Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-06-23 15:30:45 -04:00
Code Issues Packages Projects Releases Wiki Activity

0.9.8+

Browse Source
This commit is contained in:
GRiker 2012-12-06 04:55:14 -07:00
commit 69f8b36eae
121 changed files with 53394 additions and 48586 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
recipes/aksiyon_derigisi.recipe
View File

@ -20,6 +20,7 @@ class Aksiyon (BasicNewsRecipe):
auto_cleanup = True
cover_img_url = 'http://www.aksiyon.com.tr/aksiyon/images/aksiyon/top-page/aksiyon_top_r2_c1.jpg'
masthead_url = 'http://aksiyon.com.tr/aksiyon/images/aksiyon/top-page/aksiyon_top_r2_c1.jpg'
ignore_duplicate_articles = { 'title', 'url' }
remove_empty_feeds= True
feeds = [
( u'KAPAK', u'http://www.aksiyon.com.tr/aksiyon/rss?sectionId=26'),

7
recipes/endgadget.recipe
View File

@ -21,10 +21,11 @@ class Engadget(BasicNewsRecipe):
use_embedded_content = False
remove_javascript = True
remove_empty_feeds = True
auto_cleanup = True
keep_only_tags = [dict(name='div', attrs={'class':['post_content permalink ','post_content permalink alt-post-full']})]
remove_tags = [dict(name='div', attrs={'class':['filed_under','post_footer']})]
remove_tags_after = [dict(name='div', attrs={'class':['post_footer']})]
#keep_only_tags = [dict(name='div', attrs={'class':['post_content permalink ','post_content permalink alt-post-full']})]
#remove_tags = [dict(name='div', attrs={'class':['filed_under','post_footer']})]
#remove_tags_after = [dict(name='div', attrs={'class':['post_footer']})]
feeds = [(u'Posts', u'http://www.engadget.com/rss.xml')]

633
recipes/nytimes.recipe
View File

@ -6,22 +6,41 @@ __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
nytimes.com
'''
import re, string, time
from calibre import entity_to_unicode, strftime
from calibre import strftime
from datetime import timedelta, date
from time import sleep
from calibre.web.feeds.recipes import BasicNewsRecipe
from calibre.ebooks.BeautifulSoup import BeautifulSoup, Tag, BeautifulStoneSoup
class NYTimes(BasicNewsRecipe):
recursions=1 # set this to zero to omit Related articles lists
# set getTechBlogs to True to include the technology blogs
# set tech_oldest_article to control article age
# set tech_max_articles_per_feed to control article count
getTechBlogs = True
remove_empty_feeds = True
tech_oldest_article = 14
tech_max_articles_per_feed = 25
# set headlinesOnly to True for the headlines-only version. If True, webEdition is ignored.
headlinesOnly = True
# set webEdition to True for the Web edition of the newspaper. Set oldest_article to the
# number of days old an article can be for inclusion. If oldest_article = 0 all articles
# will be included. Note: oldest_article is ignored if webEdition = False
# set webEdition to True for the Web edition of the newspaper. Set oldest_web_article to the
# number of days old an article can be for inclusion. If oldest_web_article = None all articles
# will be included. Note: oldest_web_article is ignored if webEdition = False
webEdition = False
oldest_article = 7
oldest_web_article = 7
# download higher resolution images than the small thumbnails typically included in the article
# the down side of having large beautiful images is the file size is much larger, on the order of 7MB per paper
useHighResImages = True
# replace paid Kindle Version: the name will be changed to "The New York Times" to cause
# previous paid versions of the new york times to best sent to the back issues folder on the kindle
replaceKindleVersion = False
# includeSections: List of sections to include. If empty, all sections found will be included.
# Otherwise, only the sections named will be included. For example,
@ -82,57 +101,68 @@ class NYTimes(BasicNewsRecipe):
('Education',u'education'),
('Multimedia',u'multimedia'),
(u'Obituaries',u'obituaries'),
(u'Sunday Magazine',u'magazine'),
(u'Week in Review',u'weekinreview')]
(u'Sunday Magazine',u'magazine')
]
tech_feeds = [
(u'Tech - Pogues Posts', u'http://pogue.blogs.nytimes.com/feed/'),
(u'Tech - Bits', u'http://bits.blogs.nytimes.com/feed/'),
(u'Tech - Gadgetwise', u'http://gadgetwise.blogs.nytimes.com/feed/'),
(u'Tech - Open', u'http://open.blogs.nytimes.com/feed/')
]
if headlinesOnly:
title='New York Times Headlines'
description = 'Headlines from the New York Times. Needs a subscription from http://www.nytimes.com'
needs_subscription = 'optional'
description = 'Headlines from the New York Times'
needs_subscription = False
elif webEdition:
title='New York Times (Web)'
description = 'New York Times on the Web'
needs_subscription = True
needs_subscription = False
elif replaceKindleVersion:
title='The New York Times'
description = 'Today\'s New York Times'
needs_subscription = False
else:
title='New York Times'
description = 'Today\'s New York Times'
needs_subscription = True
needs_subscription = False
month_list = ['january','february','march','april','may','june','july','august','september','october','november','december']
def decode_us_date(self,datestr):
udate = datestr.strip().lower().split()
def decode_url_date(self,url):
urlitems = url.split('/')
try:
m = self.month_list.index(udate[0])+1
d = date(int(urlitems[3]),int(urlitems[4]),int(urlitems[5]))
except:
return date.today()
d = int(udate[1])
y = int(udate[2])
try:
d = date(y,m,d)
d = date(int(urlitems[4]),int(urlitems[5]),int(urlitems[6]))
except:
d = date.today
return None
return d
earliest_date = date.today() - timedelta(days=oldest_article)
if oldest_web_article is None:
earliest_date = date.today()
else:
earliest_date = date.today() - timedelta(days=oldest_web_article)
oldest_article = 365 # by default, a long time ago
__author__ = 'GRiker/Kovid Goyal/Nick Redding'
language = 'en'
requires_version = (0, 7, 5)
encoding = 'utf-8'
timefmt = ''
masthead_url = 'http://graphics8.nytimes.com/images/misc/nytlogo379x64.gif'
simultaneous_downloads = 1
cover_margins = (18,18,'grey99')
remove_tags_before = dict(id='article')
remove_tags_after = dict(id='article')
remove_tags = [dict(attrs={'class':[
remove_tags = [
dict(attrs={'class':[
'articleFooter',
'articleTools',
'columnGroup doubleRule',
'columnGroup singleRule',
'columnGroup last',
'columnGroup last',
@ -140,7 +170,6 @@ class NYTimes(BasicNewsRecipe):
'dottedLine',
'entry-meta',
'entry-response module',
'icon enlargeThis',
'leftNavTabs',
'metaFootnote',
'module box nav',
@ -150,10 +179,43 @@ class NYTimes(BasicNewsRecipe):
'relatedSearchesModule',
'side_tool',
'singleAd',
'entry entry-utility', #added for DealBook
'entry-tags', #added for DealBook
'footer promos clearfix', #added for DealBook
'footer links clearfix', #added for DealBook
'tabsContainer', #added for other blog downloads
'column lastColumn', #added for other blog downloads
'pageHeaderWithLabel', #added for other gadgetwise downloads
'column two', #added for other blog downloads
'column two last', #added for other blog downloads
'column three', #added for other blog downloads
'column three last', #added for other blog downloads
'column four',#added for other blog downloads
'column four last',#added for other blog downloads
'column last', #added for other blog downloads
'entry entry-related',
'subNavigation tabContent active', #caucus blog navigation
'mediaOverlay slideshow',
'wideThumb',
'video', #added 02-11-2011
'videoHeader',#added 02-11-2011
'articleInlineVideoHolder', #added 02-11-2011
'assetCompanionAd',
re.compile('^subNavigation'),
re.compile('^leaderboard'),
re.compile('^module'),
re.compile('commentCount')
]}),
dict(name='div', attrs={'class':re.compile('toolsList')}), # bits
dict(name='div', attrs={'class':re.compile('postNavigation')}), # bits
dict(name='div', attrs={'class':'tweet'}),
dict(name='span', attrs={'class':'commentCount meta'}),
dict(name='div', attrs={'id':'header'}),
dict(name='div', attrs={'id':re.compile('commentsContainer')}), # bits, pogue, gadgetwise, open
dict(name='ul', attrs={'class':re.compile('entry-tools')}), # pogue, gadgetwise
dict(name='div', attrs={'class':re.compile('nocontent')}), # pogue, gadgetwise
dict(name='div', attrs={'id':re.compile('respond')}), # open
dict(name='div', attrs={'class':re.compile('entry-tags')}), # pogue
dict(id=[
'adxLeaderboard',
'adxSponLink',
@ -183,22 +245,29 @@ class NYTimes(BasicNewsRecipe):
'side_index',
'side_tool',
'toolsRight',
'skybox', #added for DealBook
'TopAd', #added for DealBook
'related-content', #added for DealBook
]),
dict(name=['script', 'noscript', 'style','form','hr'])]
no_stylesheets = True
extra_css = '''
.articleHeadline { text-align: left; margin-top:0.5em; margin-bottom:0.25em; }
.credit { text-align: right; font-size: small; line-height:1em; margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.byline { text-align: left; font-size: small; line-height:1em; margin-top:10px; margin-left:0; margin-right:0; margin-bottom: 0; }
.dateline { text-align: left; font-size: small; line-height:1em;margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.kicker { font-size: small; line-height:1em;margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.timestamp { text-align: left; font-size: small; }
.caption { font-size: small; font-style:italic; line-height:1em; margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.credit { font-weight: normal; text-align: right; font-size: 50%; line-height:1em; margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.byline { text-align: left; font-size: 50%; line-height:1em; margin-top:10px; margin-left:0; margin-right:0; margin-bottom: 0; }
.dateline { text-align: left; font-size: 50%; line-height:1em;margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.kicker { font-size: 50%; line-height:1em;margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.timestamp { font-weight: normal; text-align: left; font-size: 50%; }
.caption { font-size: 50%; font-style:italic; line-height:1em; margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
a:link {text-decoration: none; }
.date{font-size: 50%; }
.update{font-size: 50%; }
.articleBody { }
.authorId {text-align: left; }
.authorId {text-align: left; font-size: 50%; }
.image {text-align: center;}
.source {text-align: left; }'''
.aside {color:blue;margin:0px 0px 0px 0px; padding: 0px 0px 0px 0px; font-size:100%;}
.asidenote {color:blue;margin:0px 0px 0px 0px; padding: 0px 0px 0px 0px; font-size:100%;font-weight:bold;}
.source {text-align: left; font-size: x-small; }'''
articles = {}
@ -237,7 +306,7 @@ class NYTimes(BasicNewsRecipe):
def exclude_url(self,url):
if not url.startswith("http"):
return True
if not url.endswith(".html"):
if not url.endswith(".html") and 'dealbook.nytimes.com' not in url: #added for DealBook
return True
if 'nytimes.com' not in url:
return True
@ -280,88 +349,91 @@ class NYTimes(BasicNewsRecipe):
def get_browser(self):
br = BasicNewsRecipe.get_browser()
if self.username is not None and self.password is not None:
br.open('http://www.nytimes.com/auth/login')
br.form = br.forms().next()
br['userid'] = self.username
br['password'] = self.password
raw = br.submit().read()
if 'Please try again' in raw:
raise Exception('Your username and password are incorrect')
return br
def skip_ad_pages(self, soup):
# Skip ad pages served before actual article
skip_tag = soup.find(True, {'name':'skip'})
if skip_tag is not None:
self.log.warn("Found forwarding link: %s" % skip_tag.parent['href'])
url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
url += '?pagewanted=all'
self.log.warn("Skipping ad to article at '%s'" % url)
return self.index_to_soup(url, raw=True)
## This doesn't work (and probably never did). It either gets another serve of the advertisement,
## or if it gets the article then get_soup (from which it is invoked) traps trying to do xml decoding.
##
## def skip_ad_pages(self, soup):
## # Skip ad pages served before actual article
## skip_tag = soup.find(True, {'name':'skip'})
## if skip_tag is not None:
## self.log.warn("Found forwarding link: %s" % skip_tag.parent['href'])
## url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
## url += '?pagewanted=all'
## self.log.warn("Skipping ad to article at '%s'" % url)
## return self.index_to_soup(url, raw=True)
cover_tag = 'NY_NYT'
def get_cover_url(self):
cover = None
st = time.localtime()
year = str(st.tm_year)
month = "%.2d" % st.tm_mon
day = "%.2d" % st.tm_mday
cover = 'http://graphics8.nytimes.com/images/' + year + '/' + month +'/' + day +'/nytfrontpage/scan.jpg'
from datetime import timedelta, date
cover = 'http://webmedia.newseum.org/newseum-multimedia/dfp/jpg'+str(date.today().day)+'/lg/'+self.cover_tag+'.jpg'
br = BasicNewsRecipe.get_browser()
daysback=1
try:
br.open(cover)
except:
while daysback<7:
cover = 'http://webmedia.newseum.org/newseum-multimedia/dfp/jpg'+str((date.today() - timedelta(days=daysback)).day)+'/lg/'+self.cover_tag+'.jpg'
br = BasicNewsRecipe.get_browser()
try:
br.open(cover)
except:
daysback = daysback+1
continue
break
if daysback==7:
self.log("\nCover unavailable")
cover = None
return cover
masthead_url = 'http://graphics8.nytimes.com/images/misc/nytlogo379x64.gif'
def short_title(self):
return self.title
def index_to_soup(self, url_or_raw, raw=False):
'''
OVERRIDE of class method
deals with various page encodings between index and articles
'''
def get_the_soup(docEncoding, url_or_raw, raw=False) :
def article_to_soup(self, url_or_raw, raw=False):
from contextlib import closing
import copy
from calibre.ebooks.chardet import xml_to_unicode
if re.match(r'\w+://', url_or_raw):
br = self.clone_browser(self.browser)
f = br.open_novisit(url_or_raw)
open_func = getattr(br, 'open_novisit', br.open)
with closing(open_func(url_or_raw)) as f:
_raw = f.read()
f.close()
if not _raw:
raise RuntimeError('Could not fetch index from %s'%url_or_raw)
else:
_raw = url_or_raw
if raw:
return _raw
if not isinstance(_raw, unicode) and self.encoding:
_raw = _raw.decode(docEncoding, 'replace')
massage = list(BeautifulSoup.MARKUP_MASSAGE)
massage.append((re.compile(r'&(\S+?);'), lambda match: entity_to_unicode(match, encoding=self.encoding)))
return BeautifulSoup(_raw, markupMassage=massage)
if callable(self.encoding):
_raw = self.encoding(_raw)
else:
_raw = _raw.decode(self.encoding, 'replace')
# Entry point
soup = get_the_soup( self.encoding, url_or_raw )
contentType = soup.find(True,attrs={'http-equiv':'Content-Type'})
docEncoding = str(contentType)[str(contentType).find('charset=') + len('charset='):str(contentType).rfind('"')]
if docEncoding == '' :
docEncoding = self.encoding
nmassage = copy.copy(BeautifulSoup.MARKUP_MASSAGE)
nmassage.extend(self.preprocess_regexps)
nmassage += [(re.compile(r'<!DOCTYPE .+?>', re.DOTALL), lambda m: '')]
# Some websites have buggy doctype declarations that mess up beautifulsoup
# Remove comments as they can leave detritus when extracting tags leaves
# multiple nested comments
nmassage.append((re.compile(r'<!--.*?-->', re.DOTALL), lambda m: ''))
usrc = xml_to_unicode(_raw, self.verbose, strip_encoding_pats=True)[0]
usrc = self.preprocess_raw_html(usrc, url_or_raw)
return BeautifulSoup(usrc, markupMassage=nmassage)
if self.verbose > 2:
self.log( " document encoding: '%s'" % docEncoding)
if docEncoding != self.encoding :
soup = get_the_soup(docEncoding, url_or_raw)
return soup
def massageNCXText(self, description):
# Kindle TOC descriptions won't render certain characters
if description:
massaged = unicode(BeautifulStoneSoup(description, convertEntities=BeautifulStoneSoup.HTML_ENTITIES))
# Replace '&' with '&'
massaged = re.sub("&","&", massaged)
massaged = re.sub("&#038;","&", massaged)
massaged = re.sub("&amp;","&", massaged)
return self.fixChars(massaged)
else:
return description
@ -383,6 +455,16 @@ class NYTimes(BasicNewsRecipe):
if self.filterDuplicates:
if url in self.url_list:
return
if self.webEdition:
date_tag = self.decode_url_date(url)
if date_tag is not None:
if self.oldest_web_article is not None:
if date_tag < self.earliest_date:
self.log("Skipping article %s" % url)
return
else:
self.log("Skipping article %s" % url)
return
self.url_list.append(url)
title = self.tag_to_string(a, use_alt=True).strip()
description = ''
@ -407,6 +489,31 @@ class NYTimes(BasicNewsRecipe):
description=description, author=author,
content=''))
def get_tech_feeds(self,ans):
if self.getTechBlogs:
tech_articles = {}
key_list = []
save_oldest_article = self.oldest_article
save_max_articles_per_feed = self.max_articles_per_feed
self.oldest_article = self.tech_oldest_article
self.max_articles_per_feed = self.tech_max_articles_per_feed
self.feeds = self.tech_feeds
tech = self.parse_feeds()
