[tika] branch branch_1x updated: TIKA-3253 -- add report for files missing in B

[ta...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

tallison pushed a commit to branch branch_1x
in repository https://gitbox.apache.org/repos/asf/tika.git


The following commit(s) were added to refs/heads/branch_1x by this push:
     new 84eb396  TIKA-3253 -- add report for files missing in B
84eb396 is described below

commit 84eb3963264818ab4f183681ed28977f6fbde32e
Author: tballison <ta...@apache.org>
AuthorDate: Fri Dec 18 12:02:48 2020 -0500

    TIKA-3253 -- add report for files missing in B
---
 tika-eval/src/main/resources/comparison-reports.xml | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/tika-eval/src/main/resources/comparison-reports.xml b/tika-eval/src/main/resources/comparison-reports.xml
index e23ec5e..b31f46e 100644
--- a/tika-eval/src/main/resources/comparison-reports.xml
+++ b/tika-eval/src/main/resources/comparison-reports.xml
@@ -1532,6 +1532,17 @@
             order by cnt desc
         </sql>
     </report>
+    <report reportName="Container files missing in B details"
+            reportFilename="attachments/container_files_missing_in_B_details.xlsx"
+            format="xlsx"
+            includeSql="true">
+        <sql>
+            select pa.file_name from profiles_a pa
+            left join profiles_b pb on pa.id=pb.id
+            where pb.id is null
+            and pa.is_embedded = false
+        </sql>
+    </report>
     <report reportName="Embedded files missing in B by Mime"
             reportFilename="attachments/embedded_files_missing_in_B_by_mime.xlsx"
             format="xlsx"