Mercurial > public > finance-parser

diff analyze_document/app.py @ 9:bf19235a9636

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
minor bugs and add sample reports
author Dennis C. M. <dennis@denniscm.com>
date Wed, 07 Jun 2023 11:07:14 +0100
parents d15ccf5f1373
children 2350662483a3
line wrap: on
line diff
--- a/analyze_document/app.py	Mon Jun 05 20:07:22 2023 +0100
+++ b/analyze_document/app.py	Wed Jun 07 11:07:14 2023 +0100
@@ -14,7 +14,7 @@
     object_key = event_detail['object']['key']
 
     company_ticker = re.search('unprocessed/(.*)_', object_key).group(1)
-    doc_type = re.search(f'unprocessed/{company_ticker}_(.*).pdf', object_key).group(1)
+    doc_type = re.search(f'unprocessed/{company_ticker}_(.*)_', object_key).group(1)
     file_id = str(uuid.uuid4())
 
     data_dict = textract_client.analyze_document(