]> granicus.if.org Git - postgresql/commit
Unbreak index optimization for LIKE on bytea
authorPeter Eisentraut <peter@eisentraut.org>
Mon, 15 Apr 2019 07:26:23 +0000 (09:26 +0200)
committerPeter Eisentraut <peter@eisentraut.org>
Mon, 15 Apr 2019 07:29:17 +0000 (09:29 +0200)
commitabb9c63b2c00109183cdb9f588d2f23fc383ed38
tree0ad13c544a03873ed8c5ec0ec08d717e9fd0b62d
parentc34677fdaa73f089d557554a9cd479b9bd5b5143
Unbreak index optimization for LIKE on bytea

The same code is used to handle both text and bytea, but bytea is not
collation-aware, so we shouldn't call get_collation_isdeterministic()
in that case, since that will error out with an invalid collation.

Reported-by: Jeevan Chalke <jeevan.chalke@enterprisedb.com>
Discussion: https://www.postgresql.org/message-id/flat/CAM2%2B6%3DWaf3qJ1%3DyVTUH8_yG-SC0xcBMY%2BSFLhvKKNnWNXSUDBw%40mail.gmail.com
src/backend/utils/adt/like_support.c
src/test/regress/expected/strings.out
src/test/regress/sql/strings.sql