def body_to_article(headline, byline, text, date, source, meta):
"""
Create an Article-object based on given parameters. It raises an
error (Medium.DoesNotExist) when the given source does not have
an entry in the database.
@param headline: headline of new Article-object
@type headline: str
@param byline: byline for new Article
@type byline: NoneType, str
@param text: text for new Article
@type text: str
@param date: date(time) for new Article
@type date: datetime.date, datetime.datetime
@param source: medium-label for new Article
@type source: str
@param meta: object containing all sorts of meta-information, most of
it suitable for metastring. However, some information
(author, length) will be extracted.
@type meta: dictionary
@return Article-object
"""
log.debug("Creating article object for {headline!r}".format(**locals()))
art = Article(headline=headline, byline=byline, text=text, date=date)
art.medium = Medium.get_or_create(source)
# Author / Section
meta = meta.copy()
art.author = meta.pop('author', None)
art.section = meta.pop('section', None)
if 'length' in meta:
art.length = int(meta.pop('length').split()[0])
else:
art.length = art.text.count(" ")
if 'url' in meta:
art.url = meta.pop('url')
art.url = re.sub("\s+", "", art.url)
art.metastring = str(meta)
return art
def body_to_article(self, headline, byline, text, date, source, meta):
"""
Create an Article-object based on given parameters. It raises an
error (Medium.DoesNotExist) when the given source does not have
an entry in the database.
@param headline: headline of new Article-object
@type headline: unicode / str
@param byline: byline for new Article
@type byline: NoneType, unicode, str
@param text: text for new Article
@type text: unicode / str
@param date: date(time) for new Article
@type date: datetime.date, datetime.datetime
@param source: medium-label for new Article
@type source: unicode / str
@param meta: object containing all sorts of meta-information, most of
it suitable for metastring. However, some information
(author, length) will be extracted.
@type meta: dictionary
@return Article-object
"""
log.debug(
"Creating article object for {headline!r}".format(**locals()))
art = Article(headline=headline, byline=byline, text=text, date=date)
art.medium = get_or_create(Medium, name=source)
# Author / Section
meta = meta.copy()
art.author = meta.pop('author', None)
art.section = meta.pop('section', None)
if 'length' in meta:
art.length = int(meta.pop('length').split()[0])
else:
art.length = art.text.count(" ")
art.metastring = str(meta)
art.project = self.options['project']
return art
def create_article(self):
"""Convert the document object into an article"""
art = Article()
# All properties in _ARTICLES_PROPS are set on a new Article,
# else in Article.metastring.
_metastring = dict()
for prop, value in self.getprops().items():
value = self._convert(value)
if prop in _ARTICLE_PROPS:
setattr(art, prop, value)
else:
_metastring[prop] = value
art.metastring = str(_metastring)
self.article = art
return art
def create_article(self):
"""Convert the document object into an article"""
art = Article()
# All properties in _ARTICLES_PROPS are set on a new Article,
# else in Article.metastring.
_metastring = dict()
for prop, value in self.getprops().items():
value = self._convert(value)
if prop in _ARTICLE_PROPS:
setattr(art, prop, value)
else:
_metastring[prop] = value
art.metastring = str(_metastring)
self.article = art
return art
def scrape_3(self, _html):
"""Some ugly MS Word format, as of 2014-03-03"""
# Partition articles
part = []
articles = []
for tag in _html.cssselect("body > div > *"):
if tag.cssselect("hr"):
articles.append(part)
part = []
else:
part.append(tag)
for tags in articles[1:]:
article = Article()
dateline = tags[1].text_content().strip()
article = self.parse_dateline(dateline, article)
article.headline = tags[1].text_content().strip()
html_str = "".join([html.tostring(t) for t in tags[2:]])
article.text = html2text(html_str)
article.metastring = {'html': html_str}
yield article
def scrape_3(self, _html):
"""Some ugly MS Word format, as of 2014-03-03"""
# Partition articles
part = []
articles = []
for tag in _html.cssselect("body > div > *"):
if tag.cssselect("hr"):
articles.append(part)
part = []
else:
part.append(tag)
for tags in articles[1:]:
article = Article()
dateline = tags[1].text_content().strip()
article = self.parse_dateline(dateline, article)
article.headline = tags[1].text_content().strip()
html_str = "".join([html.tostring(t) for t in tags[2:]])
article.text = html2text(html_str)
article.metastring = {'html': html_str}
yield article
