Horton

Re: libpq debug log

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Alvaro Herrera <alvherre@alvh.no-ip.org>
Cc: Kyotaro Horiguchi <horikyota.ntt@gmail.com>, iwata.aya@fujitsu.com, tsunakawa.takay@fujitsu.com, k.jamison@fujitsu.com, pgsql-hackers@lists.postgresql.org
Date: 2021-04-02T14:04:50Z
Lists: pgsql-hackers 
Alvaro Herrera <alvherre@alvh.no-ip.org> writes:
> As in the attached patch.

+1, but the comment could be more specific.  Maybe like "In regress mode,
suppress the length of ErrorResponse and NoticeResponse messages --- the F
(file name) field, in particular, can vary in length depending on compiler
used."

Actually though ... I recall now that elog.c tries to standardize the F
output by stripping the path part:

		/* keep only base name, useful especially for vpath builds */
		slash = strrchr(filename, '/');
		if (slash)
			filename = slash + 1;

I bet what is happening on drongo is that the compiler has generated a
__FILE__ value that contains backslashes not slashes, and this code
doesn't know how to shorten those.  So maybe instead of lobotomizing
this test, we should fix that.

		slash = strrchr(filename, '/');
+		if (!slash)
+			slash = strrchr(filename, '\\');
		if (slash)
			filename = slash + 1;

			regards, tom lane

Commits

  1. Rename PQtraceSetFlags() to PQsetTraceFlags().

  2. Suppress length of Notice/Error msgs in PQtrace regress mode

  3. Strip file names reported in error messages on Windows, too.

  4. Fix setvbuf()-induced crash in libpq_pipeline

  5. libpq_pipeline: Must strdup(optarg) to avoid crash

  6. Remove setvbuf() call from PQtrace()

  7. Initialize conn->Pfdebug to NULL when creating a connection

  8. Disable force_parallel_mode in libpq_pipeline

  9. libpq_pipeline: add PQtrace() support and tests

  10. Improve PQtrace() output format

  11. Re-simplify management of inStart in pqParseInput3's subroutines.