Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-31 14:33:54 -04:00
Code Issues Packages Projects Releases Wiki Activity

Updated NYTimes

Browse Source
This commit is contained in:
Kovid Goyal 2011-01-11 13:15:29 -07:00
parent 9d22690ce2
commit b61fdf0eac
2 changed files with 48 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
resources/recipes/nytimes.recipe
View File

@ -685,3 +685,28 @@ class NYTimes(BasicNewsRecipe):
divTag.replaceWith(tag)
return soup
def populate_article_metadata(self, article, soup, first):
shortparagraph = ""
try:
if len(article.text_summary.strip()) == 0:
articlebodies = soup.findAll('div',attrs={'class':'articleBody'})
if articlebodies:
for articlebody in articlebodies:
if articlebody:
paras = articlebody.findAll('p')
for p in paras:
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) > 70: #approximately one line of text
article.summary = article.text_summary = shortparagraph + refparagraph
return
else:
shortparagraph = refparagraph + " "
if shortparagraph.strip().find(" ") == -1 and not shortparagraph.strip().endswith(":"):
shortparagraph = shortparagraph + "- "
except:
self.log("Error creating article descriptions")
return

23
resources/recipes/nytimes_sub.recipe
View File

@ -685,4 +685,27 @@ class NYTimes(BasicNewsRecipe):
divTag.replaceWith(tag)
return soup
def populate_article_metadata(self, article, soup, first):
shortparagraph = ""
try:
if len(article.text_summary.strip()) == 0:
articlebodies = soup.findAll('div',attrs={'class':'articleBody'})
if articlebodies:
for articlebody in articlebodies:
if articlebody:
paras = articlebody.findAll('p')
for p in paras:
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) > 70: #approximately one line of text
article.summary = article.text_summary = shortparagraph + refparagraph
return
else:
shortparagraph = refparagraph + " "
if shortparagraph.strip().find(" ") == -1 and not shortparagraph.strip().endswith(":"):
shortparagraph = shortparagraph + "- "
except:
self.log("Error creating article descriptions")
return