Merge pull request #116 from ImagingDataCommons/fix-s3-url

BUG: use trim to remove any extraneous spaces while parsing s3 url in manifest

Author: fedorov

idc_index/index.py

@@ -665,7 +665,7 @@ def _validate_update_manifest_and_get_download_size(
             SELECT
                 manifest_cp_cmd,
                 REGEXP_EXTRACT(manifest_cp_cmd, '(?:.*?\\/){{3}}([^\\/?#]+)', 1) AS manifest_crdc_series_uuid,
-                REGEXP_REPLACE(regexp_replace(manifest_cp_cmd, 'cp ', ''), '\\s[^\\s]*$', '') AS s3_url,
+                REGEXP_EXTRACT(manifest_cp_cmd, 's3://\\S+') AS s3_url,
             FROM
                 manifest_df )
             SELECT