Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Updated Ming Pao

Browse Source
This commit is contained in:
Kovid Goyal 2011-12-20 09:19:38 +05:30
parent cef64ff0e7
commit ff6dd9c16a
3 changed files with 1106 additions and 359 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

257
recipes/ming_pao.recipe
View File

@ -10,6 +10,10 @@ __MakePeriodical__ = True
__UseChineseTitle__ = False
# Set it to False if you want to skip images (Default: True)
__KeepImages__ = True
# Set it to True if you want to include a summary in Kindle's article view (Default: False)
__IncludeSummary__ = False
# Set it to True if you want thumbnail images in Kindle's article view (Default: True)
__IncludeThumbnails__ = True
# (HK only) Turn below to True if you wish to use life.mingpao.com as the main article source (Default: True)
__UseLife__ = True
# (HK only) It is to disable premium content (Default: False)
@ -24,6 +28,9 @@ __Date__ = ''
'''
Change Log:
2011/12/18: update the overridden create_odf(.) routine with the one from Calibre version 0.8.31. Move __UseChineseTitle__ usage away
from create_odf(.). Optional support of text_summary and thumbnail images in Kindle's article view. Start new day
download of Hong Kong Mingpao at 4.30am. Set the actual publication date shown on kindle device.
2011/12/01: take care of situation that in txt source parsing, the article content does start with special character u'\u3010'
2011/10/21: fix a bug that hi-res img is unavailable in pages parsed from source txt
2011/10/19: fix a bug in txt source parsing
@ -53,6 +60,8 @@ Change Log:
2010/10/31: skip repeated articles in section pages
'''
from calibre import (browser, iswindows, __appname__, force_unicode, preferred_encoding, as_unicode)
from calibre.utils.date import now as nowf
import os, datetime, re, mechanize
from calibre.web.feeds.recipes import BasicNewsRecipe
from contextlib import nested
@ -60,11 +69,15 @@ from calibre.ebooks.BeautifulSoup import BeautifulSoup
from calibre.ebooks.metadata.opf2 import OPFCreator
from calibre.ebooks.metadata.toc import TOC
from calibre.ebooks.metadata import MetaInformation
from calibre.utils.localization import canonicalize_lang
# MAIN CLASS
class MPRecipe(BasicNewsRecipe):
if __Region__ == 'Hong Kong':
title = 'Ming Pao - Hong Kong'
if __UseChineseTitle__ == True:
title = u'\u660e\u5831 (\u9999\u6e2f)'
else:
title = 'Ming Pao - Hong Kong'
description = 'Hong Kong Chinese Newspaper (http://news.mingpao.com)'
category = 'Chinese, News, Hong Kong'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px; max-height:90%;} font>b {font-size:200%; font-weight:bold;} div[class=heading] {font-size:200%; font-weight:bold;} div[class=images] {font-size:50%;}'
@ -109,7 +122,10 @@ class MPRecipe(BasicNewsRecipe):
lambda match: "</b>")
]
elif __Region__ == 'Vancouver':
title = 'Ming Pao - Vancouver'
if __UseChineseTitle__ == True:
title = u'\u660e\u5831 (\u6eab\u54e5\u83ef)'
else:
title = 'Ming Pao - Vancouver'
description = 'Vancouver Chinese Newspaper (http://www.mingpaovan.com)'
category = 'Chinese, News, Vancouver'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}'
@ -127,7 +143,10 @@ class MPRecipe(BasicNewsRecipe):
lambda match: ''),
]
elif __Region__ == 'Toronto':
title = 'Ming Pao - Toronto'
if __UseChineseTitle__ == True:
title = u'\u660e\u5831 (\u591a\u502b\u591a)'
else:
title = 'Ming Pao - Toronto'
description = 'Toronto Chinese Newspaper (http://www.mingpaotor.com)'
category = 'Chinese, News, Toronto'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}'
@ -161,9 +180,9 @@ class MPRecipe(BasicNewsRecipe):
def get_dtlocal(self):
dt_utc = datetime.datetime.utcnow()
if __Region__ == 'Hong Kong':
# convert UTC to local hk time - at HKT 5.30am, all news are available
dt_local = dt_utc + datetime.timedelta(8.0/24) - datetime.timedelta(5.5/24)
# dt_local = dt_utc.astimezone(pytz.timezone('Asia/Hong_Kong')) - datetime.timedelta(5.5/24)
# convert UTC to local hk time - at HKT 4.30am, all news are available
dt_local = dt_utc + datetime.timedelta(8.0/24) - datetime.timedelta(4.5/24)
# dt_local = dt_utc.astimezone(pytz.timezone('Asia/Hong_Kong')) - datetime.timedelta(4.5/24)
elif __Region__ == 'Vancouver':
# convert UTC to local Vancouver time - at PST time 5.30am, all news are available
dt_local = dt_utc + datetime.timedelta(-8.0/24) - datetime.timedelta(5.5/24)
@ -185,6 +204,18 @@ class MPRecipe(BasicNewsRecipe):
return __Date__[0:4]+'-'+__Date__[4:6]+'-'+__Date__[6:8]
else:
return self.get_dtlocal().strftime("%Y-%m-%d")
def get_fetchyear(self):
if __Date__ <> '':
return __Date__[0:4]
else:
return self.get_dtlocal().strftime("%Y")
def get_fetchmonth(self):
if __Date__ <> '':
return __Date__[4:6]
else:
return self.get_dtlocal().strftime("%m")
def get_fetchday(self):
if __Date__ <> '':
@ -654,77 +685,153 @@ class MPRecipe(BasicNewsRecipe):
del item['absmiddle']
return soup
def populate_article_metadata(self, article, soup, first):
# thumbnails shouldn't be available if using hi-res images
if __IncludeThumbnails__ and __HiResImg__ == False and first and hasattr(self, 'add_toc_thumbnail'):
img = soup.find('img')
if img is not None:
self.add_toc_thumbnail(article, img['src'])
try:
if __IncludeSummary__ and len(article.text_summary.strip()) == 0:
# look for content
articlebodies = soup.findAll('div',attrs={'id':'newscontent'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'id':'newscontent01'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'class':'content'})
if not articlebodies:
articlebodies = soup.findAll('div', attrs={'id':'font'})
if articlebodies:
for articlebody in articlebodies:
if articlebody:
# the text may or may not be enclosed in <p></p> tag
paras = articlebody.findAll('p')
if not paras:
paras = articlebody
textFound = False
for p in paras:
if not textFound:
summary_candidate = self.tag_to_string(p).strip()
summary_candidate = summary_candidate.replace(u'\u3010\u660e\u5831\u5c08\u8a0a\u3011', '', 1)
if len(summary_candidate) > 0:
article.summary = article.text_summary = summary_candidate
textFound = True
else:
# display a simple text
#article.summary = article.text_summary = u'\u66f4\u591a......'
# display word counts
counts = 0
articlebodies = soup.findAll('div',attrs={'id':'newscontent'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'id':'newscontent01'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'class':'content'})
if not articlebodies:
articlebodies = soup.findAll('div', attrs={'id':'font'})
if articlebodies:
for articlebody in articlebodies:
# the text may or may not be enclosed in <p></p> tag
paras = articlebody.findAll('p')
if not paras:
paras = articlebody
for p in paras:
summary_candidate = self.tag_to_string(p).strip()
counts += len(summary_candidate)
article.summary = article.text_summary = u'\uff08' + str(counts) + u'\u5b57\uff09'
except:
self.log("Error creating article descriptions")
return
# override from the one in version 0.8.31
def create_opf(self, feeds, dir=None):
if dir is None:
dir = self.output_dir
if __UseChineseTitle__ == True:
if __Region__ == 'Hong Kong':
title = u'\u660e\u5831 (\u9999\u6e2f)'
elif __Region__ == 'Vancouver':
title = u'\u660e\u5831 (\u6eab\u54e5\u83ef)'
elif __Region__ == 'Toronto':
title = u'\u660e\u5831 (\u591a\u502b\u591a)'
else:
title = self.short_title()
# if not generating a periodical, force date to apply in title
if __MakePeriodical__ == False:
title = self.short_title()
# change 1: allow our own flag to tell if a periodical is to be generated
# also use customed date instead of current time
if __MakePeriodical__ == False or self.output_profile.periodical_date_in_title:
title = title + ' ' + self.get_fetchformatteddate()
if True:
mi = MetaInformation(title, [self.publisher])
mi.publisher = self.publisher
mi.author_sort = self.publisher
if __MakePeriodical__ == True:
mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title()
else:
mi.publication_type = self.publication_type+':'+self.short_title()
#mi.timestamp = nowf()
mi.timestamp = self.get_dtlocal()
mi.comments = self.description
if not isinstance(mi.comments, unicode):
mi.comments = mi.comments.decode('utf-8', 'replace')
#mi.pubdate = nowf()
mi.pubdate = self.get_dtlocal()