self.oldest_article = save_oldest_article
self.max_articles_per_feed = save_max_articles_per_feed
self.feeds = None
for f in tech:
key_list.append(f.title)
tech_articles[f.title] = []
for a in f.articles:
tech_articles[f.title].append(
dict(title=a.title, url=a.url, date=a.date,
description=a.summary, author=a.author,
content=a.content))
tech_ans = [(k, tech_articles[k]) for k in key_list if tech_articles.has_key(k)]
for x in tech_ans:
ans.append(x)
return ans
def parse_web_edition(self):
@ -418,31 +525,41 @@ class NYTimes(BasicNewsRecipe):
if sec_title in self.excludeSections:
print "SECTION EXCLUDED: ",sec_title
continue
print 'Index URL: '+'http://www.nytimes.com/pages/'+index_url+'/index.html'
try:
soup = self.index_to_soup('http://www.nytimes.com/pages/'+index_url+'/index.html')
except:
continue
print 'Index URL: '+'http://www.nytimes.com/pages/'+index_url+'/index.html'
self.key = sec_title
# Find each article
for div in soup.findAll(True,
attrs={'class':['section-headline', 'story', 'story headline','sectionHeader','headlinesOnly multiline flush']}):
if div['class'] in ['story', 'story headline'] :
attrs={'class':['section-headline', 'ledeStory', 'story', 'story headline','sectionHeader','headlinesOnly multiline flush']}):
if div['class'] in ['story', 'story headline', 'storyHeader'] :
self.handle_article(div)
elif div['class'] == 'ledeStory':
divsub = div.find('div','storyHeader')
if divsub is not None:
self.handle_article(divsub)
ulrefer = div.find('ul','refer')
if ulrefer is not None:
for lidiv in ulrefer.findAll('li'):
self.handle_article(lidiv)
elif div['class'] == 'headlinesOnly multiline flush':
for lidiv in div.findAll('li'):
self.handle_article(lidiv)
self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
return self.filter_ans(self.ans)
return self.filter_ans(self.get_tech_feeds(self.ans))
def parse_todays_index(self):
soup = self.index_to_soup('http://www.nytimes.com/pages/todayspaper/index.html')
skipping = False
# Find each article
for div in soup.findAll(True,
attrs={'class':['section-headline', 'story', 'story headline','sectionHeader','headlinesOnly multiline flush']}):
if div['class'] in ['section-headline','sectionHeader']:
self.key = string.capwords(self.feed_title(div))
self.key = self.key.replace('Op-ed','Op-Ed')
@ -466,7 +583,7 @@ class NYTimes(BasicNewsRecipe):
self.handle_article(lidiv)
self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
return self.filter_ans(self.ans)
return self.filter_ans(self.get_tech_feeds(self.ans))
def parse_headline_index(self):
@ -514,7 +631,7 @@ class NYTimes(BasicNewsRecipe):
for h3_item in search_div.findAll('h3'):
byline = h3_item.h6
if byline is not None:
author = self.tag_to_string(byline,usa_alt=False)
author = self.tag_to_string(byline,use_alt=False)
else:
author = ''
a = h3_item.find('a', href=True)
@ -540,7 +657,7 @@ class NYTimes(BasicNewsRecipe):
self.articles[section_name].append(dict(title=title, url=url, date=pubdate, description=description, author=author, content=''))
self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
return self.filter_ans(self.ans)
return self.filter_ans(self.get_tech_feeds(self.ans))
def parse_index(self):
if self.headlinesOnly:
@ -550,32 +667,190 @@ class NYTimes(BasicNewsRecipe):
else:
return self.parse_todays_index()
def strip_anchors(self,soup):
def strip_anchors(self,soup,kill_all=False):
paras = soup.findAll(True)
for para in paras:
aTags = para.findAll('a')
for a in aTags:
if a.img is None:
a.replaceWith(a.renderContents().decode('cp1252','replace'))
if kill_all or (self.recursions==0):
a.replaceWith(self.tag_to_string(a,False))
else:
if a.has_key('href'):
if a['href'].startswith('http://www.nytimes'):
if not a['href'].endswith('pagewanted=all'):
url = re.sub(r'\?.*', '', a['href'])
if self.exclude_url(url):
a.replaceWith(self.tag_to_string(a,False))
else:
a['href'] = url+'?pagewanted=all'
elif not (a['href'].startswith('http://pogue') or \
a['href'].startswith('http://bits') or \
a['href'].startswith('http://travel') or \
a['href'].startswith('http://business') or \
a['href'].startswith('http://tech') or \
a['href'].startswith('http://health') or \
a['href'].startswith('http://dealbook') or \
a['href'].startswith('http://open')):
a.replaceWith(self.tag_to_string(a,False))
return soup
def handle_tags(self,soup):
try:
print("HANDLE TAGS: TITLE = "+self.tag_to_string(soup.title))
except:
print("HANDLE TAGS: NO TITLE")
if soup is None:
print("ERROR: handle_tags received NoneType")
return None
## print("HANDLING AD FORWARD:")
## print(soup)
if self.keep_only_tags:
body = Tag(soup, 'body')
try:
if isinstance(self.keep_only_tags, dict):
self.keep_only_tags = [self.keep_only_tags]
for spec in self.keep_only_tags:
for tag in soup.find('body').findAll(**spec):
body.insert(len(body.contents), tag)
soup.find('body').replaceWith(body)
except AttributeError: # soup has no body element
pass
def remove_beyond(tag, next):
while tag is not None and getattr(tag, 'name', None) != 'body':
after = getattr(tag, next)
while after is not None:
ns = getattr(tag, next)
after.extract()
after = ns
tag = tag.parent
if self.remove_tags_after is not None:
rt = [self.remove_tags_after] if isinstance(self.remove_tags_after, dict) else self.remove_tags_after
for spec in rt:
tag = soup.find(**spec)
remove_beyond(tag, 'nextSibling')
if self.remove_tags_before is not None:
tag = soup.find(**self.remove_tags_before)
remove_beyond(tag, 'previousSibling')
for kwds in self.remove_tags:
for tag in soup.findAll(**kwds):
tag.extract()
return soup
def preprocess_html(self, soup):
print("PREPROCESS TITLE="+self.tag_to_string(soup.title))
skip_tag = soup.find(True, {'name':'skip'})
if skip_tag is not None:
url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
url += '?pagewanted=all'
self.log.warn("Skipping ad to article at '%s'" % url)
sleep(5)
soup = self.handle_tags(self.article_to_soup(url))
if self.webEdition & (self.oldest_article>0):
date_tag = soup.find(True,attrs={'class': ['dateline','date']})
if date_tag:
date_str = self.tag_to_string(date_tag,use_alt=False)
date_str = date_str.replace('Published:','')
date_items = date_str.split(',')
try:
datestring = date_items[0]+' '+date_items[1]
article_date = self.decode_us_date(datestring)
except:
article_date = date.today()
if article_date < self.earliest_date:
self.log("Skipping article dated %s" % date_str)
return None
# check if the article is from one of the tech blogs
blog=soup.find('div',attrs={'id':['pogue','bits','gadgetwise','open']})
if blog is not None:
old_body = soup.find('body')
new_body=Tag(soup,'body')
new_body.append(soup.find('div',attrs={'id':'content'}))
new_body.find('div',attrs={'id':'content'})['id']='blogcontent' # identify for postprocess_html
old_body.replaceWith(new_body)
for divr in soup.findAll('div',attrs={'class':re.compile('w190 right')}):
if divr.find(text=re.compile('Sign up')):
divr.extract()
divr = soup.find('div',attrs={'id':re.compile('related-content')})
if divr is not None:
# handle related articles
rlist = []
ul = divr.find('ul')
if ul is not None:
for li in ul.findAll('li'):
atag = li.find('a')
if atag is not None:
if atag['href'].startswith('http://pogue') or atag['href'].startswith('http://bits') or \
atag['href'].startswith('http://open'):
atag.find(text=True).replaceWith(self.massageNCXText(self.tag_to_string(atag,False)))
rlist.append(atag)
divr.extract()
if rlist != []:
asidediv = Tag(soup,'div',[('class','aside')])
if soup.find('hr') is None:
asidediv.append(Tag(soup,'hr'))
h4 = Tag(soup,'h4',[('class','asidenote')])
h4.insert(0,"Related Posts")
asidediv.append(h4)
ul = Tag(soup,'ul')
for r in rlist:
li = Tag(soup,'li',[('class','aside')])
r['class'] = 'aside'
li.append(r)
ul.append(li)
asidediv.append(ul)
asidediv.append(Tag(soup,'hr'))
smain = soup.find('body')
smain.append(asidediv)
for atag in soup.findAll('a'):
img = atag.find('img')
if img is not None:
atag.replaceWith(img)
elif not atag.has_key('href'):
atag.replaceWith(atag.renderContents().decode('cp1252','replace'))
elif not (atag['href'].startswith('http://www.nytimes') or atag['href'].startswith('http://pogue') or \
atag['href'].startswith('http://bits') or atag['href'].startswith('http://open')):
atag.replaceWith(atag.renderContents().decode('cp1252','replace'))
hdr = soup.find('address')
if hdr is not None:
hdr.name='span'
for span_credit in soup.findAll('span','credit'):
sp = Tag(soup,'span')
span_credit.replaceWith(sp)
sp.append(Tag(soup,'br'))
sp.append(span_credit)
sp.append(Tag(soup,'br'))
else: # nytimes article
related = [] # these will be the related articles
first_outer = None # first related outer tag
first_related = None # first related tag
for outerdiv in soup.findAll(attrs={'class': re.compile('articleInline runaroundLeft')}):
for rdiv in soup.findAll('div','columnGroup doubleRule'):
if rdiv.find('h3') is not None:
if self.tag_to_string(rdiv.h3,False).startswith('Related'):
rdiv.h3.find(text=True).replaceWith("Related articles")
rdiv.h3['class'] = 'asidenote'
for litag in rdiv.findAll('li'):
if litag.find('a') is not None:
if litag.find('a')['href'].startswith('http://www.nytimes.com'):
url = re.sub(r'\?.*', '', litag.find('a')['href'])
litag.find('a')['href'] = url+'?pagewanted=all'
litag.extract()
related.append(litag)
if first_related is None:
first_related = rdiv
first_outer = outerdiv
else:
litag.extract()
if related != []:
for r in related:
if r.h6: # don't want the anchor inside a h6 tag
r.h6.replaceWith(r.h6.a)
first_related.ul.append(r)
first_related.insert(0,Tag(soup,'hr'))
first_related.append(Tag(soup,'hr'))
first_related['class'] = 'aside'
first_outer.replaceWith(first_related) # replace the outer tag with the related tag
for rdiv in soup.findAll(attrs={'class': re.compile('articleInline runaroundLeft')}):
rdiv.extract()
kicker_tag = soup.find(attrs={'class':'kicker'})
if kicker_tag: # remove Op_Ed author head shots
@ -584,9 +859,77 @@ class NYTimes(BasicNewsRecipe):
img_div = soup.find('div','inlineImage module')
if img_div:
img_div.extract()
return self.strip_anchors(soup)
def postprocess_html(self,soup, True):
if self.useHighResImages:
try:
#open up all the "Enlarge this Image" pop-ups and download the full resolution jpegs
enlargeThisList = soup.findAll('div',{'class':'icon enlargeThis'})
if enlargeThisList:
for popupref in enlargeThisList:
popupreflink = popupref.find('a')
if popupreflink:
reflinkstring = str(popupreflink['href'])
refstart = reflinkstring.find("javascript:pop_me_up2('") + len("javascript:pop_me_up2('")
refend = reflinkstring.find(".html", refstart) + len(".html")
reflinkstring = reflinkstring[refstart:refend]
popuppage = self.browser.open(reflinkstring)
popuphtml = popuppage.read()
popuppage.close()
if popuphtml:
st = time.localtime()
year = str(st.tm_year)
month = "%.2d" % st.tm_mon
day = "%.2d" % st.tm_mday
imgstartpos = popuphtml.find('http://graphics8.nytimes.com/images/' + year + '/' + month +'/' + day +'/') + len('http://graphics8.nytimes.com/images/' + year + '/' + month +'/' + day +'/')
highResImageLink = 'http://graphics8.nytimes.com/images/' + year + '/' + month +'/' + day +'/' + popuphtml[imgstartpos:popuphtml.find('.jpg',imgstartpos)+4]
popupSoup = BeautifulSoup(popuphtml)
highResTag = popupSoup.find('img', {'src':highResImageLink})
if highResTag:
try:
newWidth = highResTag['width']
newHeight = highResTag['height']
imageTag = popupref.parent.find("img")
except:
self.log("Error: finding width and height of img")
popupref.extract()
if imageTag:
try:
imageTag['src'] = highResImageLink
imageTag['width'] = newWidth
imageTag['height'] = newHeight
except:
self.log("Error setting the src width and height parameters")
except Exception:
self.log("Error pulling high resolution images")
try:
#in case pulling images failed, delete the enlarge this text
enlargeThisList = soup.findAll('div',{'class':'icon enlargeThis'})
if enlargeThisList:
for popupref in enlargeThisList:
popupref.extract()
except:
self.log("Error removing Enlarge this text")
return self.strip_anchors(soup,False)
def postprocess_html(self,soup,first_fetch):
if not first_fetch: # remove Related links
for aside in soup.findAll('div','aside'):
aside.extract()
soup = self.strip_anchors(soup,True)
if soup.find('div',attrs={'id':'blogcontent'}) is None:
if first_fetch:
aside = soup.find('div','aside')
if aside is not None: # move the related list to the end of the article
art = soup.find('div',attrs={'id':'article'})
if art is None:
art = soup.find('div',attrs={'class':'article'})
if art is not None:
art.append(aside)
try:
if self.one_picture_per_article:
# Remove all images after first
@ -642,6 +985,7 @@ class NYTimes(BasicNewsRecipe):
try:
# Change <nyt_headline> to <h2>
h1 = soup.find('h1')
blogheadline = str(h1) #added for dealbook
if h1:
headline = h1.find("nyt_headline")
if headline:
@ -649,13 +993,19 @@ class NYTimes(BasicNewsRecipe):
tag['class'] = "headline"
tag.insert(0, self.fixChars(headline.contents[0]))
h1.replaceWith(tag)
elif blogheadline.find('entry-title'):#added for dealbook
tag = Tag(soup, "h2")#added for dealbook
tag['class'] = "headline"#added for dealbook
tag.insert(0, self.fixChars(h1.contents[0]))#added for dealbook
h1.replaceWith(tag)#added for dealbook
else:
# Blog entry - replace headline, remove <hr> tags
# Blog entry - replace headline, remove <hr> tags - BCC I think this is no longer functional 1-18-2011
headline = soup.find('title')
if headline:
tag = Tag(soup, "h2")
tag['class'] = "headline"
tag.insert(0, self.fixChars(headline.contents[0]))
tag.insert(0, self.fixChars(headline.renderContents()))
soup.insert(0, tag)
hrs = soup.findAll('hr')
for hr in hrs:
@ -663,6 +1013,29 @@ class NYTimes(BasicNewsRecipe):
except:
self.log("ERROR: Problem in Change <nyt_headline> to <h2>")
try:
#if this is from a blog (dealbook, fix the byline format
bylineauthor = soup.find('address',attrs={'class':'byline author vcard'})
if bylineauthor:
tag = Tag(soup, "h6")
tag['class'] = "byline"
tag.insert(0, self.fixChars(bylineauthor.renderContents()))
bylineauthor.replaceWith(tag)
except:
self.log("ERROR: fixing byline author format")
try:
#if this is a blog (dealbook) fix the credit style for the pictures
blogcredit = soup.find('div',attrs={'class':'credit'})
if blogcredit:
tag = Tag(soup, "h6")
tag['class'] = "credit"
tag.insert(0, self.fixChars(blogcredit.renderContents()))
blogcredit.replaceWith(tag)
except:
self.log("ERROR: fixing credit format")
try:
# Change <h1> to <h3> - used in editorial blogs
masthead = soup.find("h1")
@ -685,6 +1058,13 @@ class NYTimes(BasicNewsRecipe):
subhead.replaceWith(bTag)
except:
self.log("ERROR: Problem in Change <h1> to <h3> - used in editorial blogs")
try:
#remove the <strong> update tag
blogupdated = soup.find('span', {'class':'update'})
if blogupdated:
blogupdated.replaceWith("")
except:
self.log("ERROR: Removing strong tag")
try:
divTag = soup.find('div',attrs={'id':'articleBody'})
@ -708,16 +1088,16 @@ class NYTimes(BasicNewsRecipe):
return soup
def populate_article_metadata(self, article, soup, first):
if first and hasattr(self, 'add_toc_thumbnail'):
if not first:
return
idxdiv = soup.find('div',attrs={'class':'articleSpanImage'})
if idxdiv is not None:
if idxdiv.img:
self.add_toc_thumbnail(article, idxdiv.img['src'])
self.add_toc_thumbnail(article, re.sub(r'links\\link\d+\\','',idxdiv.img['src']))
else:
img = soup.find('img')
img = soup.find('body').find('img')
if img is not None:
self.add_toc_thumbnail(article, img['src'])
self.add_toc_thumbnail(article, re.sub(r'links\\link\d+\\','',img['src']))
shortparagraph = ""
try:
if len(article.text_summary.strip()) == 0:
@ -731,13 +1111,22 @@ class NYTimes(BasicNewsRecipe):
#account for blank paragraphs and short paragraphs by appending them to longer ones
if len(refparagraph) > 0:
if len(refparagraph) > 70: #approximately one line of text
article.summary = article.text_summary = shortparagraph + refparagraph
newpara = shortparagraph + refparagraph
newparaDateline,newparaEm,newparaDesc = newpara.partition('&mdash;')
if newparaEm == '':
newparaDateline,newparaEm,newparaDesc = newpara.partition('—')
if newparaEm == '':
newparaDesc = newparaDateline
article.summary = article.text_summary = newparaDesc.strip()
return
else:
shortparagraph = refparagraph + " "
if shortparagraph.strip().find(" ") == -1 and not shortparagraph.strip().endswith(":"):
shortparagraph = shortparagraph + "- "
else:
article.summary = article.text_summary = self.massageNCXText(article.text_summary)
except:
self.log("Error creating article descriptions")
return

