pgsql: Improve pg_regress's error reporting for schedule-file problems.

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: pgsql-committers(at)postgresql(dot)org
Subject: pgsql: Improve pg_regress's error reporting for schedule-file problems.
Date: 2017-10-07 22:04:32
Message-ID: E1e0xCy-0000OF-6B@gemulon.postgresql.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-committers

Improve pg_regress's error reporting for schedule-file problems.

The previous coding here trashed the line buffer as it scanned it,
making it impossible to print the source line in subsequent error
messages. With a few save/restore/strdup pushups we can improve
that situation.

In passing, move the free'ing of the various strings that are collected
while processing one set of tests down to the bottom of the loop.
That's simpler, less surprising, and should make valgrind less unhappy
about the strings that were previously leaked by the last iteration.

Branch
------
master

Details
-------
https://git.postgresql.org/pg/commitdiff/b11f0d36b224a9673863b4e592f40f179dba3016

Modified Files
--------------
src/test/regress/pg_regress.c | 62 +++++++++++++++++++++++++------------------
1 file changed, 36 insertions(+), 26 deletions(-)

Browse pgsql-committers by date

  From Date Subject
Next Message Peter Geoghegan 2017-10-07 23:22:08 Re: [COMMITTERS] pgsql: Fix freezing of a dead HOT-updated tuple
Previous Message Tom Lane 2017-10-07 21:20:16 pgsql: Enforce our convention about max number of parallel regression t