opf_path = os.path.join(dir, 'index.opf')
ncx_path = os.path.join(dir, 'index.ncx')
opf = OPFCreator(dir, mi)
# Add mastheadImage entry to <guide> section
mp = getattr(self, 'masthead_path', None)
if mp is not None and os.access(mp, os.R_OK):
from calibre.ebooks.metadata.opf2 import Guide
ref = Guide.Reference(os.path.basename(self.masthead_path), os.getcwdu())
ref.type = 'masthead'
ref.title = 'Masthead Image'
opf.guide.append(ref)
# end of change 1
# change 2: __appname__ replaced by newspaper publisher
__appname__ = self.publisher
mi = MetaInformation(title, [__appname__])
mi.publisher = __appname__
mi.author_sort = __appname__
# change 3: use __MakePeriodical__ flag to tell if a periodical should be generated
if __MakePeriodical__ == True:
mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title()
else:
mi.publication_type = self.publication_type+':'+self.short_title()
#mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title()
# change 4: in the following, all the nowf() are changed to adjusted time
# This one doesn't matter
mi.timestamp = nowf()
# change 5: skip listing the articles
#article_titles, aseen = [], set()
#for f in feeds:
# for a in f:
# if a.title and a.title not in aseen:
# aseen.add(a.title)
# article_titles.append(force_unicode(a.title, 'utf-8'))
manifest = [os.path.join(dir, 'feed_%d'%i) for i in range(len(feeds))]
manifest.append(os.path.join(dir, 'index.html'))
manifest.append(os.path.join(dir, 'index.ncx'))
#mi.comments = self.description
#if not isinstance(mi.comments, unicode):
# mi.comments = mi.comments.decode('utf-8', 'replace')
#mi.comments += ('\n\n' + _('Articles in this issue: ') + '\n' +
# '\n\n'.join(article_titles))
# Get cover
cpath = getattr(self, 'cover_path', None)
if cpath is None:
pf = open(os.path.join(dir, 'cover.jpg'), 'wb')
if self.default_cover(pf):
cpath = pf.name
if cpath is not None and os.access(cpath, os.R_OK):
opf.cover = cpath
manifest.append(cpath)
language = canonicalize_lang(self.language)
if language is not None:
mi.language = language
# This one affects the pub date shown in kindle title
#mi.pubdate = nowf()
# now appears to need the time field to be > 12.00noon as well
mi.pubdate = datetime.datetime(int(self.get_fetchyear()), int(self.get_fetchmonth()), int(self.get_fetchday()), 12, 30, 0)
opf_path = os.path.join(dir, 'index.opf')
ncx_path = os.path.join(dir, 'index.ncx')
# Get masthead
mpath = getattr(self, 'masthead_path', None)
if mpath is not None and os.access(mpath, os.R_OK):
manifest.append(mpath)
opf = OPFCreator(dir, mi)
# Add mastheadImage entry to <guide> section
mp = getattr(self, 'masthead_path', None)
if mp is not None and os.access(mp, os.R_OK):
from calibre.ebooks.metadata.opf2 import Guide
ref = Guide.Reference(os.path.basename(self.masthead_path), os.getcwdu())
ref.type = 'masthead'
ref.title = 'Masthead Image'
opf.guide.append(ref)
manifest = [os.path.join(dir, 'feed_%d'%i) for i in range(len(feeds))]
manifest.append(os.path.join(dir, 'index.html'))
manifest.append(os.path.join(dir, 'index.ncx'))
# Get cover
cpath = getattr(self, 'cover_path', None)
if cpath is None:
pf = open(os.path.join(dir, 'cover.jpg'), 'wb')
if self.default_cover(pf):
cpath = pf.name
if cpath is not None and os.access(cpath, os.R_OK):
opf.cover = cpath
manifest.append(cpath)
# Get masthead
mpath = getattr(self, 'masthead_path', None)
if mpath is not None and os.access(mpath, os.R_OK):
manifest.append(mpath)
opf.create_manifest_from_files_in(manifest)
for mani in opf.manifest:
if mani.path.endswith('.ncx'):
mani.id = 'ncx'
if mani.path.endswith('mastheadImage.jpg'):
mani.id = 'masthead-image'
entries = ['index.html']
toc = TOC(base_path=dir)
self.play_order_counter = 0
self.play_order_map = {}
opf.create_manifest_from_files_in(manifest)
for mani in opf.manifest:
if mani.path.endswith('.ncx'):
mani.id = 'ncx'
if mani.path.endswith('mastheadImage.jpg'):
mani.id = 'masthead-image'
entries = ['index.html']
toc = TOC(base_path=dir)
self.play_order_counter = 0
self.play_order_map = {}
def feed_index(num, parent):
f = feeds[num]
@ -739,13 +846,16 @@ class MPRecipe(BasicNewsRecipe):
desc = None
else:
desc = self.description_limiter(desc)
tt = a.toc_thumbnail if a.toc_thumbnail else None
entries.append('%sindex.html'%adir)
po = self.play_order_map.get(entries[-1], None)
if po is None:
self.play_order_counter += 1
po = self.play_order_counter
parent.add_item('%sindex.html'%adir, None, a.title if a.title else _('Untitled Article'),
play_order=po, author=auth, description=desc)
parent.add_item('%sindex.html'%adir, None,
a.title if a.title else _('Untitled Article'),
play_order=po, author=auth,
description=desc, toc_thumbnail=tt)
last = os.path.join(self.output_dir, ('%sindex.html'%adir).replace('/', os.sep))
for sp in a.sub_pages:
prefix = os.path.commonprefix([opf_path, sp])
@ -762,7 +872,7 @@ class MPRecipe(BasicNewsRecipe):
prefix = '/'.join('..'for i in range(2*len(re.findall(r'link\d+', last))))
templ = self.navbar.generate(True, num, j, len(f),
not self.has_single_feed,
a.orig_url, self.publisher, prefix=prefix,
a.orig_url, __appname__, prefix=prefix,
center=self.center_navbar)
elem = BeautifulSoup(templ.render(doctype='xhtml').decode('utf-8')).find('div')
body.insert(len(body.contents), elem)
@ -785,7 +895,7 @@ class MPRecipe(BasicNewsRecipe):
if not desc:
desc = None
feed_index(i, toc.add_item('feed_%d/index.html'%i, None,
f.title, play_order=po, description=desc, author=auth))
f.title, play_order=po, description=desc, author=auth))
else:
entries.append('feed_%d/index.html'%0)
@ -798,4 +908,5 @@ class MPRecipe(BasicNewsRecipe):
with nested(open(opf_path, 'wb'), open(ncx_path, 'wb')) as (opf_file, ncx_file):
opf.render(opf_file, ncx_file)

604
recipes/ming_pao_toronto.recipe
View File

@ -4,18 +4,41 @@ __copyright__ = '2010-2011, Eddie Lau'
# Region - Hong Kong, Vancouver, Toronto
__Region__ = 'Toronto'
# Users of Kindle 3 with limited system-level CJK support
# please replace the following "True" with "False".
# please replace the following "True" with "False". (Default: True)
__MakePeriodical__ = True
# Turn below to true if your device supports display of CJK titles
# Turn below to True if your device supports display of CJK titles (Default: False)
__UseChineseTitle__ = False
# Set it to False if you want to skip images
# Set it to False if you want to skip images (Default: True)
__KeepImages__ = True
# (HK only) Turn below to true if you wish to use life.mingpao.com as the main article source
# Set it to True if you want to include a summary in Kindle's article view (Default: False)
__IncludeSummary__ = False
# Set it to True if you want thumbnail images in Kindle's article view (Default: True)
__IncludeThumbnails__ = True
# (HK only) Turn below to True if you wish to use life.mingpao.com as the main article source (Default: True)
__UseLife__ = True
# (HK only) It is to disable premium content (Default: False)
__InclPremium__ = False
# (HK only) Turn below to True if you wish to parse articles in news.mingpao.com with their printer-friendly formats (Default: True)
__ParsePFF__ = True
# (HK only) Turn below to True if you wish hi-res images (Default: False)
__HiResImg__ = False
# Override the date returned by the program if specifying a YYYYMMDD below
__Date__ = ''
'''
Change Log:
2011/12/18: update the overridden create_odf(.) routine with the one from Calibre version 0.8.31. Move __UseChineseTitle__ usage away
from create_odf(.). Optional support of text_summary and thumbnail images in Kindle's article view. Start new day
download of Hong Kong Mingpao at 4.30am. Set the actual publication date shown on kindle device.
2011/12/01: take care of situation that in txt source parsing, the article content does start with special character u'\u3010'
2011/10/21: fix a bug that hi-res img is unavailable in pages parsed from source txt
2011/10/19: fix a bug in txt source parsing
2011/10/17: disable fetching of premium content, also improved txt source parsing
2011/10/04: option to get hi-res photos for the articles
2011/09/21: fetching "column" section is made optional.
2011/09/18: parse "column" section stuff from source text file directly.
2011/09/07: disable "column" section as it is no longer offered free.