560
recipes/nytimes_sub.recipe
View File

@ -6,31 +6,42 @@ __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
nytimes.com
'''
import re, string, time
from calibre import entity_to_unicode, strftime
from calibre import strftime
from datetime import timedelta, date
from time import sleep
from calibre.web.feeds.recipes import BasicNewsRecipe
from calibre.ebooks.BeautifulSoup import BeautifulSoup, Tag, BeautifulStoneSoup
class NYTimes(BasicNewsRecipe):
recursions=1 # set this to zero to omit Related articles lists
# set getTechBlogs to True to include the technology blogs
# set tech_oldest_article to control article age
# set tech_max_articles_per_feed to control article count
getTechBlogs = True
remove_empty_feeds = True
tech_oldest_article = 14
tech_max_articles_per_feed = 25
# set headlinesOnly to True for the headlines-only version. If True, webEdition is ignored.
headlinesOnly = False
# set webEdition to True for the Web edition of the newspaper. Set oldest_article to the
# number of days old an article can be for inclusion. If oldest_article = 0 all articles
# will be included. Note: oldest_article is ignored if webEdition = False
# set webEdition to True for the Web edition of the newspaper. Set oldest_web_article to the
# number of days old an article can be for inclusion. If oldest_web_article = None all articles
# will be included. Note: oldest_web_article is ignored if webEdition = False
webEdition = False
oldest_article = 7
# replace paid Kindle Version: the name will be changed to "The New York Times" to cause
# previous paid versions of the new york times to best sent to the back issues folder on the kindle
replaceKindleVersion = False
oldest_web_article = 7
# download higher resolution images than the small thumbnails typically included in the article
# the down side of having large beautiful images is the file size is much larger, on the order of 7MB per paper
useHighResImages = True
# replace paid Kindle Version: the name will be changed to "The New York Times" to cause
# previous paid versions of the new york times to best sent to the back issues folder on the kindle
replaceKindleVersion = False
# includeSections: List of sections to include. If empty, all sections found will be included.
# Otherwise, only the sections named will be included. For example,
#
@ -90,60 +101,68 @@ class NYTimes(BasicNewsRecipe):
('Education',u'education'),
('Multimedia',u'multimedia'),
(u'Obituaries',u'obituaries'),
(u'Sunday Magazine',u'magazine'),
(u'Week in Review',u'weekinreview')]
(u'Sunday Magazine',u'magazine')
]
tech_feeds = [
(u'Tech - Pogues Posts', u'http://pogue.blogs.nytimes.com/feed/'),
(u'Tech - Bits', u'http://bits.blogs.nytimes.com/feed/'),
(u'Tech - Gadgetwise', u'http://gadgetwise.blogs.nytimes.com/feed/'),
(u'Tech - Open', u'http://open.blogs.nytimes.com/feed/')
]
if headlinesOnly:
title='New York Times Headlines'
description = 'Headlines from the New York Times'
needs_subscription = True
needs_subscription = False
elif webEdition:
title='New York Times (Web)'
description = 'New York Times on the Web'
needs_subscription = True
needs_subscription = False
elif replaceKindleVersion:
title='The New York Times'
description = 'Today\'s New York Times'
needs_subscription = True
needs_subscription = False
else:
title='New York Times'
description = 'Today\'s New York Times. Needs subscription from http://www.nytimes.com'
needs_subscription = True
description = 'Today\'s New York Times'
needs_subscription = False
month_list = ['january','february','march','april','may','june','july','august','september','october','november','december']
def decode_us_date(self,datestr):
udate = datestr.strip().lower().split()
def decode_url_date(self,url):
urlitems = url.split('/')
try:
m = self.month_list.index(udate[0])+1
d = date(int(urlitems[3]),int(urlitems[4]),int(urlitems[5]))
except:
return date.today()
d = int(udate[1])
y = int(udate[2])
try:
d = date(y,m,d)
d = date(int(urlitems[4]),int(urlitems[5]),int(urlitems[6]))
except:
d = date.today
return None
return d
earliest_date = date.today() - timedelta(days=oldest_article)
if oldest_web_article is None:
earliest_date = date.today()
else:
earliest_date = date.today() - timedelta(days=oldest_web_article)
oldest_article = 365 # by default, a long time ago
__author__ = 'GRiker/Kovid Goyal/Nick Redding/Ben Collier'
__author__ = 'GRiker/Kovid Goyal/Nick Redding'
language = 'en'
requires_version = (0, 7, 5)
encoding = 'utf-8'
timefmt = ''
masthead_url = 'http://graphics8.nytimes.com/images/misc/nytlogo379x64.gif'
simultaneous_downloads = 1
cover_margins = (18,18,'grey99')
remove_tags_before = dict(id='article')
remove_tags_after = dict(id='article')
remove_tags = [dict(attrs={'class':[
remove_tags = [
dict(attrs={'class':[
'articleFooter',
'articleTools',
'columnGroup doubleRule',
'columnGroup singleRule',
'columnGroup last',
'columnGroup last',
@ -151,7 +170,6 @@ class NYTimes(BasicNewsRecipe):
'dottedLine',
'entry-meta',
'entry-response module',
#'icon enlargeThis', #removed to provide option for high res images
'leftNavTabs',
'metaFootnote',
'module box nav',
@ -175,12 +193,9 @@ class NYTimes(BasicNewsRecipe):
'column four',#added for other blog downloads
'column four last',#added for other blog downloads
'column last', #added for other blog downloads
'timestamp published', #added for other blog downloads
'entry entry-related',
'subNavigation tabContent active', #caucus blog navigation
'columnGroup doubleRule',
'mediaOverlay slideshow',
'headlinesOnly multiline flush',
'wideThumb',
'video', #added 02-11-2011
'videoHeader',#added 02-11-2011
@ -189,7 +204,18 @@ class NYTimes(BasicNewsRecipe):
re.compile('^subNavigation'),
re.compile('^leaderboard'),
re.compile('^module'),
re.compile('commentCount')
]}),
dict(name='div', attrs={'class':re.compile('toolsList')}), # bits
dict(name='div', attrs={'class':re.compile('postNavigation')}), # bits
dict(name='div', attrs={'class':'tweet'}),
dict(name='span', attrs={'class':'commentCount meta'}),
dict(name='div', attrs={'id':'header'}),
dict(name='div', attrs={'id':re.compile('commentsContainer')}), # bits, pogue, gadgetwise, open
dict(name='ul', attrs={'class':re.compile('entry-tools')}), # pogue, gadgetwise
dict(name='div', attrs={'class':re.compile('nocontent')}), # pogue, gadgetwise
dict(name='div', attrs={'id':re.compile('respond')}), # open
dict(name='div', attrs={'class':re.compile('entry-tags')}), # pogue
dict(id=[
'adxLeaderboard',
'adxSponLink',
@ -227,17 +253,21 @@ class NYTimes(BasicNewsRecipe):
no_stylesheets = True
extra_css = '''
.articleHeadline { text-align: left; margin-top:0.5em; margin-bottom:0.25em; }
.credit { text-align: right; font-size: small; line-height:1em; margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.byline { text-align: left; font-size: small; line-height:1em; margin-top:10px; margin-left:0; margin-right:0; margin-bottom: 0; }
.dateline { text-align: left; font-size: small; line-height:1em;margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.kicker { font-size: small; line-height:1em;margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.timestamp { text-align: left; font-size: small; }
.caption { font-size: small; font-style:italic; line-height:1em; margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.credit { font-weight: normal; text-align: right; font-size: 50%; line-height:1em; margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.byline { text-align: left; font-size: 50%; line-height:1em; margin-top:10px; margin-left:0; margin-right:0; margin-bottom: 0; }
.dateline { text-align: left; font-size: 50%; line-height:1em;margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.kicker { font-size: 50%; line-height:1em;margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
.timestamp { font-weight: normal; text-align: left; font-size: 50%; }
.caption { font-size: 50%; font-style:italic; line-height:1em; margin-top:5px; margin-left:0; margin-right:0; margin-bottom: 0; }
a:link {text-decoration: none; }
.date{font-size: 50%; }
.update{font-size: 50%; }
.articleBody { }
.authorId {text-align: left; }
.authorId {text-align: left; font-size: 50%; }
.image {text-align: center;}
.source {text-align: left; }'''
.aside {color:blue;margin:0px 0px 0px 0px; padding: 0px 0px 0px 0px; font-size:100%;}
.asidenote {color:blue;margin:0px 0px 0px 0px; padding: 0px 0px 0px 0px; font-size:100%;font-weight:bold;}
.source {text-align: left; font-size: x-small; }'''
articles = {}
@ -276,7 +306,7 @@ class NYTimes(BasicNewsRecipe):
def exclude_url(self,url):
if not url.startswith("http"):
return True
if not url.endswith(".html") and 'dealbook.nytimes.com' not in url and 'blogs.nytimes.com' not in url: #added for DealBook
if not url.endswith(".html") and 'dealbook.nytimes.com' not in url: #added for DealBook
return True
if 'nytimes.com' not in url:
return True
@ -319,88 +349,91 @@ class NYTimes(BasicNewsRecipe):
def get_browser(self):
br = BasicNewsRecipe.get_browser()
if self.username is not None and self.password is not None:
br.open('http://www.nytimes.com/auth/login')
br.form = br.forms().next()
br['userid'] = self.username
br['password'] = self.password
raw = br.submit().read()
if 'Please try again' in raw:
raise Exception('Your username and password are incorrect')
return br
def skip_ad_pages(self, soup):
# Skip ad pages served before actual article
skip_tag = soup.find(True, {'name':'skip'})
if skip_tag is not None:
self.log.warn("Found forwarding link: %s" % skip_tag.parent['href'])
url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
url += '?pagewanted=all'
self.log.warn("Skipping ad to article at '%s'" % url)
return self.index_to_soup(url, raw=True)
## This doesn't work (and probably never did). It either gets another serve of the advertisement,
## or if it gets the article then get_soup (from which it is invoked) traps trying to do xml decoding.
##
## def skip_ad_pages(self, soup):
## # Skip ad pages served before actual article
## skip_tag = soup.find(True, {'name':'skip'})
## if skip_tag is not None:
## self.log.warn("Found forwarding link: %s" % skip_tag.parent['href'])
## url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
## url += '?pagewanted=all'
## self.log.warn("Skipping ad to article at '%s'" % url)
## return self.index_to_soup(url, raw=True)
cover_tag = 'NY_NYT'
def get_cover_url(self):
cover = None
st = time.localtime()
year = str(st.tm_year)
month = "%.2d" % st.tm_mon
day = "%.2d" % st.tm_mday
cover = 'http://graphics8.nytimes.com/images/' + year + '/' + month +'/' + day +'/nytfrontpage/scan.jpg'
from datetime import timedelta, date
cover = 'http://webmedia.newseum.org/newseum-multimedia/dfp/jpg'+str(date.today().day)+'/lg/'+self.cover_tag+'.jpg'
br = BasicNewsRecipe.get_browser()
daysback=1
try:
br.open(cover)
except:
while daysback<7:
cover = 'http://webmedia.newseum.org/newseum-multimedia/dfp/jpg'+str((date.today() - timedelta(days=daysback)).day)+'/lg/'+self.cover_tag+'.jpg'
br = BasicNewsRecipe.get_browser()
try:
br.open(cover)
except:
daysback = daysback+1
continue
break
if daysback==7:
self.log("\nCover unavailable")
cover = None
return cover
masthead_url = 'http://graphics8.nytimes.com/images/misc/nytlogo379x64.gif'
def short_title(self):
return self.title
def index_to_soup(self, url_or_raw, raw=False):
'''
OVERRIDE of class method
deals with various page encodings between index and articles
'''
def get_the_soup(docEncoding, url_or_raw, raw=False) :
def article_to_soup(self, url_or_raw, raw=False):
from contextlib import closing
import copy
from calibre.ebooks.chardet import xml_to_unicode
if re.match(r'\w+://', url_or_raw):
br = self.clone_browser(self.browser)
f = br.open_novisit(url_or_raw)
open_func = getattr(br, 'open_novisit', br.open)
with closing(open_func(url_or_raw)) as f:
_raw = f.read()
f.close()
if not _raw:
raise RuntimeError('Could not fetch index from %s'%url_or_raw)
else:
_raw = url_or_raw
if raw:
return _raw
if not isinstance(_raw, unicode) and self.encoding:
_raw = _raw.decode(docEncoding, 'replace')
massage = list(BeautifulSoup.MARKUP_MASSAGE)
massage.append((re.compile(r'&(\S+?);'), lambda match: entity_to_unicode(match, encoding=self.encoding)))
return BeautifulSoup(_raw, markupMassage=massage)
if callable(self.encoding):
_raw = self.encoding(_raw)
else:
_raw = _raw.decode(self.encoding, 'replace')
# Entry point
soup = get_the_soup( self.encoding, url_or_raw )
contentType = soup.find(True,attrs={'http-equiv':'Content-Type'})
docEncoding = str(contentType)[str(contentType).find('charset=') + len('charset='):str(contentType).rfind('"')]
if docEncoding == '' :
docEncoding = self.encoding
nmassage = copy.copy(BeautifulSoup.MARKUP_MASSAGE)
nmassage.extend(self.preprocess_regexps)
nmassage += [(re.compile(r'<!DOCTYPE .+?>', re.DOTALL), lambda m: '')]
# Some websites have buggy doctype declarations that mess up beautifulsoup
# Remove comments as they can leave detritus when extracting tags leaves
# multiple nested comments
nmassage.append((re.compile(r'<!--.*?-->', re.DOTALL), lambda m: ''))
usrc = xml_to_unicode(_raw, self.verbose, strip_encoding_pats=True)[0]
usrc = self.preprocess_raw_html(usrc, url_or_raw)
return BeautifulSoup(usrc, markupMassage=nmassage)
if self.verbose > 2:
self.log( " document encoding: '%s'" % docEncoding)
if docEncoding != self.encoding :
soup = get_the_soup(docEncoding, url_or_raw)
return soup
def massageNCXText(self, description):
# Kindle TOC descriptions won't render certain characters
if description:
massaged = unicode(BeautifulStoneSoup(description, convertEntities=BeautifulStoneSoup.HTML_ENTITIES))
# Replace '&' with '&'
massaged = re.sub("&","&", massaged)
massaged = re.sub("&#038;","&", massaged)
massaged = re.sub("&amp;","&", massaged)
return self.fixChars(massaged)
else:
return description
@ -422,6 +455,16 @@ class NYTimes(BasicNewsRecipe):
if self.filterDuplicates:
if url in self.url_list:
return
if self.webEdition:
date_tag = self.decode_url_date(url)
if date_tag is not None:
if self.oldest_web_article is not None:
if date_tag < self.earliest_date:
self.log("Skipping article %s" % url)
return
else:
self.log("Skipping article %s" % url)
return
self.url_list.append(url)
title = self.tag_to_string(a, use_alt=True).strip()
description = ''
@ -446,6 +489,31 @@ class NYTimes(BasicNewsRecipe):
description=description, author=author,
content=''))
def get_tech_feeds(self,ans):
if self.getTechBlogs:
tech_articles = {}
key_list = []
save_oldest_article = self.oldest_article
save_max_articles_per_feed = self.max_articles_per_feed
self.oldest_article = self.tech_oldest_article
self.max_articles_per_feed = self.tech_max_articles_per_feed
self.feeds = self.tech_feeds
tech = self.parse_feeds()
self.oldest_article = save_oldest_article
self.max_articles_per_feed = save_max_articles_per_feed
self.feeds = None
for f in tech:
key_list.append(f.title)
tech_articles[f.title] = []
for a in f.articles:
tech_articles[f.title].append(
dict(title=a.title, url=a.url, date=a.date,
description=a.summary, author=a.author,
content=a.content))
tech_ans = [(k, tech_articles[k]) for k in key_list if tech_articles.has_key(k)]
for x in tech_ans:
ans.append(x)
return ans
def parse_web_edition(self):
@ -457,31 +525,41 @@ class NYTimes(BasicNewsRecipe):
if sec_title in self.excludeSections:
print "SECTION EXCLUDED: ",sec_title
continue
print 'Index URL: '+'http://www.nytimes.com/pages/'+index_url+'/index.html'
try:
soup = self.index_to_soup('http://www.nytimes.com/pages/'+index_url+'/index.html')
except:
continue
print 'Index URL: '+'http://www.nytimes.com/pages/'+index_url+'/index.html'
self.key = sec_title
# Find each article
for div in soup.findAll(True,
attrs={'class':['section-headline', 'story', 'story headline','sectionHeader','headlinesOnly multiline flush']}):
if div['class'] in ['story', 'story headline'] :
attrs={'class':['section-headline', 'ledeStory', 'story', 'story headline','sectionHeader','headlinesOnly multiline flush']}):
if div['class'] in ['story', 'story headline', 'storyHeader'] :
self.handle_article(div)
elif div['class'] == 'ledeStory':
divsub = div.find('div','storyHeader')
if divsub is not None:
self.handle_article(divsub)
ulrefer = div.find('ul','refer')
if ulrefer is not None:
for lidiv in ulrefer.findAll('li'):
self.handle_article(lidiv)
elif div['class'] == 'headlinesOnly multiline flush':
for lidiv in div.findAll('li'):
self.handle_article(lidiv)
self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
return self.filter_ans(self.ans)
return self.filter_ans(self.get_tech_feeds(self.ans))
def parse_todays_index(self):
soup = self.index_to_soup('http://www.nytimes.com/pages/todayspaper/index.html')
skipping = False
# Find each article
for div in soup.findAll(True,
attrs={'class':['section-headline', 'story', 'story headline','sectionHeader','headlinesOnly multiline flush']}):
if div['class'] in ['section-headline','sectionHeader']:
self.key = string.capwords(self.feed_title(div))
self.key = self.key.replace('Op-ed','Op-Ed')
@ -505,7 +583,7 @@ class NYTimes(BasicNewsRecipe):
self.handle_article(lidiv)
self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
return self.filter_ans(self.ans)
return self.filter_ans(self.get_tech_feeds(self.ans))
def parse_headline_index(self):
@ -553,7 +631,7 @@ class NYTimes(BasicNewsRecipe):
for h3_item in search_div.findAll('h3'):
byline = h3_item.h6
if byline is not None:
author = self.tag_to_string(byline,usa_alt=False)
author = self.tag_to_string(byline,use_alt=False)
else:
author = ''
a = h3_item.find('a', href=True)
@ -579,7 +657,7 @@ class NYTimes(BasicNewsRecipe):
self.articles[section_name].append(dict(title=title, url=url, date=pubdate, description=description, author=author, content=''))
self.ans = [(k, self.articles[k]) for k in self.ans if self.articles.has_key(k)]
return self.filter_ans(self.ans)
return self.filter_ans(self.get_tech_feeds(self.ans))
def parse_index(self):
if self.headlinesOnly:
@ -589,40 +667,198 @@ class NYTimes(BasicNewsRecipe):
else:
return self.parse_todays_index()
def strip_anchors(self,soup):
def strip_anchors(self,soup,kill_all=False):
paras = soup.findAll(True)
for para in paras:
aTags = para.findAll('a')
for a in aTags:
if a.img is None:
a.replaceWith(a.renderContents().decode('cp1252','replace'))
if kill_all or (self.recursions==0):
a.replaceWith(self.tag_to_string(a,False))
else:
if a.has_key('href'):
if a['href'].startswith('http://www.nytimes'):
if not a['href'].endswith('pagewanted=all'):
url = re.sub(r'\?.*', '', a['href'])
if self.exclude_url(url):
a.replaceWith(self.tag_to_string(a,False))
else:
a['href'] = url+'?pagewanted=all'
elif not (a['href'].startswith('http://pogue') or \
a['href'].startswith('http://bits') or \
a['href'].startswith('http://travel') or \
a['href'].startswith('http://business') or \
a['href'].startswith('http://tech') or \
a['href'].startswith('http://health') or \
a['href'].startswith('http://dealbook') or \
a['href'].startswith('http://open')):
a.replaceWith(self.tag_to_string(a,False))
return soup
def handle_tags(self,soup):
try:
print("HANDLE TAGS: TITLE = "+self.tag_to_string(soup.title))
except:
print("HANDLE TAGS: NO TITLE")
if soup is None:
print("ERROR: handle_tags received NoneType")
return None
## print("HANDLING AD FORWARD:")
## print(soup)
if self.keep_only_tags:
body = Tag(soup, 'body')
try:
if isinstance(self.keep_only_tags, dict):
self.keep_only_tags = [self.keep_only_tags]
for spec in self.keep_only_tags:
for tag in soup.find('body').findAll(**spec):
body.insert(len(body.contents), tag)
soup.find('body').replaceWith(body)
except AttributeError: # soup has no body element
pass
def remove_beyond(tag, next):
while tag is not None and getattr(tag, 'name', None) != 'body':
after = getattr(tag, next)
while after is not None:
ns = getattr(tag, next)
after.extract()
after = ns
tag = tag.parent
if self.remove_tags_after is not None:
rt = [self.remove_tags_after] if isinstance(self.remove_tags_after, dict) else self.remove_tags_after
for spec in rt:
tag = soup.find(**spec)
remove_beyond(tag, 'nextSibling')
if self.remove_tags_before is not None:
tag = soup.find(**self.remove_tags_before)
remove_beyond(tag, 'previousSibling')
for kwds in self.remove_tags:
for tag in soup.findAll(**kwds):
tag.extract()
return soup
def preprocess_html(self, soup):
if self.webEdition & (self.oldest_article>0):
date_tag = soup.find(True,attrs={'class': ['dateline','date']})
if date_tag:
date_str = self.tag_to_string(date_tag,use_alt=False)
date_str = date_str.replace('Published:','')
date_items = date_str.split(',')
try:
datestring = date_items[0]+' '+date_items[1]
article_date = self.decode_us_date(datestring)
except:
article_date = date.today()
if article_date < self.earliest_date:
self.log("Skipping article dated %s" % date_str)
return None
print("PREPROCESS TITLE="+self.tag_to_string(soup.title))
skip_tag = soup.find(True, {'name':'skip'})
if skip_tag is not None:
url = 'http://www.nytimes.com' + re.sub(r'\?.*', '', skip_tag.parent['href'])
url += '?pagewanted=all'
self.log.warn("Skipping ad to article at '%s'" % url)
sleep(5)
soup = self.handle_tags(self.article_to_soup(url))
#all articles are from today, no need to print the date on every page
try:
if not self.webEdition:
date_tag = soup.find(True,attrs={'class': ['dateline','date']})
if date_tag:
date_tag.extract()
except:
self.log("Error removing the published date")
# check if the article is from one of the tech blogs
blog=soup.find('div',attrs={'id':['pogue','bits','gadgetwise','open']})
if blog is not None:
old_body = soup.find('body')
new_body=Tag(soup,'body')
new_body.append(soup.find('div',attrs={'id':'content'}))
new_body.find('div',attrs={'id':'content'})['id']='blogcontent' # identify for postprocess_html
old_body.replaceWith(new_body)
for divr in soup.findAll('div',attrs={'class':re.compile('w190 right')}):
if divr.find(text=re.compile('Sign up')):
divr.extract()
divr = soup.find('div',attrs={'id':re.compile('related-content')})
if divr is not None:
# handle related articles
rlist = []
ul = divr.find('ul')
if ul is not None:
for li in ul.findAll('li'):
atag = li.find('a')
if atag is not None:
if atag['href'].startswith('http://pogue') or atag['href'].startswith('http://bits') or \
atag['href'].startswith('http://open'):
atag.find(text=True).replaceWith(self.massageNCXText(self.tag_to_string(atag,False)))
rlist.append(atag)
divr.extract()
if rlist != []:
asidediv = Tag(soup,'div',[('class','aside')])
if soup.find('hr') is None:
asidediv.append(Tag(soup,'hr'))
h4 = Tag(soup,'h4',[('class','asidenote')])
h4.insert(0,"Related Posts")
asidediv.append(h4)
ul = Tag(soup,'ul')
for r in rlist:
li = Tag(soup,'li',[('class','aside')])
r['class'] = 'aside'
li.append(r)
ul.append(li)
asidediv.append(ul)
asidediv.append(Tag(soup,'hr'))
smain = soup.find('body')
smain.append(asidediv)
for atag in soup.findAll('a'):
img = atag.find('img')
if img is not None:
atag.replaceWith(img)
elif not atag.has_key('href'):
atag.replaceWith(atag.renderContents().decode('cp1252','replace'))
elif not (atag['href'].startswith('http://www.nytimes') or atag['href'].startswith('http://pogue') or \
atag['href'].startswith('http://bits') or atag['href'].startswith('http://open')):
atag.replaceWith(atag.renderContents().decode('cp1252','replace'))
hdr = soup.find('address')
if hdr is not None:
hdr.name='span'
for span_credit in soup.findAll('span','credit'):
sp = Tag(soup,'span')
span_credit.replaceWith(sp)
sp.append(Tag(soup,'br'))
sp.append(span_credit)
sp.append(Tag(soup,'br'))
else: # nytimes article
related = [] # these will be the related articles
first_outer = None # first related outer tag
first_related = None # first related tag
for outerdiv in soup.findAll(attrs={'class': re.compile('articleInline runaroundLeft')}):
for rdiv in soup.findAll('div','columnGroup doubleRule'):
if rdiv.find('h3') is not None:
if self.tag_to_string(rdiv.h3,False).startswith('Related'):
rdiv.h3.find(text=True).replaceWith("Related articles")
rdiv.h3['class'] = 'asidenote'
for litag in rdiv.findAll('li'):
if litag.find('a') is not None:
if litag.find('a')['href'].startswith('http://www.nytimes.com'):
url = re.sub(r'\?.*', '', litag.find('a')['href'])
litag.find('a')['href'] = url+'?pagewanted=all'
litag.extract()
related.append(litag)
if first_related is None:
first_related = rdiv
first_outer = outerdiv
else:
litag.extract()
if related != []:
for r in related:
if r.h6: # don't want the anchor inside a h6 tag
r.h6.replaceWith(r.h6.a)
first_related.ul.append(r)
first_related.insert(0,Tag(soup,'hr'))
first_related.append(Tag(soup,'hr'))
first_related['class'] = 'aside'
first_outer.replaceWith(first_related) # replace the outer tag with the related tag
for rdiv in soup.findAll(attrs={'class': re.compile('articleInline runaroundLeft')}):
rdiv.extract()
kicker_tag = soup.find(attrs={'class':'kicker'})
if kicker_tag: # remove Op_Ed author head shots
tagline = self.tag_to_string(kicker_tag)
if tagline=='Op-Ed Columnist':
img_div = soup.find('div','inlineImage module')
if img_div:
img_div.extract()
if self.useHighResImages:
try:
@ -667,26 +903,6 @@ class NYTimes(BasicNewsRecipe):
except Exception:
self.log("Error pulling high resolution images")
try:
#remove "Related content" bar
runAroundsFound = soup.findAll('div',{'class':['articleInline runaroundLeft','articleInline doubleRule runaroundLeft','articleInline runaroundLeft firstArticleInline','articleInline runaroundLeft ','articleInline runaroundLeft lastArticleInline']})
if runAroundsFound:
for runAround in runAroundsFound:
#find all section headers
hlines = runAround.findAll(True ,{'class':['sectionHeader','sectionHeader flushBottom']})
if hlines:
for hline in hlines:
hline.extract()
#find all section headers
hlines = runAround.findAll('h6')
if hlines:
for hline in hlines:
hline.extract()
except:
self.log("Error removing related content bar")
try:
#in case pulling images failed, delete the enlarge this text
enlargeThisList = soup.findAll('div',{'class':'icon enlargeThis'})
@ -696,9 +912,24 @@ class NYTimes(BasicNewsRecipe):
except:
self.log("Error removing Enlarge this text")
return self.strip_anchors(soup)
def postprocess_html(self,soup, True):
return self.strip_anchors(soup,False)
def postprocess_html(self,soup,first_fetch):
if not first_fetch: # remove Related links
for aside in soup.findAll('div','aside'):
aside.extract()
soup = self.strip_anchors(soup,True)
if soup.find('div',attrs={'id':'blogcontent'}) is None:
if first_fetch:
aside = soup.find('div','aside')
if aside is not None: # move the related list to the end of the article
art = soup.find('div',attrs={'id':'article'})
if art is None:
art = soup.find('div',attrs={'class':'article'})
if art is not None:
art.append(aside)
try:
if self.one_picture_per_article:
# Remove all images after first
@ -855,23 +1086,22 @@ class NYTimes(BasicNewsRecipe):
self.log("ERROR: Problem in Add class=authorId to <div> so we can format with CSS")
return soup
def populate_article_metadata(self, article, soup, first):
if first and hasattr(self, 'add_toc_thumbnail'):
if not first:
return
idxdiv = soup.find('div',attrs={'class':'articleSpanImage'})
if idxdiv is not None:
if idxdiv.img:
self.add_toc_thumbnail(article, idxdiv.img['src'])
self.add_toc_thumbnail(article, re.sub(r'links\\link\d+\\','',idxdiv.img['src']))
else:
img = soup.find('img')
img = soup.find('body').find('img')
if img is not None:
self.add_toc_thumbnail(article, img['src'])
self.add_toc_thumbnail(article, re.sub(r'links\\link\d+\\','',img['src']))
shortparagraph = ""
try:
if len(article.text_summary.strip()) == 0:
articlebodies = soup.findAll('div',attrs={'class':'articleBody'})
if not articlebodies: #added to account for blog formats
articlebodies = soup.findAll('div', attrs={'class':'entry-content'}) #added to account for blog formats
if articlebodies:
for articlebody in articlebodies:
if articlebody:
@ -880,15 +1110,23 @@ class NYTimes(BasicNewsRecipe):
refparagraph = self.massageNCXText(self.tag_to_string(p,use_alt=False)).strip()
#account for blank paragraphs and short paragraphs by appending them to longer ones
if len(refparagraph) > 0:
if len(refparagraph) > 140: #approximately two lines of text
article.summary = article.text_summary = shortparagraph + refparagraph
if len(refparagraph) > 70: #approximately one line of text
newpara = shortparagraph + refparagraph
newparaDateline,newparaEm,newparaDesc = newpara.partition('&mdash;')
if newparaEm == '':
newparaDateline,newparaEm,newparaDesc = newpara.partition('—')
if newparaEm == '':
newparaDesc = newparaDateline
article.summary = article.text_summary = newparaDesc.strip()
return
else:
shortparagraph = refparagraph + " "
if shortparagraph.strip().find(" ") == -1 and not shortparagraph.strip().endswith(":"):
shortparagraph = shortparagraph + "- "
else:
article.summary = article.text_summary = self.massageNCXText(article.text_summary)
except:
self.log("Error creating article descriptions")
return

