: + try: + self.dispatcher(f, func_name, args, kwargs) + except Exception as e: + if not f.data_written: + import traceback + # Try to tell the client process what error happened + try: + conn.send_bytes(_encode(('failed', (unicode(e), + as_unicode(traceback.format_exc()))))) + except: + pass + raise + diff --git a/src/calibre/utils/opensearch/__init__.py b/src/calibre/utils/opensearch/__init__.py new file mode 100644 index 0000000000..e69de29bb2 diff --git a/src/calibre/utils/opensearch/description.py b/src/calibre/utils/opensearch/description.py new file mode 100644 index 0000000000..0b5afd8a7e --- /dev/null +++ b/src/calibre/utils/opensearch/description.py @@ -0,0 +1,114 @@ +# -*- coding: utf-8 -*- + +from __future__ import (unicode_literals, division, absolute_import, print_function) + +__license__ = 'GPL 3' +__copyright__ = ''' +2011, John Schember , +2006, Ed Summers +''' +__docformat__ = 'restructuredtext en' + +from contextlib import closing + +from lxml import etree + +from calibre import browser +from calibre.utils.opensearch.url import URL + +class Description(object): + ''' + A class for representing OpenSearch Description files. + ''' + + def __init__(self, url=""): + ''' + The constructor which may pass an optional url to load from. + + d = Description("http://www.example.com/description") + ''' + if url: + self.load(url) + + + def load(self, url): + ''' + For loading up a description object from a url. Normally + you'll probably just want to pass a URL into the constructor. + ''' + br = browser() + with closing(br.open(url, timeout=15)) as f: + doc = etree.fromstring(f.read()) + + # version 1.1 has repeating Url elements + self.urls = [] + for element in doc.xpath('//*[local-name() = "Url"]'): + template = element.get('template') + type = element.get('type') + if template and type: + url = URL() + url.template = template + url.type = type + self.urls.append(url) + + # this is version 1.0 specific + self.url = ''.join(doc.xpath('//*[local-name() = "Url"][1]//text()')) + self.format = ''.join(doc.xpath('//*[local-name() = "Format"][1]//text()')) + + self.shortname = ''.join(doc.xpath('//*[local-name() = "ShortName"][1]//text()')) + self.longname = ''.join(doc.xpath('//*[local-name() = "LongName"][1]//text()')) + self.description = ''.join(doc.xpath('//*[local-name() = "Description"][1]//text()')) + self.image = ''.join(doc.xpath('//*[local-name() = "Image"][1]//text()')) + self.sameplesearch = ''.join(doc.xpath('//*[local-name() = "SampleSearch"][1]//text()')) + self.developer = ''.join(doc.xpath('//*[local-name() = "Developer"][1]//text()')) + self.contact = ''.join(doc.xpath('/*[local-name() = "Contact"][1]//text()')) + self.attribution = ''.join(doc.xpath('//*[local-name() = "Attribution"][1]//text()')) + self.syndicationright = ''.join(doc.xpath('//*[local-name() = "SyndicationRight"][1]//text()')) + + tag_text = ' '.join(doc.xpath('//*[local-name() = "Tags"]//text()')) + if tag_text != None: + self.tags = tag_text.split(' ') + + self.adultcontent = doc.xpath('boolean(//*[local-name() = "AdultContent" and contains(., "true")])') + + def get_url_by_type(self, type): + ''' + Walks available urls and returns them by type. Only + appropriate in opensearch v1.1 where there can be multiple + query targets. Returns none if no such type is found. + + url = description.get_url_by_type('application/rss+xml') + ''' + for url in self.urls: + if url.type == type: + return url + return None + + def get_best_template(self): + ''' + OK, best is a value judgement, but so be it. You'll get + back either the atom, rss or first template available. This + method handles the main difference between opensearch v1.0 and v1.1 + ''' + # version 1.0 + if self.url: + return self.url + + # atom + if self.get_url_by_type('application/atom+xml'): + return self.get_url_by_type('application/atom+xml').template + + # rss + if self.get_url_by_type('application/rss+xml'): + return self.get_url_by_type('application/rss+xml').template + + # other possible rss type + if self.get_url_by_type('text/xml'): + return self.get_url_by_Type('text/xml').template + + # otherwise just the first one + if len(self.urls) > 0: + return self.urls[0].template + + # out of luck + return None diff --git a/src/calibre/utils/opensearch/query.py b/src/calibre/utils/opensearch/query.py new file mode 100644 index 0000000000..d87eb0145f --- /dev/null +++ b/src/calibre/utils/opensearch/query.py @@ -0,0 +1,74 @@ +# -*- coding: utf-8 -*- + +from __future__ import (unicode_literals, division, absolute_import, print_function) + +__license__ = 'GPL 3' +__copyright__ = '2006, Ed Summers ' +__docformat__ = 'restructuredtext en' + +from urlparse import