in genai-for-marketing/backend_apis/app/utils_trendspotting.py [0:0]
def _get_documents(articles: dict) -> list[dict]:
"""Gets a list of documents from a list of articles.
Args:
articles:
A list of articles.
Returns:
A list of documents.
"""
documents = []
unique_docs = set()
for article in articles['articles']:
parsed_article = _parse_article(article['url'])
if (parsed_article and parsed_article.text and
(article['title'] not in unique_docs)):
unique_docs.add(article['title'])
document = {
'page_content': parsed_article.text,
'title': article['title'],
'url': article['url'],
'domain': article['domain'],
'date': article['seendate']
}
documents.append(document)
return documents