Horton

Re: [HACKERS] Time to change pg_regress diffs to unified by default?

Christoph Berg <myon@debian.org>

View thread View on postgr.es
From: Christoph Berg <myon@debian.org>
To: Andrew Gierth <andrew@tao11.riddles.org.uk>
Cc: Peter Eisentraut <peter.eisentraut@2ndquadrant.com>, Daniel Gustafsson <daniel@yesql.se>, Noah Misch <noah@leadboat.com>, Tom Lane <tgl@sss.pgh.pa.us>, Andres Freund <andres@anarazel.de>, pgsql-hackers@postgresql.org
Date: 2019-02-15T15:05:19Z
Lists: pgsql-hackers 
Re: Andrew Gierth 2019-02-15 <874l95m8w7.fsf@news-spur.riddles.org.uk>
> Also while we're tweaking regression test output, would it be possible
> to have some indicator of whether a test passed because a variant file
> in the resultmap was ignored in favor of the standard result?
> 
> The current system of silently ignoring the resultmap means that nobody
> ever notices when resultmap entries become obsolete....

By the same argument, it should always print which variant file was
used so determining which _N.out files are still in use is possible.

Christoph

Commits

  1. pg_regress: Don't use absolute paths for the diff

  2. Use standard diff separator for regression.diffs