urlparse, urlunparse, parse_qs +from urllib import urlencode + +class Query(object): + ''' + Represents an opensearch query Really this class is just a + helper for substituting values into the macros in a format. + + format = 'http://beta.indeed.com/opensearch?q={searchTerms}&start={startIndex}&limit={count}' + q = Query(format) + q.searchTerms('zx81') + q.startIndex = 1 + q.count = 25 + print q.url() + ''' + + standard_macros = ['searchTerms', 'count', 'startIndex', 'startPage', + 'language', 'outputEncoding', 'inputEncoding'] + + def __init__(self, format): + ''' + Create a query object by passing it the url format obtained + from the opensearch Description. + ''' + self.format = format + + # unpack the url to a tuple + self.url_parts = urlparse(format) + + # unpack the query string to a dictionary + self.query_string = parse_qs(self.url_parts[4]) + + # look for standard macros and create a mapping of the + # opensearch names to the service specific ones + # so q={searchTerms} will result in a mapping between searchTerms and q + self.macro_map = {} + for key,values in self.query_string.items(): + # TODO eventually optional/required params should be + # distinguished somehow (the ones with/without trailing ? + macro = values[0].replace('{', '').replace('}', '').replace('?', '') + if macro in Query.standard_macros: + self.macro_map[macro] = key + + def url(self): + # copy the original query string + query_string = dict(self.query_string) + + # iterate through macros and set the position in the querystring + for macro, name in self.macro_map.items(): + if hasattr(self, macro): + # set the name/value pair + query_string[name] = [getattr(self, macro)] + else: + # remove the name/value pair + del(query_string[name]) + + # copy the url parts and substitute in our new query string + url_parts = list(self.url_parts) + url_parts[4] = urlencode(query_string, 1) + + # recompose and return url + return urlunparse(tuple(url_parts)) + + def has_macro(self, macro): + return self.macro_map.has_key(macro) + diff --git a/src/calibre/utils/opensearch/url.py b/src/calibre/utils/opensearch/url.py new file mode 100644 index 0000000000..f05ec3205a --- /dev/null +++ b/src/calibre/utils/opensearch/url.py @@ -0,0 +1,18 @@ +# -*- coding: utf-8 -*- + +from __future__ import (unicode_literals, division, absolute_import, print_function) + +__license__ = 'GPL 3' +__copyright__ = '2006, Ed Summers ' +__docformat__ = 'restructuredtext en' + +class URL(object): + ''' + Class for representing a URL in an opensearch v1.1 query + ''' + + def __init__(self, type='', template='', method='GET'): + self.type = type + self.template = template + self.method = 'GET' + self.params = [] diff --git a/src/calibre/web/feeds/recipes/collection.py b/src/calibre/web/feeds/recipes/collection.py index 87a65dd9e6..13bae3a554 100644 --- a/src/calibre/web/feeds/recipes/collection.py +++ b/src/calibre/web/feeds/recipes/collection.py @@ -85,7 +85,7 @@ def serialize_builtin_recipes(): return serialize_collection(recipe_mapping) def get_builtin_recipe_collection(): - return etree.parse(P('builtin_recipes.xml')).getroot() + return etree.parse(P('builtin_recipes.xml', allow_user_override=False)).getroot() def get_custom_recipe_collection(*args): from calibre.web.feeds.recipes import compile_recipe, \ @@ -179,7 +179,7 @@ def download_builtin_recipe(urn): return br.open_novisit('http://status.calibre-ebook.com/recipe/'+urn).read() def get_builtin_recipe(urn): - with zipfile.ZipFile(P('builtin_recipes.zip'), 'r') as zf: + with zipfile.ZipFile(P('builtin_recipes.zip', allow_user_override=False), 'r') as zf: return zf.read(urn+'.recipe') def get_builtin_recipe_by_title(title, log=None, download_recipe=False): diff --git a/src/calibre/web/feeds/recipes/model.py b/src/calibre/web/feeds/recipes/model.py index 2c52c85f83..5f8d906e61 100644 --- a/src/calibre/web/feeds/recipes/model.py +++ b/src/calibre/web/feeds/recipes/model.py @@ -140,7 +140,8 @@ class RecipeModel(QAbstractItemModel, SearchQueryParser): self.builtin_recipe_collection = get_builtin_recipe_collection() self.scheduler_config = SchedulerConfig() try: - with zipfile.ZipFile(P('builtin_recipes.zip'), 'r') as zf: + with zipfile.ZipFile(P('builtin_recipes.zip', + allow_user_override=False), 'r') as zf: self.favicons = dict([(x.filename, x) for x in zf.infolist() if x.filename.endswith('.png')]) except: @@ -181,7 +182,7 @@ class RecipeModel(QAbstractItemModel, SearchQueryParser): def do_refresh(self, restrict_to_urns=set([])): self.custom_recipe_collection = get_custom_recipe_collection() - zf = P('builtin_recipes.zip') + zf = P('builtin_recipes.zip', allow_user_override=False) def factory(cls, parent, *args): args = list(args)