2011/06/26: add fetching Vancouver and Toronto versions of the paper, also provide captions for images using life.mingpao fetch source
provide options to remove all images in the file
2011/05/12: switch the main parse source to life.mingpao.com, which has more photos on the article pages
@ -37,30 +60,39 @@ Change Log:
2010/10/31: skip repeated articles in section pages
'''
import os, datetime, re
from calibre import (browser, iswindows, __appname__, force_unicode, preferred_encoding, as_unicode)
from calibre.utils.date import now as nowf
import os, datetime, re, mechanize
from calibre.web.feeds.recipes import BasicNewsRecipe
from contextlib import nested
from calibre.ebooks.BeautifulSoup import BeautifulSoup
from calibre.ebooks.metadata.opf2 import OPFCreator
from calibre.ebooks.metadata.toc import TOC
from calibre.ebooks.metadata import MetaInformation
from calibre.utils.localization import canonicalize_lang
# MAIN CLASS
class MPRecipe(BasicNewsRecipe):
if __Region__ == 'Hong Kong':
title = 'Ming Pao - Hong Kong'
if __UseChineseTitle__ == True:
title = u'\u660e\u5831 (\u9999\u6e2f)'
else:
title = 'Ming Pao - Hong Kong'
description = 'Hong Kong Chinese Newspaper (http://news.mingpao.com)'
category = 'Chinese, News, Hong Kong'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} font>b {font-size:200%; font-weight:bold;}'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px; max-height:90%;} font>b {font-size:200%; font-weight:bold;} div[class=heading] {font-size:200%; font-weight:bold;} div[class=images] {font-size:50%;}'
masthead_url = 'http://news.mingpao.com/image/portals_top_logo_news.gif'
keep_only_tags = [dict(name='h1'),
dict(name='font', attrs={'style':['font-size:14pt; line-height:160%;']}), # for entertainment page title
dict(name='font', attrs={'color':['AA0000']}), # for column articles title
dict(attrs={'class':['heading']}), # for heading from txt
dict(attrs={'id':['newscontent']}), # entertainment and column page content
dict(attrs={'id':['newscontent01','newscontent02']}),
dict(attrs={'class':['content']}), # for content from txt
dict(attrs={'class':['photo']}),
dict(name='table', attrs={'width':['100%'], 'border':['0'], 'cellspacing':['5'], 'cellpadding':['0']}), # content in printed version of life.mingpao.com
dict(name='img', attrs={'width':['180'], 'alt':['按圖放大']}) # images for source from life.mingpao.com
dict(name='img', attrs={'width':['180'], 'alt':['????']}), # images for source from life.mingpao.com
dict(attrs={'class':['images']}) # for images from txt
]
if __KeepImages__:
remove_tags = [dict(name='style'),
@ -90,7 +122,10 @@ class MPRecipe(BasicNewsRecipe):
lambda match: "</b>")
]
elif __Region__ == 'Vancouver':
title = 'Ming Pao - Vancouver'
if __UseChineseTitle__ == True:
title = u'\u660e\u5831 (\u6eab\u54e5\u83ef)'
else:
title = 'Ming Pao - Vancouver'
description = 'Vancouver Chinese Newspaper (http://www.mingpaovan.com)'
category = 'Chinese, News, Vancouver'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}'
@ -108,7 +143,10 @@ class MPRecipe(BasicNewsRecipe):
lambda match: ''),
]
elif __Region__ == 'Toronto':
title = 'Ming Pao - Toronto'
if __UseChineseTitle__ == True:
title = u'\u660e\u5831 (\u591a\u502b\u591a)'
else:
title = 'Ming Pao - Toronto'
description = 'Toronto Chinese Newspaper (http://www.mingpaotor.com)'
category = 'Chinese, News, Toronto'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}'
@ -139,49 +177,12 @@ class MPRecipe(BasicNewsRecipe):
conversion_options = {'linearize_tables':True}
timefmt = ''
def image_url_processor(cls, baseurl, url):
# trick: break the url at the first occurance of digit, add an additional
# '_' at the front
# not working, may need to move this to preprocess_html() method
# minIdx = 10000
# i0 = url.find('0')
# if i0 >= 0 and i0 < minIdx:
# minIdx = i0
# i1 = url.find('1')
# if i1 >= 0 and i1 < minIdx:
# minIdx = i1
# i2 = url.find('2')
# if i2 >= 0 and i2 < minIdx:
# minIdx = i2
# i3 = url.find('3')
# if i3 >= 0 and i0 < minIdx:
# minIdx = i3
# i4 = url.find('4')
# if i4 >= 0 and i4 < minIdx:
# minIdx = i4
# i5 = url.find('5')
# if i5 >= 0 and i5 < minIdx:
# minIdx = i5
# i6 = url.find('6')
# if i6 >= 0 and i6 < minIdx:
# minIdx = i6
# i7 = url.find('7')
# if i7 >= 0 and i7 < minIdx:
# minIdx = i7
# i8 = url.find('8')
# if i8 >= 0 and i8 < minIdx:
# minIdx = i8
# i9 = url.find('9')
# if i9 >= 0 and i9 < minIdx:
# minIdx = i9
return url
def get_dtlocal(self):
dt_utc = datetime.datetime.utcnow()
if __Region__ == 'Hong Kong':
# convert UTC to local hk time - at HKT 5.30am, all news are available
dt_local = dt_utc + datetime.timedelta(8.0/24) - datetime.timedelta(5.5/24)
# dt_local = dt_utc.astimezone(pytz.timezone('Asia/Hong_Kong')) - datetime.timedelta(5.5/24)
# convert UTC to local hk time - at HKT 4.30am, all news are available
dt_local = dt_utc + datetime.timedelta(8.0/24) - datetime.timedelta(4.5/24)
# dt_local = dt_utc.astimezone(pytz.timezone('Asia/Hong_Kong')) - datetime.timedelta(4.5/24)
elif __Region__ == 'Vancouver':
# convert UTC to local Vancouver time - at PST time 5.30am, all news are available
dt_local = dt_utc + datetime.timedelta(-8.0/24) - datetime.timedelta(5.5/24)
@ -193,13 +194,34 @@ class MPRecipe(BasicNewsRecipe):
return dt_local
def get_fetchdate(self):
return self.get_dtlocal().strftime("%Y%m%d")
if __Date__ <> '':
return __Date__
else:
return self.get_dtlocal().strftime("%Y%m%d")
def get_fetchformatteddate(self):
return self.get_dtlocal().strftime("%Y-%m-%d")
if __Date__ <> '':
return __Date__[0:4]+'-'+__Date__[4:6]+'-'+__Date__[6:8]
else:
return self.get_dtlocal().strftime("%Y-%m-%d")
def get_fetchyear(self):
if __Date__ <> '':
return __Date__[0:4]
else:
return self.get_dtlocal().strftime("%Y")
def get_fetchmonth(self):
if __Date__ <> '':
return __Date__[4:6]
else:
return self.get_dtlocal().strftime("%m")
def get_fetchday(self):
return self.get_dtlocal().strftime("%d")
if __Date__ <> '':
return __Date__[6:8]
else:
return self.get_dtlocal().strftime("%d")
def get_cover_url(self):
if __Region__ == 'Hong Kong':
@ -230,12 +252,23 @@ class MPRecipe(BasicNewsRecipe):
(u'\u570b\u969b World', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalta', 'nal'),
(u'\u7d93\u6fdf Finance', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea', 'nal'),
(u'\u9ad4\u80b2 Sport', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalsp', 'nal'),
(u'\u5f71\u8996 Film/TV', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalma', 'nal'),
(u'\u5c08\u6b04 Columns', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn', 'ncl')]:
articles = self.parse_section2(url, keystr)
(u'\u5f71\u8996 Film/TV', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalma', 'nal')
]:
if __InclPremium__ == True:
articles = self.parse_section2_txt(url, keystr)
else:
articles = self.parse_section2(url, keystr)
if articles:
feeds.append((title, articles))
if __InclPremium__ == True:
# parse column section articles directly from .txt files
for title, url, keystr in [(u'\u5c08\u6b04 Columns', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn', 'ncl')
]:
articles = self.parse_section2_txt(url, keystr)
if articles:
feeds.append((title, articles))
for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'),
(u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]:
articles = self.parse_section(url)
@ -244,15 +277,16 @@ class MPRecipe(BasicNewsRecipe):
else:
for title, url in [(u'\u8981\u805e Headline', 'http://news.mingpao.com/' + dateStr + '/gaindex.htm'),
(u'\u6e2f\u805e Local', 'http://news.mingpao.com/' + dateStr + '/gbindex.htm'),
(u'\u6559\u80b2 Education', 'http://news.mingpao.com/' + dateStr + '/gfindex.htm')]:
(u'\u6559\u80b2 Education', 'http://news.mingpao.com/' + dateStr + '/gfindex.htm'),
(u'\u793e\u8a55/\u7b46\u9663 Editorial', 'http://news.mingpao.com/' + dateStr + '/mrindex.htm')]:
articles = self.parse_section(url)
if articles:
feeds.append((title, articles))
# special- editorial
ed_articles = self.parse_ed_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr')
if ed_articles:
feeds.append((u'\u793e\u8a55/\u7b46\u9663 Editorial', ed_articles))
#ed_articles = self.parse_ed_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr')
#if ed_articles:
# feeds.append((u'\u793e\u8a55/\u7b46\u9663 Editorial', ed_articles))
for title, url in [(u'\u8ad6\u58c7 Forum', 'http://news.mingpao.com/' + dateStr + '/faindex.htm'),
(u'\u4e2d\u570b China', 'http://news.mingpao.com/' + dateStr + '/caindex.htm'),
@ -263,32 +297,46 @@ class MPRecipe(BasicNewsRecipe):
# special - finance
#fin_articles = self.parse_fin_section('http://www.mpfinance.com/htm/Finance/' + dateStr + '/News/ea,eb,ecindex.htm')
fin_articles = self.parse_fin_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea')
if fin_articles:
feeds.append((u'\u7d93\u6fdf Finance', fin_articles))
#fin_articles = self.parse_fin_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea')
#if fin_articles:
# feeds.append((u'\u7d93\u6fdf Finance', fin_articles))
for title, url in [('Tech News', 'http://news.mingpao.com/' + dateStr + '/naindex.htm'),
(u'\u9ad4\u80b2 Sport', 'http://news.mingpao.com/' + dateStr + '/spindex.htm')]:
articles = self.parse_section(url)
for title, url, keystr in [(u'\u7d93\u6fdf Finance', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea', 'nal')]:
articles = self.parse_section2_txt(url, keystr)
if articles:
feeds.append((title, articles))
#for title, url in [('Tech News', 'http://news.mingpao.com/' + dateStr + '/naindex.htm'),
# (u'\u9ad4\u80b2 Sport', 'http://news.mingpao.com/' + dateStr + '/spindex.htm')]:
# articles = self.parse_section(url)
# if articles:
# feeds.append((title, articles))
# special - entertainment
ent_articles = self.parse_ent_section('http://ol.mingpao.com/cfm/star1.cfm')
if ent_articles:
feeds.append((u'\u5f71\u8996 Film/TV', ent_articles))
#ent_articles = self.parse_ent_section('http://ol.mingpao.com/cfm/star1.cfm')
#if ent_articles:
# feeds.append((u'\u5f71\u8996 Film/TV', ent_articles))
for title, url, keystr in [(u'\u5f71\u8996 Film/TV', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalma', 'nal')
]:
articles = self.parse_section2_txt(url, keystr)
if articles:
feeds.append((title, articles))
if __InclPremium__ == True:
# parse column section articles directly from .txt files
for title, url, keystr in [(u'\u5c08\u6b04 Columns', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn', 'ncl')
