FreshRSS/rssguard
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

scraper

Browse Source
This commit is contained in:
Martin Rotter 2021-05-20 10:14:11 +02:00
parent 487b660c44
commit 5569465daf
1 changed files with 8 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
resources/scripts/scrapers/wiki-qotd.py
View File

@ -22,9 +22,14 @@ for lst in lists:
quote_link = "https://en.wikiquote.org" + quote_link quote_link = "https://en.wikiquote.org" + quote_link
quote_author = last_link.get_text() quote_author = last_link.get_text()
quote_text = lst.find("li").decode_contents()
quote_heading = (quote_text[:75] + '...') if len(quote_text) > 75 else quote_text if not quote_author:
quote_text = "<span>" + quote_text + "</span>" continue
quote_text = "<span>" + lst.find("li").decode_contents() + "</span>"
quote_heading = lst.find("li")
quote_heading = (quote_heading.get_text()[:75] + '...') if len(quote_heading) > 75 else quote_heading.get_text()
quote_heading = quote_heading.split(" ~")[0]
items.append("{{\"title\": {title}, \"authors\": [{{\"name\": {author}}}], \"content_html\": {html}, \"url\": {url}, \"date_published\": {date}}}".format( items.append("{{\"title\": {title}, \"authors\": [{{\"name\": {author}}}], \"content_html\": {html}, \"url\": {url}, \"date_published\": {date}}}".format(
title = json.dumps(quote_heading), title = json.dumps(quote_heading),