def prep_page_content(article_id, article, wiki, photos, user_obj):
"""
Prepare the formatted article content
:param article_id: int; article id
:param article: str; article name
:param wiki_resp: wikipedia; mediawiki api response
:param photos: list; list of photo json
:param user_obj: User; user object for request
:return: dict; formatted page response passed to jinja template
"""
html = parse_strip_elements(wiki.html())
html = parse_convert_links(html)
html = add_formatting_generic(html)
photo_ids = process_photos(article_id, photos, user_obj)
html = handle_photo_integrate(photos, html, article)
page_content = {
'title': format_title_link(wiki.title, article),
'content': html,
'section_img_class': settings.SECTION_IMG_CLASS,
'num_photos': len(photos),
'article_id': article_id,
'user_id': user_obj.get_id(),
'photo_ids': photo_ids
}
return page_content
def process(self, article):
"""
Freshens the redis entry for `article`
"""
DataIORedis().connect()
key = hmac(article)
# Get wiki
try:
wiki = wikipedia.WikipediaPage(article, preload=True)
except DisambiguationError as e:
# choose a disambiguation
return
except PageError as e:
# bad page proceed with crawl
return
# extract & parse html
html = parse_strip_elements(wiki.html())
html = parse_convert_links(html)
# Get flickr content
res = flickr.call('photos_search', {'text': article,
'format': 'json',
'sort': 'relevance',
})
# TODO - detect failed responses
res_json = json.loads(res[14:-1])
# Extract data for the first photo returned
owner = res_json['photos']['photo'][0]['owner']
photo_id = res_json['photos']['photo'][0]['id']
farm = res_json['photos']['photo'][0]['farm']
server = res_json['photos']['photo'][0]['server']
title = res_json['photos']['photo'][0]['title']
secret = res_json['photos']['photo'][0]['secret']
page_content = {
'content': html,
'owner': owner,
'photo_id': photo_id,
'farm': farm,
'server': server,
'title': title,
'secret': secret
}
DataIORedis().write(key, json.dumps(page_content))
# return the links
return wiki.links
