Mercurial > code > home > repos > sco-bot
diff scobot/index/build_index_flow.py @ 13:403eff4a16c8
fix up indexer flow and fastapi server
author | drewp@bigasterisk.com |
---|---|
date | Thu, 11 Jul 2024 21:32:24 -0700 |
parents | 6622bacb0b84 |
children | b9c2b7fedbcd |
line wrap: on
line diff
--- a/scobot/index/build_index_flow.py Thu Jul 11 18:16:20 2024 -0700 +++ b/scobot/index/build_index_flow.py Thu Jul 11 21:32:24 2024 -0700 @@ -71,9 +71,9 @@ pass else: html = getCityPermanent(agendaUrl) - text = extractMeetingText(html) - # todo group phrases phrasesFromFile - index.addDoc(title=f'{mtg["date"]} {mtg["title"]}', content=text) + texts = extractMeetingText(html) + for se in nltk.sent_tokenize(' '.join(texts)): + index.addDoc(sourceTitle=f'{mtg["date"]} {mtg["title"]}', phrase=se) try: videoUrl = mtg['videoUrl'] @@ -102,4 +102,4 @@ if __name__ == '__main__': - buildIndex.serve() + buildIndex.serve(name='buildIndex')