22
recipes/sueddeutsche.recipe
View File

@ -8,19 +8,19 @@ Fetch sueddeutsche.de
from calibre.web.feeds.news import BasicNewsRecipe
class Sueddeutsche(BasicNewsRecipe):
title = u'Süddeutsche.de' # 2012-01-26 AGe Correct Title
description = 'News from Germany, Access to online content' # 2012-01-26 AGe
__author__ = 'Oliver Niesner and Armin Geller' #Update AGe 2012-01-26
publisher = u'Süddeutsche Zeitung' # 2012-01-26 AGe add
category = 'news, politics, Germany' # 2012-01-26 AGe add
timefmt = ' [%a, %d %b %Y]' # 2012-01-26 AGe add %a
title = u'Süddeutsche.de'
description = 'News from Germany, Access to online content'
__author__ = 'Oliver Niesner and Armin Geller' #Update AGe 2012-12-05
publisher = u'Süddeutsche Zeitung'
category = 'news, politics, Germany'
timefmt = ' [%a, %d %b %Y]'
oldest_article = 7
max_articles_per_feed = 100
language = 'de'
encoding = 'utf-8'
publication_type = 'newspaper' # 2012-01-26 add
publication_type = 'newspaper'
cover_source = 'http://www.sueddeutsche.de/verlag' # 2012-01-26 AGe add from Darko Miletic paid content source
masthead_url = 'http://www.sueddeutsche.de/static_assets/build/img/sdesiteheader/logo_homepage.441d531c.png' # 2012-01-26 AGe add
masthead_url = 'http://www.sueddeutsche.de/static_assets/img/sdesiteheader/logo_standard.a152b0df.png' # 2012-12-05 AGe add
use_embedded_content = False
no_stylesheets = True
@ -40,9 +40,9 @@ class Sueddeutsche(BasicNewsRecipe):
(u'Sport', u'http://suche.sueddeutsche.de/query/%23/sort/-docdatetime/drilldown/%C2%A7ressort%3A%5ESport%24?output=rss'),
(u'Leben', u'http://suche.sueddeutsche.de/query/%23/sort/-docdatetime/drilldown/%C2%A7ressort%3A%5ELeben%24?output=rss'),
(u'Karriere', u'http://suche.sueddeutsche.de/query/%23/sort/-docdatetime/drilldown/%C2%A7ressort%3A%5EKarriere%24?output=rss'),
(u'Bildung', u'http://rss.sueddeutsche.de/rss/bildung'), #2012-01-26 AGe New
(u'Gesundheit', u'http://rss.sueddeutsche.de/rss/gesundheit'), #2012-01-26 AGe New
(u'Stil', u'http://rss.sueddeutsche.de/rss/stil'), #2012-01-26 AGe New
(u'Bildung', u'http://rss.sueddeutsche.de/rss/bildung'),
(u'Gesundheit', u'http://rss.sueddeutsche.de/rss/gesundheit'),
(u'Stil', u'http://rss.sueddeutsche.de/rss/stil'),
(u'München & Region', u'http://suche.sueddeutsche.de/query/%23/sort/-docdatetime/drilldown/%C2%A7ressort%3A%5EMünchen&Region%24?output=rss'),
(u'Bayern', u'http://suche.sueddeutsche.de/query/%23/sort/-docdatetime/drilldown/%C2%A7ressort%3A%5EBayern%24?output=rss'),
(u'Medien', u'http://suche.sueddeutsche.de/query/%23/sort/-docdatetime/drilldown/%C2%A7ressort%3A%5EMedien%24?output=rss'),

29
recipes/weblogs_sl.recipe
View File

@ -2,8 +2,8 @@
__license__ = 'GPL v3'
__copyright__ = '4 February 2011, desUBIKado'
__author__ = 'desUBIKado'
__version__ = 'v0.08'
__date__ = '30, June 2012'
__version__ = 'v0.09'
__date__ = '02, December 2012'
'''
http://www.weblogssl.com/
'''
@ -37,6 +37,7 @@ class weblogssl(BasicNewsRecipe):
,(u'Xataka Mexico', u'http://feeds.weblogssl.com/xatakamx')
,(u'Xataka M\xf3vil', u'http://feeds.weblogssl.com/xatakamovil')
,(u'Xataka Android', u'http://feeds.weblogssl.com/xatakandroid')
,(u'Xataka Windows', u'http://feeds.weblogssl.com/xatakawindows')
,(u'Xataka Foto', u'http://feeds.weblogssl.com/xatakafoto')
,(u'Xataka ON', u'http://feeds.weblogssl.com/xatakaon')
,(u'Xataka Ciencia', u'http://feeds.weblogssl.com/xatakaciencia')
@ -80,19 +81,31 @@ class weblogssl(BasicNewsRecipe):
keep_only_tags = [dict(name='div', attrs={'id':'infoblock'}),
dict(name='div', attrs={'class':'post'}),
dict(name='div', attrs={'id':'blog-comments'})
dict(name='div', attrs={'id':'blog-comments'}),
dict(name='div', attrs={'class':'container'}) #m.xataka.com
]
remove_tags = [dict(name='div', attrs={'id':'comment-nav'})]
remove_tags = [dict(name='div', attrs={'id':'comment-nav'}),
dict(name='menu', attrs={'class':'social-sharing'}), #m.xataka.com
dict(name='section' , attrs={'class':'comments'}), #m.xataka.com
dict(name='div' , attrs={'class':'article-comments'}), #m.xataka.com
dict(name='nav' , attrs={'class':'article-taxonomy'}) #m.xataka.com
]
remove_tags_after = dict(name='section' , attrs={'class':'comments'})
def print_version(self, url):
return url.replace('http://www.', 'http://m.')
preprocess_regexps = [
# Para poner una linea en blanco entre un comentario y el siguiente
(re.compile(r'<li id="c', re.DOTALL|re.IGNORECASE), lambda match: '<br><br><li id="c')
(re.compile(r'<li id="c', re.DOTALL|re.IGNORECASE), lambda match: '<br><br><li id="c'),
# Para ver las imágenes en las noticias de m.xataka.com
(re.compile(r'<noscript>', re.DOTALL|re.IGNORECASE), lambda m: ''),
(re.compile(r'</noscript>', re.DOTALL|re.IGNORECASE), lambda m: '')
]
# Para sustituir el video incrustado de YouTube por una imagen
def preprocess_html(self, soup):
@ -108,14 +121,16 @@ class weblogssl(BasicNewsRecipe):
# Para obtener la url original del articulo a partir de la de "feedsportal"
# El siguiente código es gracias al usuario "bosplans" de www.mobileread.com
# http://www.mobileread.com/forums/sho...d.php?t=130297
# http://www.mobileread.com/forums/showthread.php?t=130297
def get_article_url(self, article):
link = article.get('link', None)
if link is None:
return article
# if link.split('/')[-4]=="xataka2":
# return article.get('feedburner_origlink', article.get('link', article.get('guid')))
if link.split('/')[-4]=="xataka2":
return article.get('feedburner_origlink', article.get('link', article.get('guid')))
return article.get('guid', None)
if link.split('/')[-1]=="story01.htm":
link=link.split('/')[-2]
a=['0B','0C','0D','0E','0F','0G','0N' ,'0L0S','0A']

30
recipes/zaman.recipe
View File

@ -9,15 +9,15 @@ class Zaman (BasicNewsRecipe):
__author__ = u'thomass'
oldest_article = 2
max_articles_per_feed =50
# no_stylesheets = True
no_stylesheets = True
#delay = 1
#use_embedded_content = False
encoding = 'ISO 8859-9'
publisher = 'Zaman'
use_embedded_content = False
encoding = 'utf-8'
publisher = 'Feza Gazetecilik'
category = 'news, haberler,TR,gazete'
language = 'tr'
publication_type = 'newspaper '
extra_css = '.buyukbaslik{font-weight: bold; font-size: 18px;color:#0000FF}'#body{ font-family: Verdana,Helvetica,Arial,sans-serif } .introduction{font-weight: bold} .story-feature{display: block; padding: 0; border: 1px solid; width: 40%; font-size: small} .story-feature h2{text-align: center; text-transform: uppercase} '
extra_css = 'h1{text-transform: capitalize; font-weight: bold; font-size: 22px;color:#0000FF} p{text-align:justify} ' #.introduction{font-weight: bold} .story-feature{display: block; padding: 0; border: 1px solid; width: 40%; font-size: small} .story-feature h2{text-align: center; text-transform: uppercase} '
conversion_options = {
'tags' : category
,'language' : language
@ -26,25 +26,26 @@ class Zaman (BasicNewsRecipe):
}
cover_img_url = 'https://fbcdn-profile-a.akamaihd.net/hprofile-ak-snc4/188140_81722291869_2111820_n.jpg'
masthead_url = 'http://medya.zaman.com.tr/extentions/zaman.com.tr/img/section/logo-section.png'
ignore_duplicate_articles = { 'title', 'url' }
auto_cleanup = False
remove_empty_feeds= True
#keep_only_tags = [dict(name='div', attrs={'id':[ 'news-detail-content']}), dict(name='td', attrs={'class':['columnist-detail','columnist_head']}) ]
remove_tags = [ dict(name='img', attrs={'src':['http://medya.zaman.com.tr/zamantryeni/pics/zamanonline.gif']})]#,dict(name='div', attrs={'class':['radioEmbedBg','radyoProgramAdi']}),dict(name='a', attrs={'class':['webkit-html-attribute-value webkit-html-external-link']}),dict(name='table', attrs={'id':['yaziYorumTablosu']}),dict(name='img', attrs={'src':['http://medya.zaman.com.tr/pics/paylas.gif','http://medya.zaman.com.tr/extentions/zaman.com.tr/img/columnist/ma-16.png']})
#keep_only_tags = [dict(name='div', attrs={'id':[ 'contentposition19']})]#,dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'news-detail-content']}), dict(name='td', attrs={'class':['columnist-detail','columnist_head']}), ]
remove_tags = [ dict(name='img', attrs={'src':['http://cmsmedya.zaman.com.tr/images/logo/logo.bmp']}),dict(name='hr', attrs={'class':['interactive-hr']})]# remove_tags = [ dict(name='div', attrs={'class':[ 'detayUyari']}),dict(name='div', attrs={'class':[ 'detayYorum']}),dict(name='div', attrs={'class':[ 'addthis_toolbox addthis_default_style ']}),dict(name='div', attrs={'id':[ 'tumYazi']})]#,dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'xxx']}),dict(name='div', attrs={'id':[ 'xxx']}),dict(name='img', attrs={'src':['http://medya.zaman.com.tr/zamantryeni/pics/zamanonline.gif']}),dict(name='div', attrs={'class':['radioEmbedBg','radyoProgramAdi']}),dict(name='a', attrs={'class':['webkit-html-attribute-value webkit-html-external-link']}),dict(name='table', attrs={'id':['yaziYorumTablosu']}),dict(name='img', attrs={'src':['http://medya.zaman.com.tr/pics/paylas.gif','http://medya.zaman.com.tr/extentions/zaman.com.tr/img/columnist/ma-16.png']}),dict(name='div', attrs={'id':[ 'news-detail-gallery']}),dict(name='div', attrs={'id':[ 'news-detail-title-bottom-part']}),dict(name='div', attrs={'id':[ 'news-detail-news-paging-main']})]#
#remove_attributes = ['width','height']
remove_empty_feeds= True
feeds = [
( u'Anasayfa', u'http://www.zaman.com.tr/anasayfa.rss'),
( u'Son Dakika', u'http://www.zaman.com.tr/sondakika.rss'),
#( u'En çok Okunanlar', u'http://www.zaman.com.tr/max_all.rss'),
#( u'Manşet', u'http://www.zaman.com.tr/manset.rss'),
( u'Gündem', u'http://www.zaman.com.tr/gundem.rss'),
( u'Manşet', u'http://www.zaman.com.tr/manset.rss'),
( u'Yazarlar', u'http://www.zaman.com.tr/yazarlar.rss'),
( u'Politika', u'http://www.zaman.com.tr/politika.rss'),
( u'Ekonomi', u'http://www.zaman.com.tr/ekonomi.rss'),
( u'Dış Haberler', u'http://www.zaman.com.tr/dishaberler.rss'),
( u'Son Dakika', u'http://www.zaman.com.tr/sondakika.rss'),
( u'Gündem', u'http://www.zaman.com.tr/gundem.rss'),
( u'Yorumlar', u'http://www.zaman.com.tr/yorumlar.rss'),
( u'Röportaj', u'http://www.zaman.com.tr/roportaj.rss'),
( u'Dizi Yazı', u'http://www.zaman.com.tr/dizi.rss'),
@ -59,8 +60,9 @@ class Zaman (BasicNewsRecipe):
( u'Cuma Eki', u'http://www.zaman.com.tr/cuma.rss'),
( u'Cumaertesi Eki', u'http://www.zaman.com.tr/cumaertesi.rss'),
( u'Pazar Eki', u'http://www.zaman.com.tr/pazar.rss'),
( u'En çok Okunanlar', u'http://www.zaman.com.tr/max_all.rss'),
( u'Anasayfa', u'http://www.zaman.com.tr/anasayfa.rss'),
]
def print_version(self, url):
return url.replace('http://www.zaman.com.tr/haber.do?haberno=', 'http://www.zaman.com.tr/yazdir.do?haberno=')
return url.replace('http://www.zaman.com.tr/newsDetail_getNewsById.action?newsId=', 'http://www.zaman.com.tr/newsDetail_openPrintPage.action?newsId=')

2
setup/__init__.py
View File

@ -215,6 +215,8 @@ class Command(object):
sys.stdout.flush()
def installer_name(ext, is64bit=False):
if is64bit and ext == 'msi':
return 'dist/%s-64bit-%s.msi'%(__appname__, __version__)
if ext in ('exe', 'msi'):
return 'dist/%s-%s.%s'%(__appname__, __version__, ext)
if ext == 'dmg':

3
setup/build_environment.py
View File

@ -11,12 +11,11 @@ from distutils.spawn import find_executable
from PyQt4 import pyqtconfig
from setup import isosx, iswindows, islinux
from setup import isosx, iswindows, islinux, is64bit
OSX_SDK = '/Developer/SDKs/MacOSX10.5.sdk'
os.environ['MACOSX_DEPLOYMENT_TARGET'] = '10.5'
is64bit = sys.maxsize > 2**32
NMAKE = RC = msvc = MT = win_inc = win_lib = win_ddk = win_ddk_lib_dirs = None
if iswindows:

5
setup/commands.py
View File

@ -20,7 +20,7 @@ __all__ = [
'upload_user_manual', 'upload_demo', 'reupload',
'linux32', 'linux64', 'linux', 'linux_freeze',
'osx32_freeze', 'osx', 'rsync', 'push',
'win32_freeze', 'win32', 'win',
'win32_freeze', 'win32', 'win64', 'win',
'stage1', 'stage2', 'stage3', 'stage4', 'stage5', 'publish'
]
@ -91,9 +91,10 @@ osx = OSX()
from setup.installer.osx.app.main import OSX32_Freeze
osx32_freeze = OSX32_Freeze()
from setup.installer.windows import Win, Win32
from setup.installer.windows import Win, Win32, Win64
win = Win()
win32 = Win32()
win64 = Win64()
from setup.installer.windows.freeze import Win32Freeze
win32_freeze = Win32Freeze()

459
setup/installer/windows/MemoryModule.c
View File

@ -1,12 +1,12 @@
/*
* Memory DLL loading code
* Version 0.0.2 with additions from Thomas Heller
* Version 0.0.3
*
* Copyright (c) 2004-2005 by Joachim Bauch / mail@joachim-bauch.de
* Copyright (c) 2004-2012 by Joachim Bauch / mail@joachim-bauch.de
* http://www.joachim-bauch.de
*
* The contents of this file are subject to the Mozilla Public License Version
* 1.1 (the "License"); you may not use this file except in compliance with
* 2.0 (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
* http://www.mozilla.org/MPL/
*
@ -19,156 +19,48 @@
*
* The Initial Developer of the Original Code is Joachim Bauch.
*
* Portions created by Joachim Bauch are Copyright (C) 2004-2005
* Portions created by Joachim Bauch are Copyright (C) 2004-2012
* Joachim Bauch. All Rights Reserved.
*
* Portions Copyright (C) 2005 Thomas Heller.
*
*/
#ifndef __GNUC__
// disable warnings about pointer <-> DWORD conversions
#pragma warning( disable : 4311 4312 )
#endif
#ifdef _WIN64
#define POINTER_TYPE ULONGLONG
#else
#define POINTER_TYPE DWORD
#endif
#include <Windows.h>
#include <winnt.h>
#if DEBUG_OUTPUT
#ifdef DEBUG_OUTPUT
#include <stdio.h>
#endif
#ifndef IMAGE_SIZEOF_BASE_RELOCATION
// Vista SDKs no longer define IMAGE_SIZEOF_BASE_RELOCATION!?
# define IMAGE_SIZEOF_BASE_RELOCATION (sizeof(IMAGE_BASE_RELOCATION))
#define IMAGE_SIZEOF_BASE_RELOCATION (sizeof(IMAGE_BASE_RELOCATION))
#endif
#include "MemoryModule.h"
/*
XXX We need to protect at least walking the 'loaded' linked list with a lock!
*/
/******************************************************************/
FINDPROC findproc;
void *findproc_data = NULL;
struct NAME_TABLE {
char *name;
DWORD ordinal;
};
typedef struct tagMEMORYMODULE {
typedef struct {
PIMAGE_NT_HEADERS headers;
unsigned char *codeBase;
HMODULE *modules;
int numModules;
int initialized;
struct NAME_TABLE *name_table;
char *name;
int refcount;
struct tagMEMORYMODULE *next, *prev;
} MEMORYMODULE, *PMEMORYMODULE;
typedef BOOL (WINAPI *DllEntryProc)(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID lpReserved);
#define GET_HEADER_DICTIONARY(module, idx) &(module)->headers->OptionalHeader.DataDirectory[idx]
MEMORYMODULE *loaded; /* linked list of loaded memory modules */
/* private - insert a loaded library in a linked list */
static void _Register(char *name, MEMORYMODULE *module)
{
module->next = loaded;
if (loaded)
loaded->prev = module;
module->prev = NULL;
loaded = module;
}
/* private - remove a loaded library from a linked list */
static void _Unregister(MEMORYMODULE *module)
{
free(module->name);
if (module->prev)
module->prev->next = module->next;
if (module->next)
module->next->prev = module->prev;
if (module == loaded)
loaded = module->next;
}
/* public - replacement for GetModuleHandle() */
HMODULE MyGetModuleHandle(LPCTSTR lpModuleName)
{
MEMORYMODULE *p = loaded;
while (p) {
// If already loaded, only increment the reference count
if (0 == stricmp(lpModuleName, p->name)) {
return (HMODULE)p;
}
p = p->next;
}
return GetModuleHandle(lpModuleName);
}
/* public - replacement for LoadLibrary, but searches FIRST for memory
libraries, then for normal libraries. So, it will load libraries AS memory
module if they are found by findproc().
*/
HMODULE MyLoadLibrary(char *lpFileName)
{
MEMORYMODULE *p = loaded;
HMODULE hMod;
while (p) {
// If already loaded, only increment the reference count
if (0 == stricmp(lpFileName, p->name)) {
p->refcount++;
return (HMODULE)p;
}
p = p->next;
}
if (findproc && findproc_data) {
void *pdata = findproc(lpFileName, findproc_data);
if (pdata) {
hMod = MemoryLoadLibrary(lpFileName, pdata);
free(p);
return hMod;
}
}
hMod = LoadLibrary(lpFileName);
return hMod;
}
/* public - replacement for GetProcAddress() */
FARPROC MyGetProcAddress(HMODULE hModule, LPCSTR lpProcName)
{
MEMORYMODULE *p = loaded;
while (p) {
if ((HMODULE)p == hModule)
return MemoryGetProcAddress(p, lpProcName);
p = p->next;
}
return GetProcAddress(hModule, lpProcName);
}
/* public - replacement for FreeLibrary() */
BOOL MyFreeLibrary(HMODULE hModule)
{
MEMORYMODULE *p = loaded;
while (p) {
if ((HMODULE)p == hModule) {
if (--p->refcount == 0) {
_Unregister(p);
MemoryFreeLibrary(p);
}
return TRUE;
}
p = p->next;
}
return FreeLibrary(hModule);
}
#if DEBUG_OUTPUT
#ifdef DEBUG_OUTPUT
static void
OutputLastError(const char *msg)
{
@ -184,20 +76,6 @@ OutputLastError(const char *msg)
}
#endif
/*
static int dprintf(char *fmt, ...)
{
char Buffer[4096];
va_list marker;
int result;
va_start(marker, fmt);
result = vsprintf(Buffer, fmt, marker);
OutputDebugString(Buffer);
return result;
}
*/
static void
CopySections(const unsigned char *data, PIMAGE_NT_HEADERS old_headers, PMEMORYMODULE module)
{
@ -205,15 +83,12 @@ CopySections(const unsigned char *data, PIMAGE_NT_HEADERS old_headers, PMEMORYMO
unsigned char *codeBase = module->codeBase;
unsigned char *dest;
PIMAGE_SECTION_HEADER section = IMAGE_FIRST_SECTION(module->headers);
for (i=0; i<module->headers->FileHeader.NumberOfSections; i++, section++)
{
if (section->SizeOfRawData == 0)
{
for (i=0; i<module->headers->FileHeader.NumberOfSections; i++, section++) {
if (section->SizeOfRawData == 0) {
// section doesn't contain data in the dll itself, but may define
// uninitialized data
size = old_headers->OptionalHeader.SectionAlignment;
if (size > 0)
{
if (size > 0) {
dest = (unsigned char *)VirtualAlloc(codeBase + section->VirtualAddress,
size,
MEM_COMMIT,
@ -255,66 +130,72 @@ FinalizeSections(PMEMORYMODULE module)
{
int i;
PIMAGE_SECTION_HEADER section = IMAGE_FIRST_SECTION(module->headers);
#ifdef _WIN64
POINTER_TYPE imageOffset = (module->headers->OptionalHeader.ImageBase & 0xffffffff00000000);
#else
#define imageOffset 0
#endif
// loop through all sections and change access flags
for (i=0; i<module->headers->FileHeader.NumberOfSections; i++, section++)
{
for (i=0; i<module->headers->FileHeader.NumberOfSections; i++, section++) {
DWORD protect, oldProtect, size;
int executable = (section->Characteristics & IMAGE_SCN_MEM_EXECUTE) != 0;
int readable = (section->Characteristics & IMAGE_SCN_MEM_READ) != 0;
int writeable = (section->Characteristics & IMAGE_SCN_MEM_WRITE) != 0;
if (section->Characteristics & IMAGE_SCN_MEM_DISCARDABLE)
{
if (section->Characteristics & IMAGE_SCN_MEM_DISCARDABLE) {
// section is not needed any more and can safely be freed
VirtualFree((LPVOID)section->Misc.PhysicalAddress, section->SizeOfRawData, MEM_DECOMMIT);
VirtualFree((LPVOID)((POINTER_TYPE)section->Misc.PhysicalAddress | imageOffset), section->SizeOfRawData, MEM_DECOMMIT);
continue;
}
// determine protection flags based on characteristics
protect = ProtectionFlags[executable][readable][writeable];
if (section->Characteristics & IMAGE_SCN_MEM_NOT_CACHED)
if (section->Characteristics & IMAGE_SCN_MEM_NOT_CACHED) {
protect |= PAGE_NOCACHE;
}
// determine size of region
size = section->SizeOfRawData;
if (size == 0)
{
if (section->Characteristics & IMAGE_SCN_CNT_INITIALIZED_DATA)
if (size == 0) {
if (section->Characteristics & IMAGE_SCN_CNT_INITIALIZED_DATA) {
size = module->headers->OptionalHeader.SizeOfInitializedData;
else if (section->Characteristics & IMAGE_SCN_CNT_UNINITIALIZED_DATA)
} else if (section->Characteristics & IMAGE_SCN_CNT_UNINITIALIZED_DATA) {
size = module->headers->OptionalHeader.SizeOfUninitializedData;
}
}
if (size > 0)
{
if (size > 0) {
// change memory access flags
if (VirtualProtect((LPVOID)section->Misc.PhysicalAddress, section->SizeOfRawData, protect, &oldProtect) == 0)
#if DEBUG_OUTPUT
if (VirtualProtect((LPVOID)((POINTER_TYPE)section->Misc.PhysicalAddress | imageOffset), size, protect, &oldProtect) == 0)
#ifdef DEBUG_OUTPUT
OutputLastError("Error protecting memory page")
#endif
;
}
}
#ifndef _WIN64
#undef imageOffset
#endif
}
static void
PerformBaseRelocation(PMEMORYMODULE module, DWORD delta)
PerformBaseRelocation(PMEMORYMODULE module, SIZE_T delta)
{
DWORD i;
unsigned char *codeBase = module->codeBase;
PIMAGE_DATA_DIRECTORY directory = GET_HEADER_DICTIONARY(module, IMAGE_DIRECTORY_ENTRY_BASERELOC);
if (directory->Size > 0)
{
PIMAGE_BASE_RELOCATION relocation = (PIMAGE_BASE_RELOCATION)(codeBase + directory->VirtualAddress);
for (; relocation->VirtualAddress > 0; )
{
unsigned char *dest = (unsigned char *)(codeBase + relocation->VirtualAddress);
if (directory->Size > 0) {
PIMAGE_BASE_RELOCATION relocation = (PIMAGE_BASE_RELOCATION) (codeBase + directory->VirtualAddress);
for (; relocation->VirtualAddress > 0; ) {
unsigned char *dest = codeBase + relocation->VirtualAddress;
unsigned short *relInfo = (unsigned short *)((unsigned char *)relocation + IMAGE_SIZEOF_BASE_RELOCATION);
for (i=0; i<((relocation->SizeOfBlock-IMAGE_SIZEOF_BASE_RELOCATION) / 2); i++, relInfo++)
{
for (i=0; i<((relocation->SizeOfBlock-IMAGE_SIZEOF_BASE_RELOCATION) / 2); i++, relInfo++) {
DWORD *patchAddrHL;
#ifdef _WIN64
ULONGLONG *patchAddr64;
#endif
int type, offset;
// the upper 4 bits define the type of relocation
@ -330,10 +211,17 @@ PerformBaseRelocation(PMEMORYMODULE module, DWORD delta)
case IMAGE_REL_BASED_HIGHLOW:
// change complete 32 bit address
patchAddrHL = (DWORD *)(dest + offset);
*patchAddrHL += delta;
patchAddrHL = (DWORD *) (dest + offset);
*patchAddrHL += (DWORD)delta;
break;
#ifdef _WIN64
case IMAGE_REL_BASED_DIR64:
patchAddr64 = (ULONGLONG *) (dest + offset);
*patchAddr64 += delta;
break;
#endif
default:
//printf("Unknown relocation: %d\n", type);
break;
@ -341,7 +229,7 @@ PerformBaseRelocation(PMEMORYMODULE module, DWORD delta)
}
// advance to next relocation block
relocation = (PIMAGE_BASE_RELOCATION)(((DWORD)relocation) + relocation->SizeOfBlock);
relocation = (PIMAGE_BASE_RELOCATION) (((char *) relocation) + relocation->SizeOfBlock);
}
}
}
@ -353,18 +241,13 @@ BuildImportTable(PMEMORYMODULE module)
unsigned char *codeBase = module->codeBase;
PIMAGE_DATA_DIRECTORY directory = GET_HEADER_DICTIONARY(module, IMAGE_DIRECTORY_ENTRY_IMPORT);
if (directory->Size > 0)
{
PIMAGE_IMPORT_DESCRIPTOR importDesc = (PIMAGE_IMPORT_DESCRIPTOR)(codeBase + directory->VirtualAddress);
for (; !IsBadReadPtr(importDesc, sizeof(IMAGE_IMPORT_DESCRIPTOR)) && importDesc->Name; importDesc++)
{
DWORD *thunkRef, *funcRef;
HMODULE handle;
handle = MyLoadLibrary(codeBase + importDesc->Name);
if (handle == INVALID_HANDLE_VALUE)
{
//LastError should already be set
if (directory->Size > 0) {
PIMAGE_IMPORT_DESCRIPTOR importDesc = (PIMAGE_IMPORT_DESCRIPTOR) (codeBase + directory->VirtualAddress);
for (; !IsBadReadPtr(importDesc, sizeof(IMAGE_IMPORT_DESCRIPTOR)) && importDesc->Name; importDesc++) {
POINTER_TYPE *thunkRef;
FARPROC *funcRef;
HMODULE handle = LoadLibrary((LPCSTR) (codeBase + importDesc->Name));
if (handle == NULL) {
#if DEBUG_OUTPUT
OutputLastError("Can't load library");
#endif
@ -373,81 +256,54 @@ BuildImportTable(PMEMORYMODULE module)
}
module->modules = (HMODULE *)realloc(module->modules, (module->numModules+1)*(sizeof(HMODULE)));
if (module->modules == NULL)
{
SetLastError(ERROR_NOT_ENOUGH_MEMORY);
if (module->modules == NULL) {
result = 0;
break;
}
module->modules[module->numModules++] = handle;
if (importDesc->OriginalFirstThunk)
{
thunkRef = (DWORD *)(codeBase + importDesc->OriginalFirstThunk);
funcRef = (DWORD *)(codeBase + importDesc->FirstThunk);
if (importDesc->OriginalFirstThunk) {
thunkRef = (POINTER_TYPE *) (codeBase + importDesc->OriginalFirstThunk);
funcRef = (FARPROC *) (codeBase + importDesc->FirstThunk);
} else {
// no hint table
thunkRef = (DWORD *)(codeBase + importDesc->FirstThunk);
funcRef = (DWORD *)(codeBase + importDesc->FirstThunk);
thunkRef = (POINTER_TYPE *) (codeBase + importDesc->FirstThunk);
funcRef = (FARPROC *) (codeBase + importDesc->FirstThunk);
}
for (; *thunkRef; thunkRef++, funcRef++)
{
if IMAGE_SNAP_BY_ORDINAL(*thunkRef) {
*funcRef = (DWORD)MyGetProcAddress(handle, (LPCSTR)IMAGE_ORDINAL(*thunkRef));
for (; *thunkRef; thunkRef++, funcRef++) {
if (IMAGE_SNAP_BY_ORDINAL(*thunkRef)) {
*funcRef = (FARPROC)GetProcAddress(handle, (LPCSTR)IMAGE_ORDINAL(*thunkRef));
} else {
PIMAGE_IMPORT_BY_NAME thunkData = (PIMAGE_IMPORT_BY_NAME)(codeBase + *thunkRef);
*funcRef = (DWORD)MyGetProcAddress(handle, (LPCSTR)&thunkData->Name);
PIMAGE_IMPORT_BY_NAME thunkData = (PIMAGE_IMPORT_BY_NAME) (codeBase + (*thunkRef));
*funcRef = (FARPROC)GetProcAddress(handle, (LPCSTR)&thunkData->Name);
}
if (*funcRef == 0)
{
SetLastError(ERROR_PROC_NOT_FOUND);
if (*funcRef == 0) {
result = 0;
break;
}
}
if (!result)
if (!result) {
break;
}
}
}
return result;
}
/*
MemoryLoadLibrary - load a library AS MEMORY MODULE, or return
existing MEMORY MODULE with increased refcount.
This allows to load a library AGAIN as memory module which is
already loaded as HMODULE!
*/
HMEMORYMODULE MemoryLoadLibrary(char *name, const void *data)
HMEMORYMODULE MemoryLoadLibrary(const void *data)
{
PMEMORYMODULE result;
PIMAGE_DOS_HEADER dos_header;
PIMAGE_NT_HEADERS old_header;
unsigned char *code, *headers;
DWORD locationDelta;
SIZE_T locationDelta;
DllEntryProc DllEntry;
BOOL successfull;
MEMORYMODULE *p = loaded;
while (p) {
// If already loaded, only increment the reference count
if (0 == stricmp(name, p->name)) {
p->refcount++;
return (HMODULE)p;
}
p = p->next;
}
/* Do NOT check for GetModuleHandle here! */
dos_header = (PIMAGE_DOS_HEADER)data;
if (dos_header->e_magic != IMAGE_DOS_SIGNATURE)
{
SetLastError(ERROR_BAD_FORMAT);
if (dos_header->e_magic != IMAGE_DOS_SIGNATURE) {
#if DEBUG_OUTPUT
OutputDebugString("Not a valid executable file.\n");
#endif
@ -455,9 +311,7 @@ HMEMORYMODULE MemoryLoadLibrary(char *name, const void *data)
}
old_header = (PIMAGE_NT_HEADERS)&((const unsigned char *)(data))[dos_header->e_lfanew];
if (old_header->Signature != IMAGE_NT_SIGNATURE)
{
SetLastError(ERROR_BAD_FORMAT);
if (old_header->Signature != IMAGE_NT_SIGNATURE) {
#if DEBUG_OUTPUT
OutputDebugString("No PE header found.\n");
#endif
@ -470,31 +324,25 @@ HMEMORYMODULE MemoryLoadLibrary(char *name, const void *data)
MEM_RESERVE,
PAGE_READWRITE);
if (code == NULL)
if (code == NULL) {
// try to allocate memory at arbitrary position
code = (unsigned char *)VirtualAlloc(NULL,
old_header->OptionalHeader.SizeOfImage,
MEM_RESERVE,
PAGE_READWRITE);
if (code == NULL)
{
SetLastError(ERROR_NOT_ENOUGH_MEMORY);
if (code == NULL) {
#if DEBUG_OUTPUT
OutputLastError("Can't reserve memory");
#endif
return NULL;
}
}
result = (PMEMORYMODULE)HeapAlloc(GetProcessHeap(), 0, sizeof(MEMORYMODULE));
result->codeBase = code;
result->numModules = 0;
result->modules = NULL;
result->initialized = 0;
result->next = result->prev = NULL;
result->refcount = 1;
result->name = strdup(name);
result->name_table = NULL;
// XXX: is it correct to commit the complete memory region at once?
// calling DllEntry raises an exception if we don't...
@ -514,31 +362,30 @@ HMEMORYMODULE MemoryLoadLibrary(char *name, const void *data)
result->headers = (PIMAGE_NT_HEADERS)&((const unsigned char *)(headers))[dos_header->e_lfanew];
// update position
result->headers->OptionalHeader.ImageBase = (DWORD)code;
result->headers->OptionalHeader.ImageBase = (POINTER_TYPE)code;
// copy sections from DLL file block to new memory location
CopySections(data, old_header, result);
// adjust base address of imported data
locationDelta = (DWORD)(code - old_header->OptionalHeader.ImageBase);
if (locationDelta != 0)
locationDelta = (SIZE_T)(code - old_header->OptionalHeader.ImageBase);
if (locationDelta != 0) {
PerformBaseRelocation(result, locationDelta);
}
// load required dlls and adjust function table of imports
if (!BuildImportTable(result))
if (!BuildImportTable(result)) {
goto error;
}
// mark memory pages depending on section headers and release
// sections that are marked as "discardable"
FinalizeSections(result);
// get entry point of loaded library
if (result->headers->OptionalHeader.AddressOfEntryPoint != 0)
{
DllEntry = (DllEntryProc)(code + result->headers->OptionalHeader.AddressOfEntryPoint);
if (DllEntry == 0)
{
SetLastError(ERROR_BAD_FORMAT); /* XXX ? */
if (result->headers->OptionalHeader.AddressOfEntryPoint != 0) {
DllEntry = (DllEntryProc) (code + result->headers->OptionalHeader.AddressOfEntryPoint);
if (DllEntry == 0) {
#if DEBUG_OUTPUT
OutputDebugString("Library has no entry point.\n");
#endif
@ -547,8 +394,7 @@ HMEMORYMODULE MemoryLoadLibrary(char *name, const void *data)
// notify library about attaching to process
successfull = (*DllEntry)((HINSTANCE)code, DLL_PROCESS_ATTACH, 0);
if (!successfull)
{
if (!successfull) {
#if DEBUG_OUTPUT
OutputDebugString("Can't attach library.\n");
#endif
@ -557,99 +403,55 @@ HMEMORYMODULE MemoryLoadLibrary(char *name, const void *data)
result->initialized = 1;
}
_Register(name, result);
return (HMEMORYMODULE)result;
error:
// cleanup
free(result->name);
MemoryFreeLibrary(result);
return NULL;
}
int _compare(const struct NAME_TABLE *p1, const struct NAME_TABLE *p2)
{
return stricmp(p1->name, p2->name);
}
int _find(const char **name, const struct NAME_TABLE *p)
{
return stricmp(*name, p->name);
}
struct NAME_TABLE *GetNameTable(PMEMORYMODULE module)
{
unsigned char *codeBase;
PIMAGE_EXPORT_DIRECTORY exports;
PIMAGE_DATA_DIRECTORY directory;
DWORD i, *nameRef;
WORD *ordinal;
struct NAME_TABLE *p, *ptab;
if (module->name_table)
return module->name_table;
codeBase = module->codeBase;
directory = GET_HEADER_DICTIONARY(module, IMAGE_DIRECTORY_ENTRY_EXPORT);
exports = (PIMAGE_EXPORT_DIRECTORY)(codeBase + directory->VirtualAddress);
nameRef = (DWORD *)(codeBase + exports->AddressOfNames);
ordinal = (WORD *)(codeBase + exports->AddressOfNameOrdinals);
p = ((PMEMORYMODULE)module)->name_table = (struct NAME_TABLE *)malloc(sizeof(struct NAME_TABLE)
* exports->NumberOfNames);
if (p == NULL)
return NULL;
ptab = p;
for (i=0; i<exports->NumberOfNames; ++i) {
p->name = (char *)(codeBase + *nameRef++);
p->ordinal = *ordinal++;
++p;
}
qsort(ptab, exports->NumberOfNames, sizeof(struct NAME_TABLE), _compare);
return ptab;
}
FARPROC MemoryGetProcAddress(HMEMORYMODULE module, const char *name)
{
unsigned char *codeBase = ((PMEMORYMODULE)module)->codeBase;
int idx=-1;
DWORD i, *nameRef;
WORD *ordinal;
PIMAGE_EXPORT_DIRECTORY exports;
PIMAGE_DATA_DIRECTORY directory = GET_HEADER_DICTIONARY((PMEMORYMODULE)module, IMAGE_DIRECTORY_ENTRY_EXPORT);
if (directory->Size == 0)
if (directory->Size == 0) {
// no export table found
return NULL;
}
exports = (PIMAGE_EXPORT_DIRECTORY)(codeBase + directory->VirtualAddress);
if (exports->NumberOfNames == 0 || exports->NumberOfFunctions == 0)
exports = (PIMAGE_EXPORT_DIRECTORY) (codeBase + directory->VirtualAddress);
if (exports->NumberOfNames == 0 || exports->NumberOfFunctions == 0) {
// DLL doesn't export anything
return NULL;
}
if (HIWORD(name)) {
struct NAME_TABLE *ptab;
struct NAME_TABLE *found;
ptab = GetNameTable((PMEMORYMODULE)module);
if (ptab == NULL)
// some failure
return NULL;
found = bsearch(&name, ptab, exports->NumberOfNames, sizeof(struct NAME_TABLE), _find);
if (found == NULL)
// search function name in list of exported names
nameRef = (DWORD *) (codeBase + exports->AddressOfNames);
ordinal = (WORD *) (codeBase + exports->AddressOfNameOrdinals);
for (i=0; i<exports->NumberOfNames; i++, nameRef++, ordinal++) {
if (_stricmp(name, (const char *) (codeBase + (*nameRef))) == 0) {
idx = *ordinal;
break;
}
}
if (idx == -1) {
// exported symbol not found
return NULL;
idx = found->ordinal;
}
else
idx = LOWORD(name) - exports->Base;
if ((DWORD)idx > exports->NumberOfFunctions)
if ((DWORD)idx > exports->NumberOfFunctions) {
// name <-> ordinal number don't match
return NULL;
}
// AddressOfFunctions contains the RVAs to the "real" functions
return (FARPROC)(codeBase + *(DWORD *)(codeBase + exports->AddressOfFunctions + (idx*4)));
return (FARPROC) (codeBase + (*(DWORD *) (codeBase + exports->AddressOfFunctions + (idx*4))));
}
void MemoryFreeLibrary(HMEMORYMODULE mod)
@ -657,32 +459,29 @@ void MemoryFreeLibrary(HMEMORYMODULE mod)
int i;
PMEMORYMODULE module = (PMEMORYMODULE)mod;
if (module != NULL)
{
if (module->initialized != 0)
{
if (module != NULL) {
if (module->initialized != 0) {
// notify library about detaching from process
DllEntryProc DllEntry = (DllEntryProc)(module->codeBase + module->headers->OptionalHeader.AddressOfEntryPoint);
DllEntryProc DllEntry = (DllEntryProc) (module->codeBase + module->headers->OptionalHeader.AddressOfEntryPoint);
(*DllEntry)((HINSTANCE)module->codeBase, DLL_PROCESS_DETACH, 0);
module->initialized = 0;
}
if (module->modules != NULL)
{
if (module->modules != NULL) {
// free previously opened libraries
for (i=0; i<module->numModules; i++)
if (module->modules[i] != INVALID_HANDLE_VALUE)
MyFreeLibrary(module->modules[i]);
for (i=0; i<module->numModules; i++) {
if (module->modules[i] != INVALID_HANDLE_VALUE) {
FreeLibrary(module->modules[i]);
}
}
free(module->modules);
}
if (module->codeBase != NULL)
if (module->codeBase != NULL) {
// release memory of library
VirtualFree(module->codeBase, 0, MEM_RELEASE);
if (module->name_table != NULL)
free(module->name_table);
}
HeapFree(GetProcessHeap(), 0, module);
}

