def get_first_appearance(page: metadata_parser.MetadataParser,
prop: str) -> str:
"""
obtiene el primer elemento para la propiedad en la pagina pasada
"""
prop_data = page.get_metadatas(prop)
if type(prop_data) is list:
return prop_data[0]
else:
return None
def fetch_president_articles():
from metadata_parser import MetadataParser
created = 0
updated = 0
articles_to_fetch = PresidentCandidateArticle.objects.filter(
information__isnull=True)
for article in articles_to_fetch:
page = MetadataParser(url=article.url)
title = first_or_none(page.get_metadatas('title'))
description = first_or_none(page.get_metadatas('description'))
site = first_or_none(page.get_metadatas('site_name'))
url = page.get_url_canonical()
url = url if url else article.url
image_url = page.get_metadata_link('image')
information_obj, is_created = PresidentCandidateArticleInformation.objects.update_or_create(
article=article,
defaults={
'title': title,
'description': description,
'site': site,
'url': url
})
save_image_from_url(information_obj.image, image_url)
if is_created:
created += 1
else:
updated += 1
return {'created': created, 'updated': updated}