]:
articles = self.parse_section2_txt(url, keystr)
if articles:
feeds.append((title, articles))
for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'),
(u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]:
articles = self.parse_section(url)
if articles:
feeds.append((title, articles))
# special- columns
col_articles = self.parse_col_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn')
if col_articles:
feeds.append((u'\u5c08\u6b04 Columns', col_articles))
elif __Region__ == 'Vancouver':
for title, url in [(u'\u8981\u805e Headline', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VAindex.htm'),
(u'\u52a0\u570b Canada', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VBindex.htm'),
@ -332,6 +380,16 @@ class MPRecipe(BasicNewsRecipe):
title = self.tag_to_string(a)
url = a.get('href', False)
url = 'http://news.mingpao.com/' + dateStr + '/' +url
# replace the url to the print-friendly version
if __ParsePFF__ == True:
if url.rfind('Redirect') <> -1 and __InclPremium__ == True:
url = re.sub(dateStr + '.*' + dateStr, dateStr, url)
url = re.sub('%2F.*%2F', '/', url)
title = title.replace(u'\u6536\u8cbb\u5167\u5bb9', '')
url = url.replace('%2Etxt', '_print.htm')
url = url.replace('%5F', '_')
else:
url = url.replace('.htm', '_print.htm')
if url not in included_urls and url.rfind('Redirect') == -1:
current_articles.append({'title': title, 'url': url, 'description':'', 'date':''})
included_urls.append(url)
@ -340,6 +398,8 @@ class MPRecipe(BasicNewsRecipe):
# parse from life.mingpao.com
def parse_section2(self, url, keystr):
br = mechanize.Browser()
br.set_handle_redirect(False)
self.get_fetchdate()
soup = self.index_to_soup(url)
a = soup.findAll('a', href=True)
@ -350,12 +410,34 @@ class MPRecipe(BasicNewsRecipe):
title = self.tag_to_string(i)
url = 'http://life.mingpao.com/cfm/' + i.get('href', False)
if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind(keystr) == -1):
url = url.replace('dailynews3.cfm', 'dailynews3a.cfm') # use printed version of the article
try:
br.open_novisit(url)
url = url.replace('dailynews3.cfm', 'dailynews3a.cfm') # use printed version of the article
current_articles.append({'title': title, 'url': url, 'description': ''})
included_urls.append(url)
except:
print 'skipping a premium article'
current_articles.reverse()
return current_articles
# parse from text file of life.mingpao.com
def parse_section2_txt(self, url, keystr):
self.get_fetchdate()
soup = self.index_to_soup(url)
a = soup.findAll('a', href=True)
a.reverse()
current_articles = []
included_urls = []
for i in a:
title = self.tag_to_string(i)
url = 'http://life.mingpao.com/cfm/' + i.get('href', False)
if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind(keystr) == -1):
url = url.replace('cfm/dailynews3.cfm?File=', 'ftp/Life3/') # use printed version of the article
current_articles.append({'title': title, 'url': url, 'description': ''})
included_urls.append(url)
current_articles.reverse()
return current_articles
# parse from www.mingpaovan.com
def parse_section3(self, url, baseUrl):
self.get_fetchdate()
@ -438,6 +520,162 @@ class MPRecipe(BasicNewsRecipe):
current_articles.reverse()
return current_articles
# preprocess those .txt and javascript based files
def preprocess_raw_html(self, raw_html, url):
new_html = raw_html
if url.rfind('ftp') <> -1 or url.rfind('_print.htm') <> -1:
if url.rfind('_print.htm') <> -1:
# javascript based file
splitter = re.compile(r'\n')
new_raw_html = '<html><head><title>Untitled</title></head>'
new_raw_html = new_raw_html + '<body>'
for item in splitter.split(raw_html):
if item.startswith('var heading1 ='):
heading = item.replace('var heading1 = \'', '')
heading = heading.replace('\'', '')
heading = heading.replace(';', '')
new_raw_html = new_raw_html + '<div class="heading">' + heading
if item.startswith('var heading2 ='):
heading = item.replace('var heading2 = \'', '')
heading = heading.replace('\'', '')
heading = heading.replace(';', '')
if heading <> '':
new_raw_html = new_raw_html + '<br>' + heading + '</div>'
else:
new_raw_html = new_raw_html + '</div>'
if item.startswith('var content ='):
content = item.replace("var content = ", '')
content = content.replace('\'', '')
content = content.replace(';', '')
new_raw_html = new_raw_html + '<div class="content">' + content + '</div>'
if item.startswith('var photocontent ='):
photo = item.replace('var photocontent = \'', '')
photo = photo.replace('\'', '')
photo = photo.replace(';', '')
photo = photo.replace('<tr>', '')
photo = photo.replace('<td>', '')
photo = photo.replace('</tr>', '')
photo = photo.replace('</td>', '<br>')
photo = photo.replace('class="photo"', '')
new_raw_html = new_raw_html + '<div class="images">' + photo + '</div>'
new_html = new_raw_html + '</body></html>'
else:
# .txt based file
splitter = re.compile(r'\n') # Match non-digits
new_raw_html = '<html><head><title>Untitled</title></head><body><div class="images">'
next_is_img_txt = False
title_started = False
title_break_reached = False
met_article_start_char = False
for item in splitter.split(raw_html):
item = item.strip()
# if title already reached but break between title and content not yet found, record title_break_reached
if title_started == True and title_break_reached == False and item == '':
title_break_reached = True
# if title reached and title_break_reached and met_article_start_char == False and item is not empty
# start content
elif title_started == True and title_break_reached == True and met_article_start_char == False:
if item <> '':
met_article_start_char = True
new_raw_html = new_raw_html + '</div><div class="content"><p>' + item + '<p>\n'
#if item.startswith(u'\u3010'):
# met_article_start_char = True
# new_raw_html = new_raw_html + '</div><div class="content"><p>' + item + '<p>\n'
else:
if next_is_img_txt == False:
if item.startswith("=@"):
print 'skip movie link'
elif item.startswith("=?"):
next_is_img_txt = True
new_raw_html += '<img src="' + str(item)[2:].strip() + '.gif" /><p>\n'
elif item.startswith('=='):
next_is_img_txt = True
if False:
# TODO: check existence of .gif first
newimg = '_' + item[2:].strip() + '.jpg'
new_raw_html += '<img src="' + newimg + '" /><p>\n'
else:
new_raw_html += '<img src="' + str(item)[2:].strip() + '.jpg" /><p>\n'
elif item.startswith('='):
next_is_img_txt = True
if False:
# TODO: check existence of .gif first
newimg = '_' + item[1:].strip() + '.jpg'
new_raw_html += '<img src="' + newimg + '" /><p>\n'
else:
new_raw_html += '<img src="' + str(item)[1:].strip() + '.jpg" /><p>\n'
else:
if next_is_img_txt == False and met_article_start_char == False:
if item <> '':
if title_started == False:
#print 'Title started at ', item
new_raw_html = new_raw_html + '</div><div class="heading">' + item + '\n'
title_started = True
else:
new_raw_html = new_raw_html + item + '\n'
else:
new_raw_html = new_raw_html + item + '<p>\n'
else:
next_is_img_txt = False
new_raw_html = new_raw_html + item + '\n'
new_html = new_raw_html + '</div></body></html>'
#raw_html = raw_html.replace(u'<p>\u3010', u'\u3010')
if __HiResImg__ == True:
# TODO: add a _ in front of an image url
if url.rfind('news.mingpao.com') > -1:
imglist = re.findall('src="?.*?jpg"', new_html)
br = mechanize.Browser()
br.set_handle_redirect(False)
for img in imglist:
gifimg = img.replace('jpg"', 'gif"')
try:
br.open_novisit(url + "/../" + gifimg[5:len(gifimg)-1])
new_html = new_html.replace(img, gifimg)
except:
# find the location of the first _
pos = img.find('_')
if pos > -1:
# if found, insert _ after the first _
newimg = img[0:pos] + '_' + img[pos:]
new_html = new_html.replace(img, newimg)
else:
# if not found, insert _ after "
new_html = new_html.replace(img[1:], '"_' + img[1:])
elif url.rfind('life.mingpao.com') > -1:
imglist = re.findall('src=\'?.*?jpg\'', new_html)
br = mechanize.Browser()
br.set_handle_redirect(False)
#print 'Img list: ', imglist, '\n'
for img in imglist:
#print 'Found img: ', img
gifimg = img.replace('jpg\'', 'gif\'')
try:
gifurl = re.sub(r'dailynews.*txt', '', url)
br.open_novisit(gifurl + gifimg[5:len(gifimg)-1])
new_html = new_html.replace(img, gifimg)
except:
pos = img.rfind('/')
newimg = img[0:pos+1] + '_' + img[pos+1:]
new_html = new_html.replace(img, newimg)
# repeat with src quoted by double quotes, for text parsed from src txt
imglist = re.findall('src="?.*?jpg"', new_html)
for img in imglist:
#print 'Found img: ', img
gifimg = img.replace('jpg"', 'gif"')
try:
#print 'url', url
pos = url.rfind('/')
gifurl = url[:pos+1]
#print 'try it:', gifurl + gifimg[5:len(gifimg)-1]
br.open_novisit(gifurl + gifimg[5:len(gifimg)-1])
new_html = new_html.replace(img, gifimg)
except:
pos = img.find('"')
newimg = img[0:pos+1] + '_' + img[pos+1:]
#print 'Use hi-res img', newimg
new_html = new_html.replace(img, newimg)
return new_html
def preprocess_html(self, soup):
for item in soup.findAll(style=True):
del item['style']
@ -446,78 +684,154 @@ class MPRecipe(BasicNewsRecipe):
for item in soup.findAll(stype=True):
del item['absmiddle']
return soup
def populate_article_metadata(self, article, soup, first):
# thumbnails shouldn't be available if using hi-res images
if __IncludeThumbnails__ and __HiResImg__ == False and first and hasattr(self, 'add_toc_thumbnail'):
img = soup.find('img')
if img is not None:
self.add_toc_thumbnail(article, img['src'])
try:
if __IncludeSummary__ and len(article.text_summary.strip()) == 0:
# look for content
articlebodies = soup.findAll('div',attrs={'id':'newscontent'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'id':'newscontent01'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'class':'content'})
if not articlebodies:
articlebodies = soup.findAll('div', attrs={'id':'font'})
if articlebodies:
for articlebody in articlebodies:
if articlebody:
# the text may or may not be enclosed in <p></p> tag
paras = articlebody.findAll('p')
if not paras:
paras = articlebody
textFound = False
for p in paras:
if not textFound:
summary_candidate = self.tag_to_string(p).strip()
summary_candidate = summary_candidate.replace(u'\u3010\u660e\u5831\u5c08\u8a0a\u3011', '', 1)
if len(summary_candidate) > 0:
article.summary = article.text_summary = summary_candidate
textFound = True
else:
# display a simple text
#article.summary = article.text_summary = u'\u66f4\u591a......'