20
setup/installer/windows/MemoryModule.h
View File

@ -1,12 +1,12 @@
/*
* Memory DLL loading code
* Version 0.0.2
* Version 0.0.3
*
* Copyright (c) 2004-2005 by Joachim Bauch / mail@joachim-bauch.de
* Copyright (c) 2004-2012 by Joachim Bauch / mail@joachim-bauch.de
* http://www.joachim-bauch.de
*
* The contents of this file are subject to the Mozilla Public License Version
* 1.1 (the "License"); you may not use this file except in compliance with
* 2.0 (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
* http://www.mozilla.org/MPL/
*
@ -19,7 +19,7 @@
*
* The Initial Developer of the Original Code is Joachim Bauch.
*
* Portions created by Joachim Bauch are Copyright (C) 2004-2005
* Portions created by Joachim Bauch are Copyright (C) 2004-2012
* Joachim Bauch. All Rights Reserved.
*
*/
@ -35,22 +35,12 @@ typedef void *HMEMORYMODULE;
extern "C" {
#endif
typedef void *(*FINDPROC)();
extern FINDPROC findproc;
extern void *findproc_data;
HMEMORYMODULE MemoryLoadLibrary(char *, const void *);
HMEMORYMODULE MemoryLoadLibrary(const void *);
FARPROC MemoryGetProcAddress(HMEMORYMODULE, const char *);
void MemoryFreeLibrary(HMEMORYMODULE);
BOOL MyFreeLibrary(HMODULE hModule);
HMODULE MyLoadLibrary(char *lpFileName);
FARPROC MyGetProcAddress(HMODULE hModule, LPCSTR lpProcName);
HMODULE MyGetModuleHandle(LPCTSTR lpModuleName);
#ifdef __cplusplus
}
#endif

65
setup/installer/windows/__init__.py
View File

@ -8,53 +8,66 @@ __docformat__ = 'restructuredtext en'
import os, shutil, subprocess
from setup import Command, __appname__, __version__
from setup import Command, __appname__, __version__, installer_name
from setup.installer import VMInstaller
class Win(Command):
description = 'Build windows binary installers'
sub_commands = ['win32']
sub_commands = ['win64', 'win32']
def run(self, opts):
pass
class Win32(VMInstaller):
description = 'Build 32bit windows binary installer'
INSTALLER_EXT = 'exe'
VM_NAME = 'xp_build'
VM = '/vmware/bin/%s'%VM_NAME
VM_CHECK = 'calibre_windows_xp_home'
class WinBase(VMInstaller):
FREEZE_COMMAND = 'win32_freeze'
FREEZE_TEMPLATE = 'python -OO setup.py {freeze_command} --no-ice'
INSTALLER_EXT = 'msi'
SHUTDOWN_CMD = ['shutdown.exe', '-s', '-f', '-t', '0']
def sign_msi(self):
print ('Signing installers ...')
subprocess.check_call(['ssh', self.VM_NAME, '~/sign.sh'], shell=False)
class Win32(WinBase):
description = 'Build 32bit windows binary installer'
VM_NAME = 'xp_build'
VM = '/vmware/bin/%s'%VM_NAME
VM_CHECK = 'calibre_windows_xp_home'
@property
def msi64(self):
return installer_name('msi', is64bit=True)
def do_dl(self, installer, errmsg):
subprocess.check_call(('scp',
'%s:build/%s/%s'%(self.VM_NAME, __appname__, installer), 'dist'))
if not os.path.exists(installer):
self.warn(errmsg)
raise SystemExit(1)
def download_installer(self):
installer = self.installer()
if os.path.exists('build/winfrozen'):
shutil.rmtree('build/winfrozen')
self.sign_msi()
subprocess.check_call(('scp',
'xp_build:build/%s/%s'%(__appname__, installer), 'dist'))
if not os.path.exists(installer):
self.warn('Failed to freeze')
raise SystemExit(1)
self.do_dl(installer, 'Failed to freeze')
installer = 'dist/%s-portable-installer-%s.exe'%(__appname__, __version__)
subprocess.check_call(('scp',
'xp_build:build/%s/%s'%(__appname__, installer), 'dist'))
if not os.path.exists(installer):
self.warn('Failed to get portable installer')
raise SystemExit(1)
self.do_dl(installer, 'Failed to get portable installer')
class Win64(WinBase):
description = 'Build 64bit windows binary installer'
VM_NAME = 'win64'
VM = '/vmware/bin/%s'%VM_NAME
VM_CHECK = 'win64'
IS_64_BIT = True
BUILD_PREFIX = WinBase.BUILD_PREFIX + [
'if [ -f "$HOME/.bash_profile" ] ; then',
' source "$HOME/.bash_profile"',
'fi',
]

110
setup/installer/windows/freeze.py
View File

@ -25,6 +25,7 @@ LZMA = r'Q:\easylzma\build\easylzma-0.0.8'
VERSION = re.sub('[a-z]\d+', '', __version__)
WINVER = VERSION+'.0'
machine = 'X64' if is64bit else 'X86'
DESCRIPTIONS = {
'calibre' : 'The main calibre program',
@ -90,6 +91,7 @@ class Win32Freeze(Command, WixMixIn):
if not is64bit:
self.build_portable()
self.build_portable_installer()
self.sign_installers()
def remove_CRT_from_manifests(self):
'''
@ -110,7 +112,7 @@ class Win32Freeze(Command, WixMixIn):
self.info('Removing CRT dependency from manifest of: %s'%bn)
# Blank out the bytes corresponding to the dependency specification
nraw = repl_pat.sub(lambda m: b' '*len(m.group()), raw)
if len(nraw) != len(raw):
if len(nraw) != len(raw) or nraw == raw:
raise Exception('Something went wrong with %s'%bn)
with open(dll, 'wb') as f:
f.write(nraw)
@ -132,6 +134,23 @@ class Win32Freeze(Command, WixMixIn):
# used instead
shutil.copy2(f, tgt)
def fix_pyd_bootstraps_in(self, folder):
for dirpath, dirnames, filenames in os.walk(folder):
for f in filenames:
name, ext = os.path.splitext(f)
bpy = self.j(dirpath, name + '.py')
if ext == '.pyd' and os.path.exists(bpy):
with open(bpy, 'rb') as f:
raw = f.read().strip()
if (not raw.startswith('def __bootstrap__') or not
raw.endswith('__bootstrap__()')):
raise Exception('The file %r has non'
' bootstrap code'%self.j(dirpath, f))
for ext in ('.py', '.pyc', '.pyo'):
x = self.j(dirpath, name+ext)
if os.path.exists(x):
os.remove(x)
def freeze(self):
shutil.copy2(self.j(self.src_root, 'LICENSE'), self.base)
@ -184,23 +203,12 @@ class Win32Freeze(Command, WixMixIn):
shutil.copytree(self.j(comext, 'shell'), self.j(sp_dir, 'win32com', 'shell'))
shutil.rmtree(comext)
# Fix PyCrypto, removing the bootstrap .py modules that load the .pyd
# modules, since they do not work when in a zip file
for crypto_dir in glob.glob(self.j(sp_dir, 'pycrypto-*', 'Crypto')):
for dirpath, dirnames, filenames in os.walk(crypto_dir):
for f in filenames:
name, ext = os.path.splitext(f)
if ext == '.pyd':
with open(self.j(dirpath, name+'.py')) as f:
raw = f.read().strip()
if (not raw.startswith('def __bootstrap__') or not
raw.endswith('__bootstrap__()')):
raise Exception('The PyCrypto file %r has non'
' bootstrap code'%self.j(dirpath, f))
for ext in ('.py', '.pyc', '.pyo'):
x = self.j(dirpath, name+ext)
if os.path.exists(x):
os.remove(x)
# Fix PyCrypto and Pillow, removing the bootstrap .py modules that load
# the .pyd modules, since they do not work when in a zip file
for folder in os.listdir(sp_dir):
folder = self.j(sp_dir, folder)
if os.path.isdir(folder):
self.fix_pyd_bootstraps_in(folder)
for pat in (r'PyQt4\uic\port_v3', ):
x = glob.glob(self.j(self.lib_dir, 'site-packages', pat))[0]
@ -367,7 +375,7 @@ class Win32Freeze(Command, WixMixIn):
if not self.opts.keep_site:
os.remove(y)
def run_builder(self, cmd):
def run_builder(self, cmd, show_output=False):
p = subprocess.Popen(cmd, stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
if p.wait() != 0:
@ -376,6 +384,9 @@ class Win32Freeze(Command, WixMixIn):
self.info(p.stdout.read())
self.info(p.stderr.read())
sys.exit(1)
if show_output:
self.info(p.stdout.read())
self.info(p.stderr.read())
def build_portable_installer(self):
zf = self.a(self.j('dist', 'calibre-portable-%s.zip.lz'%VERSION))
@ -401,7 +412,7 @@ class Win32Freeze(Command, WixMixIn):
exe = self.j('dist', 'calibre-portable-installer-%s.exe'%VERSION)
if self.newer(exe, [obj, xobj]):
self.info('Linking', exe)
cmd = [msvc.linker] + ['/INCREMENTAL:NO', '/MACHINE:X86',
cmd = [msvc.linker] + ['/INCREMENTAL:NO', '/MACHINE:'+machine,
'/LIBPATH:'+self.obj_dir, '/SUBSYSTEM:WINDOWS',
'/LIBPATH:'+(LZMA+r'\lib\Release'),
'/RELEASE', '/MANIFEST', '/MANIFESTUAC:level="asInvoker" uiAccess="false"',
@ -458,7 +469,7 @@ class Win32Freeze(Command, WixMixIn):
exe = self.j(base, 'calibre-portable.exe')
if self.newer(exe, [obj]):
self.info('Linking', exe)
cmd = [msvc.linker] + ['/INCREMENTAL:NO', '/MACHINE:X86',
cmd = [msvc.linker] + ['/INCREMENTAL:NO', '/MACHINE:'+machine,
'/LIBPATH:'+self.obj_dir, '/SUBSYSTEM:WINDOWS',
'/RELEASE',
'/ENTRY:wWinMainCRTStartup',
@ -478,6 +489,17 @@ class Win32Freeze(Command, WixMixIn):
subprocess.check_call([LZMA + r'\bin\elzma.exe', '-9', '--lzip', name])
def sign_installers(self):
self.info('Signing installers...')
files = glob.glob(self.j('dist', '*.msi')) + glob.glob(self.j('dist',
'*.exe'))
if not files:
raise ValueError('No installers found')
subprocess.check_call(['signtool.exe', 'sign', '/a', '/d',
'calibre - E-book management', '/du',
'http://calibre-ebook.com', '/t',
'http://timestamp.verisign.com/scripts/timstamp.dll'] + files)
def add_dir_to_zip(self, zf, path, prefix=''):
'''
Add a directory recursively to the zip file with an optional prefix.
@ -499,9 +521,11 @@ class Win32Freeze(Command, WixMixIn):
finally:
os.chdir(cwd)
def build_launchers(self):
def build_launchers(self, debug=False):
if not os.path.exists(self.obj_dir):
os.makedirs(self.obj_dir)
dflags = (['/Zi'] if debug else [])
dlflags = (['/DEBUG'] if debug else ['/INCREMENTAL:NO'])
base = self.j(self.src_root, 'setup', 'installer', 'windows')
sources = [self.j(base, x) for x in ['util.c', 'MemoryModule.c']]
headers = [self.j(base, x) for x in ['util.h', 'MemoryModule.h']]
@ -510,20 +534,20 @@ class Win32Freeze(Command, WixMixIn):
cflags += ['/DPYDLL="python%s.dll"'%self.py_ver, '/IC:/Python%s/include'%self.py_ver]
for src, obj in zip(sources, objects):
if not self.newer(obj, headers+[src]): continue
cmd = [msvc.cc] + cflags + ['/Fo'+obj, '/Tc'+src]
self.run_builder(cmd)
cmd = [msvc.cc] + cflags + dflags + ['/Fo'+obj, '/Tc'+src]
self.run_builder(cmd, show_output=True)
dll = self.j(self.obj_dir, 'calibre-launcher.dll')
ver = '.'.join(__version__.split('.')[:2])
if self.newer(dll, objects):
cmd = [msvc.linker, '/DLL', '/INCREMENTAL:NO', '/VERSION:'+ver,
'/OUT:'+dll, '/nologo', '/MACHINE:X86'] + objects + \
cmd = [msvc.linker, '/DLL', '/VERSION:'+ver, '/OUT:'+dll,
'/nologo', '/MACHINE:'+machine] + dlflags + objects + \
[self.embed_resources(dll),
'/LIBPATH:C:/Python%s/libs'%self.py_ver,
'python%s.lib'%self.py_ver,
'/delayload:python%s.dll'%self.py_ver]
self.info('Linking calibre-launcher.dll')
self.run_builder(cmd)
self.run_builder(cmd, show_output=True)
src = self.j(base, 'main.c')
shutil.copy2(dll, self.base)
@ -541,16 +565,16 @@ class Win32Freeze(Command, WixMixIn):
dest = self.j(self.obj_dir, bname+'.obj')
if self.newer(dest, [src]+headers):
self.info('Compiling', bname)
cmd = [msvc.cc] + xflags + ['/Tc'+src, '/Fo'+dest]
cmd = [msvc.cc] + xflags + dflags + ['/Tc'+src, '/Fo'+dest]
self.run_builder(cmd)
exe = self.j(self.base, bname+'.exe')
lib = dll.replace('.dll', '.lib')
if self.newer(exe, [dest, lib, self.rc_template, __file__]):
self.info('Linking', bname)
cmd = [msvc.linker] + ['/INCREMENTAL:NO', '/MACHINE:X86',
cmd = [msvc.linker] + ['/MACHINE:'+machine,
'/LIBPATH:'+self.obj_dir, '/SUBSYSTEM:'+subsys,
'/LIBPATH:C:/Python%s/libs'%self.py_ver, '/RELEASE',
'/OUT:'+exe, self.embed_resources(exe),
'/OUT:'+exe] + dlflags + [self.embed_resources(exe),
dest, lib]
self.run_builder(cmd)
@ -563,12 +587,18 @@ class Win32Freeze(Command, WixMixIn):
for x in (self.plugins_dir, self.dll_dir):
for pyd in os.listdir(x):
if pyd.endswith('.pyd') and pyd not in {
'unrar.pyd', 'sqlite_custom.pyd', 'calibre_style.pyd'}:
# sqlite_custom has to be a file for
# sqlite_load_extension to work
# For some reason unrar.pyd crashes when processing
# password protected RAR files if loaded from inside
# pylib.zip
'sqlite_custom.pyd',
# calibre_style has to be loaded by Qt therefore it
# must be a file
'calibre_style.pyd',
# Because of https://github.com/fancycode/MemoryModule/issues/4
# any extensions that use C++ exceptions must be loaded
# from files
'unrar.pyd', 'wpd.pyd', 'podofo.pyd',
'progress_indicator.pyd',
}:
self.add_to_zipfile(zf, pyd, x)
os.remove(self.j(x, pyd))
@ -581,6 +611,7 @@ class Win32Freeze(Command, WixMixIn):
sp = self.j(self.lib_dir, 'site-packages')
# Special handling for PIL and pywin32
handled = set(['PIL.pth', 'pywin32.pth', 'PIL', 'win32'])
if not is64bit:
self.add_to_zipfile(zf, 'PIL', sp)
base = self.j(sp, 'win32', 'lib')
for x in os.listdir(base):
@ -593,16 +624,17 @@ class Win32Freeze(Command, WixMixIn):
self.add_to_zipfile(zf, x, base)
handled.add('easy-install.pth')
# We dont want the site.py from site-packages
handled.add('site.pyo')
for d in self.get_pth_dirs(self.j(sp, 'easy-install.pth')):
handled.add(self.b(d))
for x in os.listdir(d):
if x == 'EGG-INFO':
if x in {'EGG-INFO', 'site.py', 'site.pyc', 'site.pyo'}:
continue
self.add_to_zipfile(zf, x, d)
# The rest of site-packages
# We dont want the site.py from site-packages
handled.add('site.pyo')
for x in os.listdir(sp):
if x in handled or x.endswith('.egg-info'):
continue
@ -622,8 +654,10 @@ class Win32Freeze(Command, WixMixIn):
line = line.strip()
if not line or line.startswith('#') or line.startswith('import'):
continue
candidate = self.j(base, line)
candidate = os.path.abspath(self.j(base, line))
if os.path.exists(candidate):
if not os.path.isdir(candidate):
raise ValueError('%s is not a directory'%candidate)
yield candidate
def add_to_zipfile(self, zf, name, base, exclude=frozenset()):

