add excerpt and scraper details.

This commit is contained in:
Jason Schwarzenberger
2020-12-03 16:41:27 +13:00
parent d4260feb72
commit 2a2bf4d671
6 changed files with 76 additions and 14 deletions

View File

@@ -76,14 +76,14 @@ def get_article(url):
if scraper not in scrapers.keys():
continue
try:
html = scrapers[scraper].get_html(url)
if html:
return html
details = scrapers[scraper].get_details(url)
if details and details.get('content'):
return details, scraper
except KeyboardInterrupt:
raise
except:
pass
return ''
return None, None
def get_content_type(url):
try:
@@ -143,7 +143,12 @@ def update_story(story, is_manual=False, urlref=None):
return False
logging.info('Getting article ' + story['url'])
story['text'] = get_article(story['url'])
details, scraper = get_article(story['url'])
if not details: return False
story['text'] = details.get('content', '')
story['excerpt'] = details.get('excerpt', '')
story['scraper'] = scraper
story['scraper_link'] = details.get('scraper_link', '')
if not story['text']: return False
return True

View File

@@ -12,9 +12,28 @@ def get_html(url):
details = get_details(url)
if not details:
return ''
return details['html']
return details['content']
def get_details(url):
outline = _get_outline(url)
if not outline:
return None
return as_readable(outline)
def as_readable(details):
readable = {
'title': details['title'],
'byline': details['author'],
'content': details['html'],
'excerpt': _excerpt(details),
'siteName': details['site_name'],
'url': details['article_url'],
'publisher': details['site_name'],
'scraper_link': 'https://outline.com/' + details['short_code']
}
return readable
def _get_outline(url):
try:
logging.info(f"Outline Scraper: {url}")
params = {'source_url': url}
@@ -34,4 +53,11 @@ def get_details(url):
raise
except BaseException as e:
logging.error('Problem outlining article: {}'.format(str(e)))
return None
return None
def _excerpt(details):
meta = details.get('meta')
if not meta: return ''
if meta.get('description'): return meta.get('description', '')
if not meta.get('og'): return ''
return meta.get('og').get('og:description', '')