# display word counts
counts = 0
articlebodies = soup.findAll('div',attrs={'id':'newscontent'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'id':'newscontent01'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'class':'content'})
if not articlebodies:
articlebodies = soup.findAll('div', attrs={'id':'font'})
if articlebodies:
for articlebody in articlebodies:
# the text may or may not be enclosed in <p></p> tag
paras = articlebody.findAll('p')
if not paras:
paras = articlebody
for p in paras:
summary_candidate = self.tag_to_string(p).strip()
counts += len(summary_candidate)
article.summary = article.text_summary = u'\uff08' + str(counts) + u'\u5b57\uff09'
except:
self.log("Error creating article descriptions")
return
# override from the one in version 0.8.31
def create_opf(self, feeds, dir=None):
if dir is None:
dir = self.output_dir
if __UseChineseTitle__ == True:
if __Region__ == 'Hong Kong':
title = u'\u660e\u5831 (\u9999\u6e2f)'
elif __Region__ == 'Vancouver':
title = u'\u660e\u5831 (\u6eab\u54e5\u83ef)'
elif __Region__ == 'Toronto':
title = u'\u660e\u5831 (\u591a\u502b\u591a)'
else:
title = self.short_title()
# if not generating a periodical, force date to apply in title
if __MakePeriodical__ == False:
title = self.short_title()
# change 1: allow our own flag to tell if a periodical is to be generated
# also use customed date instead of current time
if __MakePeriodical__ == False or self.output_profile.periodical_date_in_title:
title = title + ' ' + self.get_fetchformatteddate()
if True:
mi = MetaInformation(title, [self.publisher])
mi.publisher = self.publisher
mi.author_sort = self.publisher
if __MakePeriodical__ == True:
mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title()
else:
mi.publication_type = self.publication_type+':'+self.short_title()
#mi.timestamp = nowf()
mi.timestamp = self.get_dtlocal()
mi.comments = self.description
if not isinstance(mi.comments, unicode):
mi.comments = mi.comments.decode('utf-8', 'replace')
#mi.pubdate = nowf()
mi.pubdate = self.get_dtlocal()
opf_path = os.path.join(dir, 'index.opf')
ncx_path = os.path.join(dir, 'index.ncx')
opf = OPFCreator(dir, mi)
# Add mastheadImage entry to <guide> section
mp = getattr(self, 'masthead_path', None)
if mp is not None and os.access(mp, os.R_OK):
from calibre.ebooks.metadata.opf2 import Guide
ref = Guide.Reference(os.path.basename(self.masthead_path), os.getcwdu())
ref.type = 'masthead'
ref.title = 'Masthead Image'
opf.guide.append(ref)
# end of change 1
# change 2: __appname__ replaced by newspaper publisher
__appname__ = self.publisher
mi = MetaInformation(title, [__appname__])
mi.publisher = __appname__
mi.author_sort = __appname__
# change 3: use __MakePeriodical__ flag to tell if a periodical should be generated
if __MakePeriodical__ == True:
mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title()
else:
mi.publication_type = self.publication_type+':'+self.short_title()
#mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title()
# change 4: in the following, all the nowf() are changed to adjusted time
# This one doesn't matter
mi.timestamp = nowf()
# change 5: skip listing the articles
#article_titles, aseen = [], set()
#for f in feeds:
# for a in f:
# if a.title and a.title not in aseen:
# aseen.add(a.title)
# article_titles.append(force_unicode(a.title, 'utf-8'))
manifest = [os.path.join(dir, 'feed_%d'%i) for i in range(len(feeds))]
manifest.append(os.path.join(dir, 'index.html'))
manifest.append(os.path.join(dir, 'index.ncx'))
#mi.comments = self.description
#if not isinstance(mi.comments, unicode):
# mi.comments = mi.comments.decode('utf-8', 'replace')
#mi.comments += ('\n\n' + _('Articles in this issue: ') + '\n' +
# '\n\n'.join(article_titles))
# Get cover
cpath = getattr(self, 'cover_path', None)
if cpath is None:
pf = open(os.path.join(dir, 'cover.jpg'), 'wb')
if self.default_cover(pf):
cpath = pf.name
if cpath is not None and os.access(cpath, os.R_OK):
opf.cover = cpath
manifest.append(cpath)
language = canonicalize_lang(self.language)
if language is not None:
mi.language = language
# This one affects the pub date shown in kindle title
#mi.pubdate = nowf()
# now appears to need the time field to be > 12.00noon as well
mi.pubdate = datetime.datetime(int(self.get_fetchyear()), int(self.get_fetchmonth()), int(self.get_fetchday()), 12, 30, 0)
opf_path = os.path.join(dir, 'index.opf')
ncx_path = os.path.join(dir, 'index.ncx')
# Get masthead
mpath = getattr(self, 'masthead_path', None)
if mpath is not None and os.access(mpath, os.R_OK):
manifest.append(mpath)
opf = OPFCreator(dir, mi)
# Add mastheadImage entry to <guide> section
mp = getattr(self, 'masthead_path', None)
if mp is not None and os.access(mp, os.R_OK):
from calibre.ebooks.metadata.opf2 import Guide
ref = Guide.Reference(os.path.basename(self.masthead_path), os.getcwdu())
ref.type = 'masthead'
ref.title = 'Masthead Image'
opf.guide.append(ref)
manifest = [os.path.join(dir, 'feed_%d'%i) for i in range(len(feeds))]
manifest.append(os.path.join(dir, 'index.html'))
manifest.append(os.path.join(dir, 'index.ncx'))
# Get cover
cpath = getattr(self, 'cover_path', None)
if cpath is None:
pf = open(os.path.join(dir, 'cover.jpg'), 'wb')
if self.default_cover(pf):
cpath = pf.name
if cpath is not None and os.access(cpath, os.R_OK):
opf.cover = cpath
manifest.append(cpath)
# Get masthead
mpath = getattr(self, 'masthead_path', None)
if mpath is not None and os.access(mpath, os.R_OK):
manifest.append(mpath)
opf.create_manifest_from_files_in(manifest)
for mani in opf.manifest:
if mani.path.endswith('.ncx'):
mani.id = 'ncx'
if mani.path.endswith('mastheadImage.jpg'):
mani.id = 'masthead-image'
entries = ['index.html']
toc = TOC(base_path=dir)
self.play_order_counter = 0
self.play_order_map = {}
opf.create_manifest_from_files_in(manifest)
for mani in opf.manifest:
if mani.path.endswith('.ncx'):
mani.id = 'ncx'
if mani.path.endswith('mastheadImage.jpg'):
mani.id = 'masthead-image'
entries = ['index.html']
toc = TOC(base_path=dir)
self.play_order_counter = 0
self.play_order_map = {}
def feed_index(num, parent):
f = feeds[num]
@ -532,13 +846,16 @@ class MPRecipe(BasicNewsRecipe):
desc = None
else:
desc = self.description_limiter(desc)
tt = a.toc_thumbnail if a.toc_thumbnail else None
entries.append('%sindex.html'%adir)
po = self.play_order_map.get(entries[-1], None)
if po is None:
self.play_order_counter += 1
po = self.play_order_counter
parent.add_item('%sindex.html'%adir, None, a.title if a.title else _('Untitled Article'),
play_order=po, author=auth, description=desc)
parent.add_item('%sindex.html'%adir, None,
a.title if a.title else _('Untitled Article'),
play_order=po, author=auth,
description=desc, toc_thumbnail=tt)
last = os.path.join(self.output_dir, ('%sindex.html'%adir).replace('/', os.sep))
for sp in a.sub_pages:
prefix = os.path.commonprefix([opf_path, sp])
@ -555,7 +872,7 @@ class MPRecipe(BasicNewsRecipe):
prefix = '/'.join('..'for i in range(2*len(re.findall(r'link\d+', last))))
templ = self.navbar.generate(True, num, j, len(f),
not self.has_single_feed,
a.orig_url, self.publisher, prefix=prefix,
a.orig_url, __appname__, prefix=prefix,
center=self.center_navbar)
elem = BeautifulSoup(templ.render(doctype='xhtml').decode('utf-8')).find('div')
body.insert(len(body.contents), elem)
@ -578,7 +895,7 @@ class MPRecipe(BasicNewsRecipe):
if not desc:
desc = None
feed_index(i, toc.add_item('feed_%d/index.html'%i, None,
f.title, play_order=po, description=desc, author=auth))
f.title, play_order=po, description=desc, author=auth))
else:
entries.append('feed_%d/index.html'%0)
@ -591,4 +908,5 @@ class MPRecipe(BasicNewsRecipe):
with nested(open(opf_path, 'wb'), open(ncx_path, 'wb')) as (opf_file, ncx_file):
opf.render(opf_file, ncx_file)

604
recipes/ming_pao_vancouver.recipe
View File

@ -4,18 +4,41 @@ __copyright__ = '2010-2011, Eddie Lau'
# Region - Hong Kong, Vancouver, Toronto
__Region__ = 'Vancouver'
# Users of Kindle 3 with limited system-level CJK support
# please replace the following "True" with "False".
# please replace the following "True" with "False". (Default: True)
__MakePeriodical__ = True
# Turn below to true if your device supports display of CJK titles
# Turn below to True if your device supports display of CJK titles (Default: False)
__UseChineseTitle__ = False
# Set it to False if you want to skip images
# Set it to False if you want to skip images (Default: True)
__KeepImages__ = True
# (HK only) Turn below to true if you wish to use life.mingpao.com as the main article source
# Set it to True if you want to include a summary in Kindle's article view (Default: False)
__IncludeSummary__ = False
# Set it to True if you want thumbnail images in Kindle's article view (Default: True)
__IncludeThumbnails__ = True
# (HK only) Turn below to True if you wish to use life.mingpao.com as the main article source (Default: True)
__UseLife__ = True
# (HK only) It is to disable premium content (Default: False)
__InclPremium__ = False
# (HK only) Turn below to True if you wish to parse articles in news.mingpao.com with their printer-friendly formats (Default: True)
__ParsePFF__ = True
# (HK only) Turn below to True if you wish hi-res images (Default: False)
__HiResImg__ = False
# Override the date returned by the program if specifying a YYYYMMDD below
__Date__ = ''
'''
Change Log:
2011/12/18: update the overridden create_odf(.) routine with the one from Calibre version 0.8.31. Move __UseChineseTitle__ usage away
from create_odf(.). Optional support of text_summary and thumbnail images in Kindle's article view. Start new day
download of Hong Kong Mingpao at 4.30am. Set the actual publication date shown on kindle device.
2011/12/01: take care of situation that in txt source parsing, the article content does start with special character u'\u3010'
2011/10/21: fix a bug that hi-res img is unavailable in pages parsed from source txt
2011/10/19: fix a bug in txt source parsing
2011/10/17: disable fetching of premium content, also improved txt source parsing
2011/10/04: option to get hi-res photos for the articles
2011/09/21: fetching "column" section is made optional.
2011/09/18: parse "column" section stuff from source text file directly.
2011/09/07: disable "column" section as it is no longer offered free.