6
setup/installer/windows/notes.rst
View File

@ -109,10 +109,8 @@ of mimetypes from the windows registry
Python packages
------------------
Install setuptools from http://pypi.python.org/pypi/setuptools If there are no
windows binaries already compiled for the version of python you are using then
download the source and run the following command in the folder where the
source has been unpacked::
Install setuptools from http://pypi.python.org/pypi/setuptools. Use the source
tarball. Edit setup.py and set zip_safe=False. Then run::
python setup.py install

5
setup/installer/windows/portable-installer.cpp
View File

@ -418,9 +418,12 @@ static BOOL move_program() {
}
if (MoveFileEx(L"Calibre Portable\\Calibre", L"..\\Calibre", 0) == 0) {
show_last_error(L"Failed to move calibre program folder");
Sleep(4000); // Sleep and try again
if (MoveFileEx(L"Calibre Portable\\Calibre", L"..\\Calibre", 0) == 0) {
show_last_error(L"Failed to move calibre program folder. This is usually caused by an antivirus program or a file sync program like DropBox. Turn them off temporarily and try again. Underlying error: ");
return false;
}
}
if (!directory_exists(L"..\\Calibre Library")) {
MoveFileEx(L"Calibre Portable\\Calibre Library", L"..\\Calibre Library", 0);

37
setup/installer/windows/util.c
View File

@ -16,6 +16,7 @@ static char python_dll[] = PYDLL;
void set_gui_app(char yes) { GUI_APP = yes; }
char is_gui_app() { return GUI_APP; }
int calibre_show_python_error(const wchar_t *preamble, int code);
// memimporter {{{
@ -63,17 +64,6 @@ static void* FindLibrary(char *name, PyObject *callback)
return p;
}
static PyObject *set_find_proc(PyObject *self, PyObject *args)
{
PyObject *callback = NULL;
if (!PyArg_ParseTuple(args, "|O:set_find_proc", &callback))
return NULL;
Py_DECREF((PyObject *)findproc_data);
Py_INCREF(callback);
findproc_data = (void *)callback;
return Py_BuildValue("i", 1);
}
static PyObject *
import_module(PyObject *self, PyObject *args)
{
@ -92,7 +82,7 @@ import_module(PyObject *self, PyObject *args)
&data, &size,
&initfuncname, &modname, &pathname))
return NULL;
hmem = MemoryLoadLibrary(pathname, data);
hmem = MemoryLoadLibrary(data);
if (!hmem) {
PyErr_Format(*DLL_ImportError,
"MemoryLoadLibrary() failed loading %s", pathname);
@ -119,14 +109,14 @@ import_module(PyObject *self, PyObject *args)
static PyMethodDef methods[] = {
{ "import_module", import_module, METH_VARARGS,
"import_module(code, initfunc, dllname[, finder]) -> module" },
{ "set_find_proc", set_find_proc, METH_VARARGS },
{ NULL, NULL }, /* Sentinel */
};
// }}}
static int _show_error(const wchar_t *preamble, const wchar_t *msg, const int code) {
wchar_t *buf, *cbuf;
wchar_t *buf;
char *cbuf;
buf = (wchar_t*)LocalAlloc(LMEM_ZEROINIT, sizeof(wchar_t)*
(wcslen(msg) + wcslen(preamble) + 80));
@ -142,7 +132,7 @@ static int _show_error(const wchar_t *preamble, const wchar_t *msg, const int co
else {
cbuf = (char*) calloc(10+(wcslen(buf)*4), sizeof(char));
if (cbuf) {
if (WideCharToMultiByte(CP_UTF8, 0, buf, -1, cbuf, 10+(wcslen(buf)*4), NULL, NULL) != 0) printf_s(cbuf);
if (WideCharToMultiByte(CP_UTF8, 0, buf, -1, cbuf, (int)(10+(wcslen(buf)*4)), NULL, NULL) != 0) printf_s(cbuf);
free(cbuf);
}
}
@ -165,6 +155,7 @@ int show_last_error_crt(wchar_t *preamble) {
int show_last_error(wchar_t *preamble) {
wchar_t *msg = NULL;
DWORD dw = GetLastError();
int ret;
FormatMessage(
FORMAT_MESSAGE_ALLOCATE_BUFFER |
@ -173,10 +164,13 @@ int show_last_error(wchar_t *preamble) {
NULL,
dw,
MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT),
&msg,
0, NULL );
(LPWSTR)&msg,
0,
NULL );
return _show_error(preamble, msg, (int)dw);
ret = _show_error(preamble, msg, (int)dw);
if (msg != NULL) LocalFree(msg);
return ret;
}
char* get_app_dir() {
@ -254,10 +248,10 @@ void setup_stream(const char *name, const char *errors, UINT cp) {
else if (cp == CP_UTF7) _snprintf_s(buf, 100, _TRUNCATE, "%s", "utf-7");
else _snprintf_s(buf, 100, _TRUNCATE, "cp%d", cp);
stream = PySys_GetObject(name);
stream = PySys_GetObject((char*)name);
if (!PyFile_SetEncodingAndErrors(stream, buf, errors))
ExitProcess(calibre_show_python_error("Failed to set stream encoding", 1));
if (!PyFile_SetEncodingAndErrors(stream, buf, (char*)errors))
ExitProcess(calibre_show_python_error(L"Failed to set stream encoding", 1));
free(buf);
@ -372,7 +366,6 @@ void initialize_interpreter(wchar_t *outr, wchar_t *errr,
}
PySys_SetObject("argv", argv);
findproc = FindLibrary;
Py_InitModule3("_memimporter", methods, module_doc);
}

55
setup/installer/windows/wix-template.xml
View File

@ -2,7 +2,7 @@
<Wix xmlns='http://schemas.microsoft.com/wix/2006/wi' xmlns:util="http://schemas.microsoft.com/wix/UtilExtension"
>
<Product Name='{app}' Id='*' UpgradeCode='{upgrade_code}'
<Product Name='{app}{x64}' Id='*' UpgradeCode='{upgrade_code}'
Language='1033' Codepage='1252' Version='{version}' Manufacturer='Kovid Goyal'>
<Package Id='*' Keywords='Installer' Description="{app} Installer"
@ -29,19 +29,24 @@
Language="1033"
Property="NEWPRODUCTFOUND"/>
</Upgrade>
<CustomAction Id="PreventDowngrading" Error="Newer version already installed."/>
<CustomAction Id="PreventDowngrading" Error="Newer version of {app} already installed. If you want to downgrade you must uninstall {app} first."/>
<Property Id="APPLICATIONFOLDER">
<RegistrySearch Id='calibreInstDir' Type='raw'
Root='HKLM' Key="Software\{app}\Installer" Name="InstallPath" />
Root='HKLM' Key="Software\{app}{x64}\Installer" Name="InstallPath" />
</Property>
<Directory Id='TARGETDIR' Name='SourceDir'>
<Directory Id='ProgramFilesFolder' Name='PFiles'>
<Directory Id='APPLICATIONFOLDER' Name='{app}' />
<Directory Id='{ProgramFilesFolder}' Name='PFiles'>
<!-- The name must be calibre on 32 bit to ensure
that the component guids dont change compared
to previous msis. However, on 64 bit it must
be Calibre2 otherwise by default it will
install to C:\Program Files\calibre -->
<Directory Id='APPLICATIONFOLDER' Name="{appfolder}" />
</Directory>
<Directory Id="ProgramMenuFolder">
<Directory Id="ApplicationProgramsFolder" Name="{app} - E-book Management"/>
<Directory Id="ApplicationProgramsFolder" Name="{app}{x64} - E-book Management"/>
</Directory>
<Directory Id="DesktopFolder" Name="Desktop"/>
</Directory>
@ -50,24 +55,24 @@
{app_components}
<Component Id="AddToPath" Guid="*">
<Environment Id='UpdatePath' Name='PATH' Action='set' System='yes' Part='last' Value='[APPLICATIONFOLDER]' />
<RegistryValue Root="HKCU" Key="Software\Microsoft\{app}" Name="system_path_updated" Type="integer" Value="1" KeyPath="yes"/>
<RegistryValue Root="HKCU" Key="Software\Microsoft\{app}{x64}" Name="system_path_updated" Type="integer" Value="1" KeyPath="yes"/>
</Component>
<Component Id="RememberInstallDir" Guid="*">
<RegistryValue Root="HKLM" Key="Software\{app}\Installer" Name="InstallPath" Type="string" Value="[APPLICATIONFOLDER]" KeyPath="yes"/>
<RegistryValue Root="HKLM" Key="Software\{app}{x64}\Installer" Name="InstallPath" Type="string" Value="[APPLICATIONFOLDER]" KeyPath="yes"/>
</Component>
</DirectoryRef>
<DirectoryRef Id="ApplicationProgramsFolder">
<Component Id="StartMenuShortcuts" Guid="*">
<Shortcut Id="s1" Name="{app} - E-book management"
<Shortcut Id="s1" Name="{app}{x64} - E-book management"
Description="Manage your e-book collection and download news"
Target="[#{exe_map[calibre]}]"
WorkingDirectory="APPLICATIONROOTDIRECTORY" />
<Shortcut Id="s2" Name="E-book viewer"
<Shortcut Id="s2" Name="E-book viewer{x64}"
Description="Viewer for all the major e-book formats"
Target="[#{exe_map[ebook-viewer]}]"
WorkingDirectory="APPLICATIONROOTDIRECTORY" />
<Shortcut Id="s3" Name="LRF viewer"
<Shortcut Id="s3" Name="LRF viewer{x64}"
Description="Viewer for LRF format e-books"
Target="[#{exe_map[lrfviewer]}]"
WorkingDirectory="APPLICATIONROOTDIRECTORY" />
@ -79,17 +84,17 @@
Target="http://calibre-ebook.com/get-involved"/>
<RemoveFolder Id="ApplicationProgramsFolder" On="uninstall"/>
<RegistryValue Root="HKCU" Key="Software\Microsoft\{app}" Name="start_menu_shortcuts_installed" Type="integer" Value="1" KeyPath="yes"/>
<RegistryValue Root="HKCU" Key="Software\Microsoft\{app}{x64}" Name="start_menu_shortcuts_installed" Type="integer" Value="1" KeyPath="yes"/>
</Component>
</DirectoryRef>
<DirectoryRef Id="DesktopFolder">
<Component Id="DesktopShortcut" Guid="*">
<Shortcut Id="ds1" Name="{app} - E-book management"
<Shortcut Id="ds1" Name="{app}{x64} - E-book management"
Description="Manage your e-book collection and download news"
Target="[#{exe_map[calibre]}]"
WorkingDirectory="APPLICATIONROOTDIRECTORY" />
<RegistryValue Root="HKCU" Key="Software\Microsoft\{app}" Name="desktop_shortcut_installed" Type="integer" Value="1" KeyPath="yes"/>
<RegistryValue Root="HKCU" Key="Software\Microsoft\{app}{x64}" Name="desktop_shortcut_installed" Type="integer" Value="1" KeyPath="yes"/>
</Component>
</DirectoryRef>
@ -122,17 +127,35 @@
<!-- Add icon to entry in Add/Remove programs -->
<Icon Id="main_icon" SourceFile="{main_icon}"/>
<Property Id="ARPPRODUCTICON" Value="main_icon" />
<Property Id="ARPURLINFOABOUT" Value="http://calibre-ebook.com" />
<Property Id='ARPHELPLINK' Value="http://calibre-ebook.com/help" />
<Property Id='ARPURLUPDATEINFO' Value="http://calibre-ebook.com/download_windows" />
<SetProperty Id="ARPINSTALLLOCATION" Value="[APPLICATIONFOLDER]" After="CostFinalize" />
<Condition
Message="This application is only supported on Windows XP SP3, or higher.">
<![CDATA[Installed OR (VersionNT >= 501)]]>
Message="This application is only supported on {minverhuman}, or higher.">
<![CDATA[Installed OR (VersionNT >= {minver})]]>
</Condition>
<!-- On 64 bit installers there is a bug in WiX that causes the
WixSetDefaultPerMachineFolder action to incorrectly set
APPLICATIONFOLDER to the x86 value, so we override it. See
http://stackoverflow.com/questions/5479790/wix-how-to-override-c-program-files-x86-on-x64-machine-in-wixui-advanced-s
-->
<CustomAction
Id="OverwriteWixSetDefaultPerMachineFolder"
Property="WixPerMachineFolder"
Value="[APPLICATIONFOLDER]"
Execute="immediate"
/>
<InstallExecuteSequence>
<Custom Action="PreventDowngrading" After="FindRelatedProducts">NEWPRODUCTFOUND</Custom>
{fix_wix}
<RemoveExistingProducts After="InstallFinalize" />
</InstallExecuteSequence>
<InstallUISequence>
<Custom Action="PreventDowngrading" After="FindRelatedProducts">NEWPRODUCTFOUND</Custom>
{fix_wix}
</InstallUISequence>
<UI>

64
setup/installer/windows/wix.py
View File

@ -6,11 +6,20 @@ __license__ = 'GPL v3'
__copyright__ = '2009, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'
import os, shutil, subprocess
import os, shutil, subprocess, sys
from setup import __appname__, __version__, basenames
from setup.build_environment import is64bit
if is64bit:
WIXP = r'C:\Program Files (x86)\WiX Toolset v3.6'
UPGRADE_CODE = '5DD881FF-756B-4097-9D82-8C0F11D521EA'
MINVERHUMAN = 'Windows Vista'
else:
WIXP = r'C:\Program Files\WiX Toolset v3.6'
UPGRADE_CODE = 'BEB2A80D-E902-4DAD-ADF9-8BD2DA42CFE1'
MINVERHUMAN = 'Windows XP SP3'
WIXP = r'C:\Program Files\Windows Installer XML v3.5'
CANDLE = WIXP+r'\bin\candle.exe'
LIGHT = WIXP+r'\bin\light.exe'
@ -28,8 +37,14 @@ class WixMixIn:
components = self.get_components_from_files()
wxs = template.format(
app = __appname__,
appfolder = 'Calibre2' if is64bit else __appname__,
version = __version__,
upgrade_code = 'BEB2A80D-E902-4DAD-ADF9-8BD2DA42CFE1',
upgrade_code = UPGRADE_CODE,
ProgramFilesFolder = 'ProgramFiles64Folder' if is64bit else 'ProgramFilesFolder',
x64 = ' 64bit' if is64bit else '',
minverhuman = MINVERHUMAN,
minver = '600' if is64bit else '501',
fix_wix = '<Custom Action="OverwriteWixSetDefaultPerMachineFolder" After="WixSetDefaultPerMachineFolder" />' if is64bit else '',
compression = self.opts.msi_compression,
app_components = components,
exe_map = self.smap,
@ -48,14 +63,15 @@ class WixMixIn:
with open(enusf, 'wb') as f:
f.write(enus)
wixobj = self.j(self.installer_dir, __appname__+'.wixobj')
cmd = [CANDLE, '-nologo', '-ext', 'WiXUtilExtension', '-o', wixobj, wxsf]
arch = 'x64' if is64bit else 'x86'
cmd = [CANDLE, '-nologo', '-arch', arch, '-ext', 'WiXUtilExtension', '-o', wixobj, wxsf]
self.info(*cmd)
subprocess.check_call(cmd)
self.run_wix(cmd)
self.installer = self.j(self.src_root, 'dist')
if not os.path.exists(self.installer):
os.makedirs(self.installer)
self.installer = self.j(self.installer, '%s-%s.msi' % (__appname__,
__version__))
self.installer = self.j(self.installer, '%s%s-%s.msi' % (__appname__,
('-64bit' if is64bit else ''), __version__))
license = self.j(self.src_root, 'LICENSE.rtf')
banner = self.j(self.src_root, 'icons', 'wix-banner.bmp')
dialog = self.j(self.src_root, 'icons', 'wix-dialog.bmp')
@ -66,13 +82,27 @@ class WixMixIn:
'-dWixUILicenseRtf='+license,
'-dWixUIBannerBmp='+banner,
'-dWixUIDialogBmp='+dialog]
cmd.append('-sice:ICE60') # No language in dlls warning
cmd.extend([
'-sice:ICE60',# No language in dlls warning
'-sice:ICE61',# Allow upgrading with same version number
'-sice:ICE40', # Re-install mode overriden
'-sice:ICE69', # Shortcut components are part of a different feature than the files they point to
])
if self.opts.no_ice:
cmd.append('-sval')
if self.opts.verbose:
cmd.append('-v')
self.info(*cmd)
subprocess.check_call(cmd)
self.run_wix(cmd)
def run_wix(self, cmd):
p = subprocess.Popen(cmd, stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
ret = p.wait()
self.info(p.stdout.read())
self.info(p.stderr.read())
if ret != 0:
sys.exit(1)
def get_components_from_files(self):
@ -103,7 +133,20 @@ class WixMixIn:
(fid, f, x, checksum),
'</Component>'
]
components.append(''.join(c))
if x.endswith('.exe') and not x.startswith('pdf'):
# Add the executable to app paths so that users can
# launch it from the run dialog even if it is not on
# the path. See http://msdn.microsoft.com/en-us/library/windows/desktop/ee872121(v=vs.85).aspx
c[-1:-1] = [
('<RegistryValue Root="HKLM" '
r'Key="SOFTWARE\Microsoft\Windows\CurrentVersion\App '
r'Paths\%s" Value="[#file_%d]" Type="string" />'%(x, fid)),
('<RegistryValue Root="HKLM" '
r'Key="SOFTWARE\Microsoft\Windows\CurrentVersion\App '
r'Paths\{0}" Name="Path" Value="[APPLICATIONFOLDER]" '
'Type="string" />'.format(x)),
]
components.append('\n'.join(c))
return components
components = process_dir(os.path.abspath(self.base))
@ -114,4 +157,3 @@ class WixMixIn:
return '\t\t\t\t'+'\n\t\t\t\t'.join(components)

