def extract_text_content_and_links_from_articletext(main_content, has_intro=True):
article_text = main_content
in_text_tagged_urls = []
all_cleaned_paragraphs = []
all_rough_paragraphs = []
all_plaintext_urls = []
embedded_tweets = []
def is_text_content(blob):
if isinstance(blob, bs.Tag) and blob.name in TEXT_MARKUP_TAGS:
return True
if isinstance(blob, bs.NavigableString):
return True
return False
text_fragments = [c for c in article_text.contents if is_text_content(c)]
if text_fragments:
# we first need to avoid treating embedded tweets as text
for paragraph in text_fragments:
if isinstance(paragraph, bs.NavigableString):
all_cleaned_paragraphs.append(remove_text_formatting_markup_from_fragments(paragraph))
all_rough_paragraphs.append(paragraph)
else:
if not paragraph.find("blockquote", {"class": "twitter-tweet"}):
in_text_links = extract_and_tag_in_text_links(paragraph)
in_text_tagged_urls.extend(in_text_links)
all_cleaned_paragraphs.append(remove_text_formatting_markup_from_fragments(paragraph))
all_rough_paragraphs.append(paragraph)
else:
embedded_tweets.extend(
twitter_utils.extract_rendered_tweet(paragraph, DHNET_NETLOC, DHNET_INTERNAL_SITES)
)
# extracting plaintext links
for paragraph in all_rough_paragraphs:
plaintext_urls = extract_plaintext_urls_from_text(
remove_text_formatting_and_links_from_fragments(paragraph)
)
for url in plaintext_urls:
tags = classify_and_tag(url, DHNET_NETLOC, DHNET_INTERNAL_SITES)
tags.update(["plaintext", "in text"])
all_plaintext_urls.append(make_tagged_url(url, url, tags))
else:
all_cleaned_paragraphs = []
return all_cleaned_paragraphs, in_text_tagged_urls + all_plaintext_urls + embedded_tweets
def extract_text_content_and_links(main_content):
article_text = main_content.find('div', {'id': 'articleText'})
in_text_tagged_urls = []
all_rough_paragraphs = []
all_clean_paragraphs = []
all_plaintext_urls = []
embedded_tweets = []
def is_text_content(blob):
if isinstance(blob, BeautifulSoup.Tag) and blob.name in TEXT_MARKUP_TAGS:
return True
if isinstance(blob, BeautifulSoup.NavigableString):
return True
return False
text_fragments = [c for c in article_text.contents if is_text_content(c)]
if text_fragments:
for paragraph in text_fragments:
if isinstance(paragraph, BeautifulSoup.NavigableString):
all_clean_paragraphs.append(remove_text_formatting_markup_from_fragments(paragraph, strip_chars=' \t\r\n'))
all_rough_paragraphs.append(paragraph)
else:
if not paragraph.find('blockquote', {'class': 'twitter-tweet'}):
in_text_links = extract_and_tag_in_text_links(paragraph)
in_text_tagged_urls.extend(in_text_links)
all_clean_paragraphs.append(remove_text_formatting_markup_from_fragments(paragraph, strip_chars=' \t\r\n'))
all_rough_paragraphs.append(paragraph)
else:
embedded_tweets.extend(
twitter_utils.extract_rendered_tweet(paragraph, LALIBRE_NETLOC, LALIBRE_ASSOCIATED_SITES))
for p in all_rough_paragraphs:
plaintext_urls = extract_plaintext_urls_from_text(remove_text_formatting_and_links_from_fragments(p))
for url in plaintext_urls:
tags = classify_and_tag(url, LALIBRE_NETLOC, LALIBRE_ASSOCIATED_SITES)
tags.update(['plaintext', 'in text'])
all_plaintext_urls.append(make_tagged_url(url, url, tags))
else:
all_clean_paragraphs = []
return all_clean_paragraphs, in_text_tagged_urls + all_plaintext_urls + embedded_tweets