2011/06/26: add fetching Vancouver and Toronto versions of the paper, also provide captions for images using life.mingpao fetch source
provide options to remove all images in the file
2011/05/12: switch the main parse source to life.mingpao.com, which has more photos on the article pages
@ -37,30 +60,39 @@ Change Log:
2010/10/31: skip repeated articles in section pages
'''
import os, datetime, re
from calibre import (browser, iswindows, __appname__, force_unicode, preferred_encoding, as_unicode)
from calibre.utils.date import now as nowf
import os, datetime, re, mechanize
from calibre.web.feeds.recipes import BasicNewsRecipe
from contextlib import nested
from calibre.ebooks.BeautifulSoup import BeautifulSoup
from calibre.ebooks.metadata.opf2 import OPFCreator
from calibre.ebooks.metadata.toc import TOC
from calibre.ebooks.metadata import MetaInformation
from calibre.utils.localization import canonicalize_lang
# MAIN CLASS
class MPRecipe(BasicNewsRecipe):
if __Region__ == 'Hong Kong':
title = 'Ming Pao - Hong Kong'
if __UseChineseTitle__ == True:
title = u'\u660e\u5831 (\u9999\u6e2f)'
else:
title = 'Ming Pao - Hong Kong'
description = 'Hong Kong Chinese Newspaper (http://news.mingpao.com)'
category = 'Chinese, News, Hong Kong'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} font>b {font-size:200%; font-weight:bold;}'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px; max-height:90%;} font>b {font-size:200%; font-weight:bold;} div[class=heading] {font-size:200%; font-weight:bold;} div[class=images] {font-size:50%;}'
masthead_url = 'http://news.mingpao.com/image/portals_top_logo_news.gif'
keep_only_tags = [dict(name='h1'),
dict(name='font', attrs={'style':['font-size:14pt; line-height:160%;']}), # for entertainment page title
dict(name='font', attrs={'color':['AA0000']}), # for column articles title
dict(attrs={'class':['heading']}), # for heading from txt
dict(attrs={'id':['newscontent']}), # entertainment and column page content
dict(attrs={'id':['newscontent01','newscontent02']}),
dict(attrs={'class':['content']}), # for content from txt
dict(attrs={'class':['photo']}),
dict(name='table', attrs={'width':['100%'], 'border':['0'], 'cellspacing':['5'], 'cellpadding':['0']}), # content in printed version of life.mingpao.com
dict(name='img', attrs={'width':['180'], 'alt':['按圖放大']}) # images for source from life.mingpao.com
dict(name='img', attrs={'width':['180'], 'alt':['????']}), # images for source from life.mingpao.com
dict(attrs={'class':['images']}) # for images from txt
]
if __KeepImages__:
remove_tags = [dict(name='style'),
@ -90,7 +122,10 @@ class MPRecipe(BasicNewsRecipe):
lambda match: "</b>")
]
elif __Region__ == 'Vancouver':
title = 'Ming Pao - Vancouver'
if __UseChineseTitle__ == True:
title = u'\u660e\u5831 (\u6eab\u54e5\u83ef)'
else:
title = 'Ming Pao - Vancouver'
description = 'Vancouver Chinese Newspaper (http://www.mingpaovan.com)'
category = 'Chinese, News, Vancouver'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}'
@ -108,7 +143,10 @@ class MPRecipe(BasicNewsRecipe):
lambda match: ''),
]
elif __Region__ == 'Toronto':
title = 'Ming Pao - Toronto'
if __UseChineseTitle__ == True:
title = u'\u660e\u5831 (\u591a\u502b\u591a)'
else:
title = 'Ming Pao - Toronto'
description = 'Toronto Chinese Newspaper (http://www.mingpaotor.com)'
category = 'Chinese, News, Toronto'
extra_css = 'img {display: block; margin-left: auto; margin-right: auto; margin-top: 10px; margin-bottom: 10px;} b>font {font-size:200%; font-weight:bold;}'
@ -139,49 +177,12 @@ class MPRecipe(BasicNewsRecipe):
conversion_options = {'linearize_tables':True}
timefmt = ''
def image_url_processor(cls, baseurl, url):
# trick: break the url at the first occurance of digit, add an additional
# '_' at the front
# not working, may need to move this to preprocess_html() method
# minIdx = 10000
# i0 = url.find('0')
# if i0 >= 0 and i0 < minIdx:
# minIdx = i0
# i1 = url.find('1')
# if i1 >= 0 and i1 < minIdx:
# minIdx = i1
# i2 = url.find('2')
# if i2 >= 0 and i2 < minIdx:
# minIdx = i2
# i3 = url.find('3')
# if i3 >= 0 and i0 < minIdx:
# minIdx = i3
# i4 = url.find('4')
# if i4 >= 0 and i4 < minIdx:
# minIdx = i4
# i5 = url.find('5')
# if i5 >= 0 and i5 < minIdx:
# minIdx = i5
# i6 = url.find('6')
# if i6 >= 0 and i6 < minIdx:
# minIdx = i6
# i7 = url.find('7')
# if i7 >= 0 and i7 < minIdx:
# minIdx = i7
# i8 = url.find('8')
# if i8 >= 0 and i8 < minIdx:
# minIdx = i8
# i9 = url.find('9')
# if i9 >= 0 and i9 < minIdx:
# minIdx = i9
return url
def get_dtlocal(self):
dt_utc = datetime.datetime.utcnow()
if __Region__ == 'Hong Kong':
# convert UTC to local hk time - at HKT 5.30am, all news are available
dt_local = dt_utc + datetime.timedelta(8.0/24) - datetime.timedelta(5.5/24)
# dt_local = dt_utc.astimezone(pytz.timezone('Asia/Hong_Kong')) - datetime.timedelta(5.5/24)
# convert UTC to local hk time - at HKT 4.30am, all news are available
dt_local = dt_utc + datetime.timedelta(8.0/24) - datetime.timedelta(4.5/24)
# dt_local = dt_utc.astimezone(pytz.timezone('Asia/Hong_Kong')) - datetime.timedelta(4.5/24)
elif __Region__ == 'Vancouver':
# convert UTC to local Vancouver time - at PST time 5.30am, all news are available
dt_local = dt_utc + datetime.timedelta(-8.0/24) - datetime.timedelta(5.5/24)
@ -193,13 +194,34 @@ class MPRecipe(BasicNewsRecipe):
return dt_local
def get_fetchdate(self):
return self.get_dtlocal().strftime("%Y%m%d")
if __Date__ <> '':
return __Date__
else:
return self.get_dtlocal().strftime("%Y%m%d")
def get_fetchformatteddate(self):
return self.get_dtlocal().strftime("%Y-%m-%d")
if __Date__ <> '':
return __Date__[0:4]+'-'+__Date__[4:6]+'-'+__Date__[6:8]
else:
return self.get_dtlocal().strftime("%Y-%m-%d")
def get_fetchyear(self):
if __Date__ <> '':
return __Date__[0:4]
else:
return self.get_dtlocal().strftime("%Y")
def get_fetchmonth(self):
if __Date__ <> '':
return __Date__[4:6]
else:
return self.get_dtlocal().strftime("%m")
def get_fetchday(self):
return self.get_dtlocal().strftime("%d")
if __Date__ <> '':
return __Date__[6:8]
else:
return self.get_dtlocal().strftime("%d")
def get_cover_url(self):
if __Region__ == 'Hong Kong':
@ -230,12 +252,23 @@ class MPRecipe(BasicNewsRecipe):
(u'\u570b\u969b World', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalta', 'nal'),
(u'\u7d93\u6fdf Finance', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea', 'nal'),
(u'\u9ad4\u80b2 Sport', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalsp', 'nal'),
(u'\u5f71\u8996 Film/TV', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalma', 'nal'),
(u'\u5c08\u6b04 Columns', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn', 'ncl')]:
articles = self.parse_section2(url, keystr)
(u'\u5f71\u8996 Film/TV', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalma', 'nal')
]:
if __InclPremium__ == True:
articles = self.parse_section2_txt(url, keystr)
else:
articles = self.parse_section2(url, keystr)
if articles:
feeds.append((title, articles))
if __InclPremium__ == True:
# parse column section articles directly from .txt files
for title, url, keystr in [(u'\u5c08\u6b04 Columns', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn', 'ncl')
]:
articles = self.parse_section2_txt(url, keystr)
if articles:
feeds.append((title, articles))
for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'),
(u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]:
articles = self.parse_section(url)
@ -244,15 +277,16 @@ class MPRecipe(BasicNewsRecipe):
else:
for title, url in [(u'\u8981\u805e Headline', 'http://news.mingpao.com/' + dateStr + '/gaindex.htm'),
(u'\u6e2f\u805e Local', 'http://news.mingpao.com/' + dateStr + '/gbindex.htm'),
(u'\u6559\u80b2 Education', 'http://news.mingpao.com/' + dateStr + '/gfindex.htm')]:
(u'\u6559\u80b2 Education', 'http://news.mingpao.com/' + dateStr + '/gfindex.htm'),
(u'\u793e\u8a55/\u7b46\u9663 Editorial', 'http://news.mingpao.com/' + dateStr + '/mrindex.htm')]:
articles = self.parse_section(url)
if articles:
feeds.append((title, articles))
# special- editorial
ed_articles = self.parse_ed_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr')
if ed_articles:
feeds.append((u'\u793e\u8a55/\u7b46\u9663 Editorial', ed_articles))
#ed_articles = self.parse_ed_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=nalmr')
#if ed_articles:
# feeds.append((u'\u793e\u8a55/\u7b46\u9663 Editorial', ed_articles))
for title, url in [(u'\u8ad6\u58c7 Forum', 'http://news.mingpao.com/' + dateStr + '/faindex.htm'),
(u'\u4e2d\u570b China', 'http://news.mingpao.com/' + dateStr + '/caindex.htm'),
@ -263,32 +297,46 @@ class MPRecipe(BasicNewsRecipe):
# special - finance
#fin_articles = self.parse_fin_section('http://www.mpfinance.com/htm/Finance/' + dateStr + '/News/ea,eb,ecindex.htm')
fin_articles = self.parse_fin_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea')
if fin_articles:
feeds.append((u'\u7d93\u6fdf Finance', fin_articles))
#fin_articles = self.parse_fin_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea')
#if fin_articles:
# feeds.append((u'\u7d93\u6fdf Finance', fin_articles))
for title, url in [('Tech News', 'http://news.mingpao.com/' + dateStr + '/naindex.htm'),
(u'\u9ad4\u80b2 Sport', 'http://news.mingpao.com/' + dateStr + '/spindex.htm')]:
articles = self.parse_section(url)
for title, url, keystr in [(u'\u7d93\u6fdf Finance', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalea', 'nal')]:
articles = self.parse_section2_txt(url, keystr)
if articles:
feeds.append((title, articles))
#for title, url in [('Tech News', 'http://news.mingpao.com/' + dateStr + '/naindex.htm'),
# (u'\u9ad4\u80b2 Sport', 'http://news.mingpao.com/' + dateStr + '/spindex.htm')]:
# articles = self.parse_section(url)
# if articles:
# feeds.append((title, articles))
# special - entertainment
ent_articles = self.parse_ent_section('http://ol.mingpao.com/cfm/star1.cfm')
if ent_articles:
feeds.append((u'\u5f71\u8996 Film/TV', ent_articles))
#ent_articles = self.parse_ent_section('http://ol.mingpao.com/cfm/star1.cfm')
#if ent_articles:
# feeds.append((u'\u5f71\u8996 Film/TV', ent_articles))
for title, url, keystr in [(u'\u5f71\u8996 Film/TV', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr + '&Category=nalma', 'nal')
]:
articles = self.parse_section2_txt(url, keystr)
if articles:
feeds.append((title, articles))
if __InclPremium__ == True:
# parse column section articles directly from .txt files
for title, url, keystr in [(u'\u5c08\u6b04 Columns', 'http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn', 'ncl')