14524
setup/iso_639/ms.po
View File

File diff suppressed because it is too large Load Diff

3
setup/upload.py
View File

@ -29,6 +29,7 @@ STAGING_DIR = '/root/staging'
def installers():
installers = list(map(installer_name, ('dmg', 'msi', 'tar.bz2')))
installers.append(installer_name('tar.bz2', is64bit=True))
installers.append(installer_name('msi', is64bit=True))
installers.insert(0, 'dist/%s-%s.tar.xz'%(__appname__, __version__))
installers.append('dist/%s-portable-installer-%s.exe'%(__appname__, __version__))
return installers
@ -40,7 +41,7 @@ def installer_description(fname):
bits = '32' if 'i686' in fname else '64'
return bits + 'bit Linux binary'
if fname.endswith('.msi'):
return 'Windows installer'
return 'Windows %sinstaller'%('64bit ' if '64bit' in fname else '')
if fname.endswith('.dmg'):
return 'OS X dmg'
if fname.endswith('.exe'):

2
src/calibre/constants.py
View File

@ -28,7 +28,7 @@ isunix = isosx or islinux
isportable = os.environ.get('CALIBRE_PORTABLE_BUILD', None) is not None
ispy3 = sys.version_info.major > 2
isxp = iswindows and sys.getwindowsversion().major < 6
is64bit = sys.maxint > (1 << 32)
is64bit = sys.maxsize > (1 << 32)
isworker = os.environ.has_key('CALIBRE_WORKER') or os.environ.has_key('CALIBRE_SIMPLE_WORKER')
if isworker:
os.environ.pop('CALIBRE_FORCE_ANSI', None)

6
src/calibre/debug.py
View File

@ -148,10 +148,10 @@ def print_basic_debug_info(out=None):
out = functools.partial(prints, file=out)
import platform
from calibre.constants import (__appname__, get_version, isportable, isosx,
isfrozen)
isfrozen, is64bit)
out(__appname__, get_version(), 'Portable' if isportable else '',
'isfrozen:', isfrozen)
out(platform.platform(), platform.system())
'isfrozen:', isfrozen, 'is64bit:', is64bit)
out(platform.platform(), platform.system(), platform.architecture())
out(platform.system_alias(platform.system(), platform.release(),
platform.version()))
out('Python', platform.python_version())

2
src/calibre/devices/__init__.py
View File

@ -182,7 +182,7 @@ def debug(ioreg_to_tmp=False, buf=None, plugins=None,
out(ioreg)
if hasattr(buf, 'getvalue'):
return buf.getvalue().decode('utf-8')
return buf.getvalue().decode('utf-8', 'replace')
finally:
sys.stdout = oldo
sys.stderr = olde

4
src/calibre/devices/android/driver.py
View File

@ -232,7 +232,7 @@ class ANDROID(USBMS):
'THINKPAD_TABLET', 'SGH-T989', 'YP-G70', 'STORAGE_DEVICE',
'ADVANCED', 'SGH-I727', 'USB_FLASH_DRIVER', 'ANDROID',
'S5830I_CARD', 'MID7042', 'LINK-CREATE', '7035', 'VIEWPAD_7E',
'NOVO7', 'MB526', '_USB#WYK7MSF8KE']
'NOVO7', 'MB526', '_USB#WYK7MSF8KE', 'TABLET_PC']
WINDOWS_CARD_A_MEM = ['ANDROID_PHONE', 'GT-I9000_CARD', 'SGH-I897',
'FILE-STOR_GADGET', 'SGH-T959_CARD', 'SGH-T959', 'SAMSUNG_ANDROID', 'GT-P1000_CARD',
'A70S', 'A101IT', '7', 'INCREDIBLE', 'A7EB', 'SGH-T849_CARD',
@ -243,7 +243,7 @@ class ANDROID(USBMS):
'FILE-CD_GADGET', 'GT-I9001_CARD', 'USB_2.0', 'XT875',
'UMS_COMPOSITE', 'PRO', '.KOBO_VOX', 'SGH-T989_CARD', 'SGH-I727',
'USB_FLASH_DRIVER', 'ANDROID', 'MID7042', '7035', 'VIEWPAD_7E',
'NOVO7', 'ADVANCED']
'NOVO7', 'ADVANCED', 'TABLET_PC']
OSX_MAIN_MEM = 'Android Device Main Memory'

2
src/calibre/devices/kobo/driver.py
View File

@ -2357,6 +2357,8 @@ class KOBOTOUCH(KOBO):
update_query = 'UPDATE content SET Series=?, SeriesNumber==? where BookID is Null and ContentID = ?'
if book.series is None:
update_values = (None, None, book.contentID, )
elif book.series_index is None: # This should never happen, but...
update_values = (book.series, None, book.contentID, )
else:
update_values = (book.series, "%g"%book.series_index, book.contentID, )

1
src/calibre/devices/mtp/unix/devices.c
View File

@ -16,6 +16,7 @@ const calibre_device_entry_t calibre_mtp_device_table[] = {
// Nexus 10
, { "Google", 0x18d1, "Nexus 10", 0x4ee2, DEVICE_FLAGS_ANDROID_BUGS}
, { "Google", 0x18d1, "Nexus 10", 0x4ee1, DEVICE_FLAGS_ANDROID_BUGS}
, { NULL, 0xffff, NULL, 0xffff, DEVICE_FLAG_NONE }
};

10
src/calibre/devices/smart_device_app/driver.py
View File

@ -54,6 +54,8 @@ def synchronous(tlockname):
class ConnectionListener (Thread):
NOT_SERVICED_COUNT = 6
def __init__(self, driver):
Thread.__init__(self)
self.daemon = True
@ -78,8 +80,8 @@ class ConnectionListener (Thread):
if not self.driver.connection_queue.empty():
queue_not_serviced_count += 1
if queue_not_serviced_count >= 3:
self.driver._debug('queue not serviced')
if queue_not_serviced_count >= self.NOT_SERVICED_COUNT:
self.driver._debug('queue not serviced', queue_not_serviced_count)
try:
sock = self.driver.connection_queue.get_nowait()
s = self.driver._json_encode(
@ -1281,10 +1283,10 @@ class SMART_DEVICE_APP(DeviceConfig, DevicePlugin):
self._close_listen_socket()
return message
else:
while i < 100: # try up to 100 random port numbers
while i < 100: # try 9090 then up to 99 random port numbers
i += 1
port = self._attach_to_port(self.listen_socket,
random.randint(8192, 32000))
9090 if i == 1 else random.randint(8192, 32000))
if port != 0:
break
if port == 0:

20
src/calibre/ebooks/mobi/reader/markup.py
View File

@ -74,11 +74,12 @@ def remove_kindlegen_markup(parts):
part = "".join(srcpieces)
parts[i] = part
# we can safely remove all of the Kindlegen generated data-AmznPageBreak tags
# we can safely remove all of the Kindlegen generated data-AmznPageBreak
# attributes
find_tag_with_AmznPageBreak_pattern = re.compile(
r'''(<[^>]*\sdata-AmznPageBreak=[^>]*>)''', re.IGNORECASE)
within_tag_AmznPageBreak_position_pattern = re.compile(
r'''\sdata-AmznPageBreak=['"][^'"]*['"]''')
r'''\sdata-AmznPageBreak=['"]([^'"]*)['"]''')
for i in xrange(len(parts)):
part = parts[i]
@ -86,10 +87,8 @@ def remove_kindlegen_markup(parts):
for j in range(len(srcpieces)):
tag = srcpieces[j]
if tag.startswith('<'):
for m in within_tag_AmznPageBreak_position_pattern.finditer(tag):
replacement = ''
tag = within_tag_AmznPageBreak_position_pattern.sub(replacement, tag, 1)
srcpieces[j] = tag
srcpieces[j] = within_tag_AmznPageBreak_position_pattern.sub(
lambda m:' style="page-break-after:%s"'%m.group(1), tag)
part = "".join(srcpieces)
parts[i] = part
@ -203,7 +202,7 @@ def update_flow_links(mobi8_reader, resource_map, log):
# All flows are now unicode and have links resolved
return flows
def insert_flows_into_markup(parts, flows, mobi8_reader):
def insert_flows_into_markup(parts, flows, mobi8_reader, log):
mr = mobi8_reader
# kindle:flow:XXXX?mime=YYYY/ZZZ (used for style sheets, svg images, etc)
@ -219,7 +218,12 @@ def insert_flows_into_markup(parts, flows, mobi8_reader):
if tag.startswith('<'):
for m in flow_pattern.finditer(tag):
num = int(m.group(1), 32)
try:
fi = mr.flowinfo[num]
except IndexError:
log.warn('Ignoring invalid flow reference: %s'%m.group())
tag = ''
else:
if fi.format == 'inline':
tag = flows[num]
else:
@ -313,7 +317,7 @@ def expand_mobi8_markup(mobi8_reader, resource_map, log):
flows = update_flow_links(mobi8_reader, resource_map, log)
# Insert inline flows into the markup
insert_flows_into_markup(parts, flows, mobi8_reader)
insert_flows_into_markup(parts, flows, mobi8_reader, log)
# Insert raster images into markup
insert_images_into_markup(parts, resource_map, log)

44
src/calibre/ebooks/mobi/reader/mobi8.py
View File

@ -44,6 +44,18 @@ def locate_beg_end_of_tag(ml, aid):
return plt, pgt
return 0, 0
def reverse_tag_iter(block):
''' Iterate over all tags in block in reverse order, i.e. last tag
to first tag. '''
end = len(block)
while True:
pgt = block.rfind(b'>', 0, end)
if pgt == -1: break
plt = block.rfind(b'<', 0, pgt)
if plt == -1: break
yield block[plt:pgt+1]
end = plt
class Mobi8Reader(object):
def __init__(self, mobi6_reader, log):
@ -275,13 +287,12 @@ class Mobi8Reader(object):
return '%s/%s'%(fi.type, fi.filename), idtext
def get_id_tag(self, pos):
# find the correct tag by actually searching in the destination
# textblock at position
# Find the first tag with a named anchor (name or id attribute) before
# pos
fi = self.get_file_info(pos)
if fi.num is None and fi.start is None:
raise ValueError('No file contains pos: %d'%pos)
textblock = self.parts[fi.num]
id_map = []
npos = pos - fi.start
pgt = textblock.find(b'>', npos)
plt = textblock.find(b'<', npos)
@ -290,28 +301,15 @@ class Mobi8Reader(object):
if plt == npos or pgt < plt:
npos = pgt + 1
textblock = textblock[0:npos]
# find id links only inside of tags
# inside any < > pair find all "id=' and return whatever is inside
# the quotes
id_pattern = re.compile(br'''<[^>]*\sid\s*=\s*['"]([^'"]*)['"][^>]*>''',
re.IGNORECASE)
for m in re.finditer(id_pattern, textblock):
id_map.append((m.start(), m.group(1)))
id_re = re.compile(br'''<[^>]+\sid\s*=\s*['"]([^'"]+)['"]''')
name_re = re.compile(br'''<\s*a\s*\sname\s*=\s*['"]([^'"]+)['"]''')
for tag in reverse_tag_iter(textblock):
m = id_re.match(tag) or name_re.match(tag)
if m is not None:
return m.group(1)
if not id_map:
# Found no id in the textblock, link must be to top of file
# No tag found, link to start of file
return b''
# if npos is before first id= inside a tag, return the first
if npos < id_map[0][0]:
return id_map[0][1]
# if npos is after the last id= inside a tag, return the last
if npos > id_map[-1][0]:
return id_map[-1][1]
# otherwise find last id before npos
for i, item in enumerate(id_map):
if npos < item[0]:
return id_map[i-1][1]
return id_map[0][1]
def create_guide(self):
guide = Guide()

6
src/calibre/ebooks/oeb/reader.py
View File

@ -320,13 +320,11 @@ class OEBReader(object):
self.logger.warn(u'Spine item %r not found' % idref)
continue
item = manifest.ids[idref]
if item.media_type.lower() in OEB_DOCS and hasattr(item.data, 'xpath'):
spine.add(item, elem.get('linear'))
for item in spine:
if item.media_type.lower() not in OEB_DOCS:
if not hasattr(item.data, 'xpath'):
else:
self.oeb.log.warn('The item %s is not a XML document.'
' Removing it from spine.'%item.href)
spine.remove(item)
if len(spine) == 0:
raise OEBError("Spine is empty")
self._spine_add_extra()

8
src/calibre/ebooks/oeb/transforms/structure.py
View File

@ -114,7 +114,9 @@ class DetectStructure(object):
def find_matches(expr, doc):
try:
return XPath(expr)(doc)
ans = XPath(expr)(doc)
len(ans)
return ans
except:
self.log.warn('Invalid chapter expression, ignoring: %s'%expr)
return []
@ -203,7 +205,9 @@ class DetectStructure(object):
def find_matches(expr, doc):
try:
return XPath(expr)(doc)
ans = XPath(expr)(doc)
len(ans)
return ans
except:
self.log.warn('Invalid ToC expression, ignoring: %s'%expr)
return []

6
src/calibre/ebooks/pdf/writer.py
View File

@ -27,10 +27,10 @@ def get_custom_size(opts):
custom_size = None
if opts.custom_size != None:
width, sep, height = opts.custom_size.partition('x')
if height != '':
if height:
try:
width = int(width)
height = int(height)
width = float(width)
height = float(height)
custom_size = (width, height)
except:
custom_size = None

4
src/calibre/gui2/store/stores/libri_de_plugin.py
View File

@ -72,8 +72,8 @@ class LibreDEStore(BasicStoreConfig, StorePlugin):
mobi = details.xpath(
'boolean(.//span[@class="bindername" and contains(text(), "mobipocket")]/text())')
cover_url = ''.join(data.xpath('.//div[@class="coverImg"]/a/img/@src'))
price = ''.join(data.xpath('.//span[@class="preis"]/text()')).replace('*', '').strip()
cover_url = ''.join(data.xpath('.//div[@class="coverimg"]/a/img/@src'))
price = ''.join(data.xpath('.//div[@class="preis"]/text()')).replace('*', '').strip()
counter -= 1

14
src/calibre/gui2/update.py
View File

@ -8,7 +8,7 @@ from PyQt4.Qt import (QThread, pyqtSignal, Qt, QUrl, QDialog, QGridLayout,
import mechanize
from calibre.constants import (__appname__, __version__, iswindows, isosx,
isportable)
isportable, is64bit)
from calibre import browser, prints, as_unicode
from calibre.utils.config import prefs
from calibre.gui2 import config, dynamic, open_url
@ -19,6 +19,13 @@ URL = 'http://status.calibre-ebook.com/latest'
NO_CALIBRE_UPDATE = '-0.0.0'
VSEP = '|'
def get_download_url():
which = ('portable' if isportable else 'windows' if iswindows
else 'osx' if isosx else 'linux')
if which == 'windows' and is64bit:
which += '64'
return 'http://calibre-ebook.com/download_' + which
def get_newest_version():
br = browser()
req = mechanize.Request(URL)
@ -116,10 +123,7 @@ class UpdateNotification(QDialog):
config.set('new_version_notification', bool(self.cb.isChecked()))
def accept(self):
url = ('http://calibre-ebook.com/download_' +
('portable' if isportable else 'windows' if iswindows
else 'osx' if isosx else 'linux'))
open_url(QUrl(url))
open_url(QUrl(get_download_url()))
QDialog.accept(self)

11
src/calibre/library/server/browse.py
View File

@ -22,6 +22,7 @@ from calibre.library.comments import comments_to_html
from calibre.library.server import custom_fields_to_display
from calibre.library.field_metadata import category_icon_map
from calibre.library.server.utils import quote, unquote
from calibre.ebooks.metadata.sources.identify import urls_from_identifiers
def xml(*args, **kwargs):
ans = prepare_string_for_xml(*args, **kwargs)
@ -823,6 +824,16 @@ class BrowseServer(object):
if field in ('title', 'formats') or not args.get(field, False) \
or not m['name']:
continue
if field == 'identifiers':
urls = urls_from_identifiers(mi.get(field, {}))
links = [u'<a class="details_category_link" target="_new" href="%s" title="%s:%s">%s</a>' % (url, id_typ, id_val, name)
for name, id_typ, id_val, url in urls]
links = u', '.join(links)
if links:
fields.append((m['name'], u'<strong>%s: </strong>%s'%(
_('Ids'), links)))
continue
if m['datatype'] == 'rating':
r = u'<strong>%s: </strong>'%xml(m['name']) + \
render_rating(mi.get(field)/2.0, self.opts.url_prefix,

1014
src/calibre/translations/af.po
View File

File diff suppressed because it is too large Load Diff

1020
src/calibre/translations/ar.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/ast.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/az.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/ber.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/bg.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/bn.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/br.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/bs.po
View File

File diff suppressed because it is too large Load Diff

1043
src/calibre/translations/ca.po
View File

File diff suppressed because it is too large Load Diff

1016
src/calibre/translations/cs.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/cy.po
View File

File diff suppressed because it is too large Load Diff

1016
src/calibre/translations/da.po
View File

File diff suppressed because it is too large Load Diff

1052
src/calibre/translations/de.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/el.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/en_AU.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/en_CA.po
View File

File diff suppressed because it is too large Load Diff

1040
src/calibre/translations/en_GB.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/eo.po
View File

File diff suppressed because it is too large Load Diff

1045
src/calibre/translations/es.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/et.po
View File

File diff suppressed because it is too large Load Diff

1022
src/calibre/translations/eu.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/fa.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/fi.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/fo.po
View File

File diff suppressed because it is too large Load Diff

1064
src/calibre/translations/fr.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/fr_CA.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/fur.po
View File

File diff suppressed because it is too large Load Diff

1017
src/calibre/translations/gl.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/gu.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/he.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/hi.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/him.po
View File

File diff suppressed because it is too large Load Diff

1016
src/calibre/translations/hr.po
View File

File diff suppressed because it is too large Load Diff

1062
src/calibre/translations/hu.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/id.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/is.po
View File

File diff suppressed because it is too large Load Diff

1024
src/calibre/translations/it.po
View File

File diff suppressed because it is too large Load Diff

1046
src/calibre/translations/ja.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/jv.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/ka.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/kn.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/ko.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/ku.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/lt.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/ltg.po
View File

File diff suppressed because it is too large Load Diff

1012
src/calibre/translations/lv.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/mk.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/ml.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/mr.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/ms.po
View File

File diff suppressed because it is too large Load Diff

1016
src/calibre/translations/nb.po
View File

File diff suppressed because it is too large Load Diff

1017
src/calibre/translations/nds.po
View File

File diff suppressed because it is too large Load Diff

1023
src/calibre/translations/nl.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/nn.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/oc.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/pa.po
View File

File diff suppressed because it is too large Load Diff

1018
src/calibre/translations/pl.po
View File

File diff suppressed because it is too large Load Diff

1017
src/calibre/translations/pt.po
View File

File diff suppressed because it is too large Load Diff

1016
src/calibre/translations/pt_BR.po
View File

File diff suppressed because it is too large Load Diff

1017
src/calibre/translations/ro.po
View File

File diff suppressed because it is too large Load Diff

1029
src/calibre/translations/ru.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/sc.po
View File

File diff suppressed because it is too large Load Diff

1014
src/calibre/translations/si.po
View File

File diff suppressed because it is too large Load Diff

Some files were not shown because too many files have changed in this diff Show More