From 78d57c4020aac00ce81e38ce4abc85369aa2c66e Mon Sep 17 00:00:00 2001
From: Andrey Fedorov <andrey.fedorov@gmail.com>
Date: Mon, 25 Nov 2024 12:56:18 -0500
Subject: [PATCH] BUG: ignore manifest lines that do not contain s3 URLs

This came up while trying to use manifests generated using Looker Studio
---
 idc_index/index.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/idc_index/index.py b/idc_index/index.py
index 00f619d..3c055c7 100644
--- a/idc_index/index.py
+++ b/idc_index/index.py
@@ -836,7 +836,9 @@ def _validate_update_manifest_and_get_download_size(
                 REGEXP_EXTRACT(manifest_cp_cmd, '(?:.*?\\/){{3}}([^\\/?#]+)', 1) AS manifest_crdc_series_uuid,
                 REGEXP_EXTRACT(manifest_cp_cmd, 's3://\\S+') AS s3_url,
             FROM
-                manifest_df )
+                manifest_df
+            WHERE
+                REGEXP_EXTRACT(manifest_cp_cmd, 's3://\\S+') IS NOT NULL)
             SELECT
                 seriesInstanceuid,
                 index_crdc_series_uuid,