]:
articles = self.parse_section2_txt(url, keystr)
if articles:
feeds.append((title, articles))
for title, url in [(u'\u526f\u520a Supplement', 'http://news.mingpao.com/' + dateStr + '/jaindex.htm'),
(u'\u82f1\u6587 English', 'http://news.mingpao.com/' + dateStr + '/emindex.htm')]:
articles = self.parse_section(url)
if articles:
feeds.append((title, articles))
# special- columns
col_articles = self.parse_col_section('http://life.mingpao.com/cfm/dailynews2.cfm?Issue=' + dateStr +'&Category=ncolumn')
if col_articles:
feeds.append((u'\u5c08\u6b04 Columns', col_articles))
elif __Region__ == 'Vancouver':
for title, url in [(u'\u8981\u805e Headline', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VAindex.htm'),
(u'\u52a0\u570b Canada', 'http://www.mingpaovan.com/htm/News/' + dateStr + '/VBindex.htm'),
@ -332,6 +380,16 @@ class MPRecipe(BasicNewsRecipe):
title = self.tag_to_string(a)
url = a.get('href', False)
url = 'http://news.mingpao.com/' + dateStr + '/' +url
# replace the url to the print-friendly version
if __ParsePFF__ == True:
if url.rfind('Redirect') <> -1 and __InclPremium__ == True:
url = re.sub(dateStr + '.*' + dateStr, dateStr, url)
url = re.sub('%2F.*%2F', '/', url)
title = title.replace(u'\u6536\u8cbb\u5167\u5bb9', '')
url = url.replace('%2Etxt', '_print.htm')
url = url.replace('%5F', '_')
else:
url = url.replace('.htm', '_print.htm')
if url not in included_urls and url.rfind('Redirect') == -1:
current_articles.append({'title': title, 'url': url, 'description':'', 'date':''})
included_urls.append(url)
@ -340,6 +398,8 @@ class MPRecipe(BasicNewsRecipe):
# parse from life.mingpao.com
def parse_section2(self, url, keystr):
br = mechanize.Browser()
br.set_handle_redirect(False)
self.get_fetchdate()
soup = self.index_to_soup(url)
a = soup.findAll('a', href=True)
@ -350,12 +410,34 @@ class MPRecipe(BasicNewsRecipe):
title = self.tag_to_string(i)
url = 'http://life.mingpao.com/cfm/' + i.get('href', False)
if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind(keystr) == -1):
url = url.replace('dailynews3.cfm', 'dailynews3a.cfm') # use printed version of the article
try:
br.open_novisit(url)
url = url.replace('dailynews3.cfm', 'dailynews3a.cfm') # use printed version of the article
current_articles.append({'title': title, 'url': url, 'description': ''})
included_urls.append(url)
except:
print 'skipping a premium article'
current_articles.reverse()
return current_articles
# parse from text file of life.mingpao.com
def parse_section2_txt(self, url, keystr):
self.get_fetchdate()
soup = self.index_to_soup(url)
a = soup.findAll('a', href=True)
a.reverse()
current_articles = []
included_urls = []
for i in a:
title = self.tag_to_string(i)
url = 'http://life.mingpao.com/cfm/' + i.get('href', False)
if (url not in included_urls) and (not url.rfind('.txt') == -1) and (not url.rfind(keystr) == -1):
url = url.replace('cfm/dailynews3.cfm?File=', 'ftp/Life3/') # use printed version of the article
current_articles.append({'title': title, 'url': url, 'description': ''})
included_urls.append(url)
current_articles.reverse()
return current_articles
# parse from www.mingpaovan.com
def parse_section3(self, url, baseUrl):
self.get_fetchdate()
@ -438,6 +520,162 @@ class MPRecipe(BasicNewsRecipe):
current_articles.reverse()
return current_articles
# preprocess those .txt and javascript based files
def preprocess_raw_html(self, raw_html, url):
new_html = raw_html
if url.rfind('ftp') <> -1 or url.rfind('_print.htm') <> -1:
if url.rfind('_print.htm') <> -1:
# javascript based file
splitter = re.compile(r'\n')
new_raw_html = '<html><head><title>Untitled</title></head>'
new_raw_html = new_raw_html + '<body>'
for item in splitter.split(raw_html):
if item.startswith('var heading1 ='):
heading = item.replace('var heading1 = \'', '')
heading = heading.replace('\'', '')
heading = heading.replace(';', '')
new_raw_html = new_raw_html + '<div class="heading">' + heading
if item.startswith('var heading2 ='):
heading = item.replace('var heading2 = \'', '')
heading = heading.replace('\'', '')
heading = heading.replace(';', '')
if heading <> '':
new_raw_html = new_raw_html + '<br>' + heading + '</div>'
else:
new_raw_html = new_raw_html + '</div>'
if item.startswith('var content ='):
content = item.replace("var content = ", '')
content = content.replace('\'', '')
content = content.replace(';', '')
new_raw_html = new_raw_html + '<div class="content">' + content + '</div>'
if item.startswith('var photocontent ='):
photo = item.replace('var photocontent = \'', '')
photo = photo.replace('\'', '')
photo = photo.replace(';', '')
photo = photo.replace('<tr>', '')
photo = photo.replace('<td>', '')
photo = photo.replace('</tr>', '')
photo = photo.replace('</td>', '<br>')
photo = photo.replace('class="photo"', '')
new_raw_html = new_raw_html + '<div class="images">' + photo + '</div>'
new_html = new_raw_html + '</body></html>'
else:
# .txt based file
splitter = re.compile(r'\n') # Match non-digits
new_raw_html = '<html><head><title>Untitled</title></head><body><div class="images">'
next_is_img_txt = False
title_started = False
title_break_reached = False
met_article_start_char = False
for item in splitter.split(raw_html):
item = item.strip()
# if title already reached but break between title and content not yet found, record title_break_reached
if title_started == True and title_break_reached == False and item == '':
title_break_reached = True
# if title reached and title_break_reached and met_article_start_char == False and item is not empty
# start content
elif title_started == True and title_break_reached == True and met_article_start_char == False:
if item <> '':
met_article_start_char = True
new_raw_html = new_raw_html + '</div><div class="content"><p>' + item + '<p>\n'
#if item.startswith(u'\u3010'):
# met_article_start_char = True
# new_raw_html = new_raw_html + '</div><div class="content"><p>' + item + '<p>\n'
else:
if next_is_img_txt == False:
if item.startswith("=@"):
print 'skip movie link'
elif item.startswith("=?"):
next_is_img_txt = True
new_raw_html += '<img src="' + str(item)[2:].strip() + '.gif" /><p>\n'
elif item.startswith('=='):
next_is_img_txt = True
if False:
# TODO: check existence of .gif first
newimg = '_' + item[2:].strip() + '.jpg'
new_raw_html += '<img src="' + newimg + '" /><p>\n'
else:
new_raw_html += '<img src="' + str(item)[2:].strip() + '.jpg" /><p>\n'
elif item.startswith('='):
next_is_img_txt = True
if False:
# TODO: check existence of .gif first
newimg = '_' + item[1:].strip() + '.jpg'
new_raw_html += '<img src="' + newimg + '" /><p>\n'
else:
new_raw_html += '<img src="' + str(item)[1:].strip() + '.jpg" /><p>\n'
else:
if next_is_img_txt == False and met_article_start_char == False:
if item <> '':
if title_started == False:
#print 'Title started at ', item
new_raw_html = new_raw_html + '</div><div class="heading">' + item + '\n'
title_started = True
else:
new_raw_html = new_raw_html + item + '\n'
else:
new_raw_html = new_raw_html + item + '<p>\n'
else:
next_is_img_txt = False
new_raw_html = new_raw_html + item + '\n'
new_html = new_raw_html + '</div></body></html>'
#raw_html = raw_html.replace(u'<p>\u3010', u'\u3010')
if __HiResImg__ == True:
# TODO: add a _ in front of an image url
if url.rfind('news.mingpao.com') > -1:
imglist = re.findall('src="?.*?jpg"', new_html)
br = mechanize.Browser()
br.set_handle_redirect(False)
for img in imglist:
gifimg = img.replace('jpg"', 'gif"')
try:
br.open_novisit(url + "/../" + gifimg[5:len(gifimg)-1])
new_html = new_html.replace(img, gifimg)
except:
# find the location of the first _
pos = img.find('_')
if pos > -1:
# if found, insert _ after the first _
newimg = img[0:pos] + '_' + img[pos:]
new_html = new_html.replace(img, newimg)
else:
# if not found, insert _ after "
new_html = new_html.replace(img[1:], '"_' + img[1:])
elif url.rfind('life.mingpao.com') > -1:
imglist = re.findall('src=\'?.*?jpg\'', new_html)
br = mechanize.Browser()
br.set_handle_redirect(False)
#print 'Img list: ', imglist, '\n'
for img in imglist:
#print 'Found img: ', img
gifimg = img.replace('jpg\'', 'gif\'')
try:
gifurl = re.sub(r'dailynews.*txt', '', url)
br.open_novisit(gifurl + gifimg[5:len(gifimg)-1])
new_html = new_html.replace(img, gifimg)
except:
pos = img.rfind('/')
newimg = img[0:pos+1] + '_' + img[pos+1:]
new_html = new_html.replace(img, newimg)
# repeat with src quoted by double quotes, for text parsed from src txt
imglist = re.findall('src="?.*?jpg"', new_html)
for img in imglist:
#print 'Found img: ', img
gifimg = img.replace('jpg"', 'gif"')
try:
#print 'url', url
pos = url.rfind('/')
gifurl = url[:pos+1]
#print 'try it:', gifurl + gifimg[5:len(gifimg)-1]
br.open_novisit(gifurl + gifimg[5:len(gifimg)-1])
new_html = new_html.replace(img, gifimg)
except:
pos = img.find('"')
newimg = img[0:pos+1] + '_' + img[pos+1:]
#print 'Use hi-res img', newimg
new_html = new_html.replace(img, newimg)
return new_html
def preprocess_html(self, soup):
for item in soup.findAll(style=True):
del item['style']
@ -446,78 +684,154 @@ class MPRecipe(BasicNewsRecipe):
for item in soup.findAll(stype=True):
del item['absmiddle']
return soup
def populate_article_metadata(self, article, soup, first):
# thumbnails shouldn't be available if using hi-res images
if __IncludeThumbnails__ and __HiResImg__ == False and first and hasattr(self, 'add_toc_thumbnail'):
img = soup.find('img')
if img is not None:
self.add_toc_thumbnail(article, img['src'])
try:
if __IncludeSummary__ and len(article.text_summary.strip()) == 0:
# look for content
articlebodies = soup.findAll('div',attrs={'id':'newscontent'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'id':'newscontent01'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'class':'content'})
if not articlebodies:
articlebodies = soup.findAll('div', attrs={'id':'font'})
if articlebodies:
for articlebody in articlebodies:
if articlebody:
# the text may or may not be enclosed in <p></p> tag
paras = articlebody.findAll('p')
if not paras:
paras = articlebody
textFound = False
for p in paras:
if not textFound:
summary_candidate = self.tag_to_string(p).strip()
summary_candidate = summary_candidate.replace(u'\u3010\u660e\u5831\u5c08\u8a0a\u3011', '', 1)
if len(summary_candidate) > 0:
article.summary = article.text_summary = summary_candidate
textFound = True
else:
# display a simple text
#article.summary = article.text_summary = u'\u66f4\u591a......'
# display word counts
counts = 0
articlebodies = soup.findAll('div',attrs={'id':'newscontent'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'id':'newscontent01'})
if not articlebodies:
articlebodies = soup.findAll('div',attrs={'class':'content'})
if not articlebodies:
articlebodies = soup.findAll('div', attrs={'id':'font'})
if articlebodies:
for articlebody in articlebodies:
# the text may or may not be enclosed in <p></p> tag
paras = articlebody.findAll('p')
if not paras:
paras = articlebody
for p in paras:
summary_candidate = self.tag_to_string(p).strip()
counts += len(summary_candidate)
article.summary = article.text_summary = u'\uff08' + str(counts) + u'\u5b57\uff09'
except:
self.log("Error creating article descriptions")
return
# override from the one in version 0.8.31
def create_opf(self, feeds, dir=None):
if dir is None:
dir = self.output_dir
if __UseChineseTitle__ == True:
if __Region__ == 'Hong Kong':
title = u'\u660e\u5831 (\u9999\u6e2f)'
elif __Region__ == 'Vancouver':
title = u'\u660e\u5831 (\u6eab\u54e5\u83ef)'
elif __Region__ == 'Toronto':
title = u'\u660e\u5831 (\u591a\u502b\u591a)'
else:
title = self.short_title()
# if not generating a periodical, force date to apply in title
if __MakePeriodical__ == False:
title = self.short_title()
# change 1: allow our own flag to tell if a periodical is to be generated
# also use customed date instead of current time
if __MakePeriodical__ == False or self.output_profile.periodical_date_in_title:
title = title + ' ' + self.get_fetchformatteddate()
if True:
mi = MetaInformation(title, [self.publisher])
mi.publisher = self.publisher
mi.author_sort = self.publisher
if __MakePeriodical__ == True:
mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title()
else:
mi.publication_type = self.publication_type+':'+self.short_title()
#mi.timestamp = nowf()
mi.timestamp = self.get_dtlocal()
mi.comments = self.description
if not isinstance(mi.comments, unicode):
mi.comments = mi.comments.decode('utf-8', 'replace')
#mi.pubdate = nowf()
mi.pubdate = self.get_dtlocal()
opf_path = os.path.join(dir, 'index.opf')
ncx_path = os.path.join(dir, 'index.ncx')
opf = OPFCreator(dir, mi)
# Add mastheadImage entry to <guide> section
mp = getattr(self, 'masthead_path', None)
if mp is not None and os.access(mp, os.R_OK):
from calibre.ebooks.metadata.opf2 import Guide
ref = Guide.Reference(os.path.basename(self.masthead_path), os.getcwdu())
ref.type = 'masthead'
ref.title = 'Masthead Image'
opf.guide.append(ref)
# end of change 1
# change 2: __appname__ replaced by newspaper publisher
__appname__ = self.publisher
mi = MetaInformation(title, [__appname__])
mi.publisher = __appname__
mi.author_sort = __appname__
# change 3: use __MakePeriodical__ flag to tell if a periodical should be generated
if __MakePeriodical__ == True:
mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title()
else:
mi.publication_type = self.publication_type+':'+self.short_title()
#mi.publication_type = 'periodical:'+self.publication_type+':'+self.short_title()
# change 4: in the following, all the nowf() are changed to adjusted time
# This one doesn't matter
mi.timestamp = nowf()
# change 5: skip listing the articles
#article_titles, aseen = [], set()
#for f in feeds:
# for a in f:
# if a.title and a.title not in aseen:
# aseen.add(a.title)
# article_titles.append(force_unicode(a.title, 'utf-8'))
manifest = [os.path.join(dir, 'feed_%d'%i) for i in range(len(feeds))]
manifest.append(os.path.join(dir, 'index.html'))
manifest.append(os.path.join(dir, 'index.ncx'))
#mi.comments = self.description
#if not isinstance(mi.comments, unicode):
# mi.comments = mi.comments.decode('utf-8', 'replace')
#mi.comments += ('\n\n' + _('Articles in this issue: ') + '\n' +
# '\n\n'.join(article_titles))
# Get cover
cpath = getattr(self, 'cover_path', None)
if cpath is None:
pf = open(os.path.join(dir, 'cover.jpg'), 'wb')
if self.default_cover(pf):
cpath = pf.name
if cpath is not None and os.access(cpath, os.R_OK):
opf.cover = cpath
manifest.append(cpath)
language = canonicalize_lang(self.language)
if language is not None:
mi.language = language
# This one affects the pub date shown in kindle title
#mi.pubdate = nowf()
# now appears to need the time field to be > 12.00noon as well
mi.pubdate = datetime.datetime(int(self.get_fetchyear()), int(self.get_fetchmonth()), int(self.get_fetchday()), 12, 30, 0)
opf_path = os.path.join(dir, 'index.opf')
ncx_path = os.path.join(dir, 'index.ncx')
# Get masthead
mpath = getattr(self, 'masthead_path', None)
if mpath is not None and os.access(mpath, os.R_OK):
manifest.append(mpath)
opf = OPFCreator(dir, mi)
# Add mastheadImage entry to <guide> section
mp = getattr(self, 'masthead_path', None)
if mp is not None and os.access(mp, os.R_OK):
from calibre.ebooks.metadata.opf2 import Guide
ref = Guide.Reference(os.path.basename(self.masthead_path), os.getcwdu())
ref.type = 'masthead'
ref.title = 'Masthead Image'
opf.guide.append(ref)
manifest = [os.path.join(dir, 'feed_%d'%i) for i in range(len(feeds))]
manifest.append(os.path.join(dir, 'index.html'))
manifest.append(os.path.join(dir, 'index.ncx'))
# Get cover
cpath = getattr(self, 'cover_path', None)
if cpath is None:
pf = open(os.path.join(dir, 'cover.jpg'), 'wb')
if self.default_cover(pf):
cpath = pf.name
if cpath is not None and os.access(cpath, os.R_OK):
opf.cover = cpath
manifest.append(cpath)
# Get masthead
mpath = getattr(self, 'masthead_path', None)
if mpath is not None and os.access(mpath, os.R_OK):
manifest.append(mpath)
opf.create_manifest_from_files_in(manifest)
for mani in opf.manifest:
if mani.path.endswith('.ncx'):
mani.id = 'ncx'
if mani.path.endswith('mastheadImage.jpg'):
mani.id = 'masthead-image'
entries = ['index.html']
toc = TOC(base_path=dir)
self.play_order_counter = 0
self.play_order_map = {}
opf.create_manifest_from_files_in(manifest)
for mani in opf.manifest:
if mani.path.endswith('.ncx'):
mani.id = 'ncx'
if mani.path.endswith('mastheadImage.jpg'):
mani.id = 'masthead-image'
entries = ['index.html']
toc = TOC(base_path=dir)
self.play_order_counter = 0
self.play_order_map = {}
def feed_index(num, parent):
f = feeds[num]
@ -532,13 +846,16 @@ class MPRecipe(BasicNewsRecipe):
desc = None
else:
desc = self.description_limiter(desc)
tt = a.toc_thumbnail if a.toc_thumbnail else None
entries.append('%sindex.html'%adir)
po = self.play_order_map.get(entries[-1], None)
if po is None:
self.play_order_counter += 1
po = self.play_order_counter
parent.add_item('%sindex.html'%adir, None, a.title if a.title else _('Untitled Article'),
play_order=po, author=auth, description=desc)
parent.add_item('%sindex.html'%adir, None,
a.title if a.title else _('Untitled Article'),
play_order=po, author=auth,
description=desc, toc_thumbnail=tt)
last = os.path.join(self.output_dir, ('%sindex.html'%adir).replace('/', os.sep))
for sp in a.sub_pages:
prefix = os.path.commonprefix([opf_path, sp])
@ -555,7 +872,7 @@ class MPRecipe(BasicNewsRecipe):
prefix = '/'.join('..'for i in range(2*len(re.findall(r'link\d+', last))))
templ = self.navbar.generate(True, num, j, len(f),
not self.has_single_feed,
a.orig_url, self.publisher, prefix=prefix,
a.orig_url, __appname__, prefix=prefix,
center=self.center_navbar)
elem = BeautifulSoup(templ.render(doctype='xhtml').decode('utf-8')).find('div')
body.insert(len(body.contents), elem)
@ -578,7 +895,7 @@ class MPRecipe(BasicNewsRecipe):
if not desc:
desc = None
feed_index(i, toc.add_item('feed_%d/index.html'%i, None,
f.title, play_order=po, description=desc, author=auth))
f.title, play_order=po, description=desc, author=auth))
else:
entries.append('feed_%d/index.html'%0)
@ -591,4 +908,5 @@ class MPRecipe(BasicNewsRecipe):
with nested(open(opf_path, 'wb'), open(ncx_path, 'wb')) as (opf_file, ncx_file):
opf.render(opf_file, ncx_file)