Re: Improving the notation for ecpg.addons rules
Tom Lane <tgl@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Michael Paquier <michael@paquier.xyz>
Cc: pgsql-hackers@lists.postgresql.org
Date: 2024-08-19T05:23:58Z
Lists: pgsql-hackers
Michael Paquier <michael@paquier.xyz> writes: > It looks like %replace_line expects all its elements to have one space > between each token, still this is not enforced with a check across its > hardcoded elements? Yeah, I was wondering about that. I wouldn't do it exactly like that, but with a check that the entry gets matched somewhere. The other patchset adds cross-checks that each ecpg.addons entry is used exactly once, but there's no such check for these arrays that are internal to parse.pl. Can't quite decide if it's worth adding. I debugged the patch in this thread by checking that the emitted preproc.y didn't change, but that might not be helpful for changes that are actually intended to change the grammar. regards, tom lane
Commits
-
ecpg: invent a saner syntax for ecpg.addons entries.
- dbedc461b4e7 18.0 landed
-
ecpg: add cross-checks to parse.pl for usage of internal tables.
- d2f41b4621de 18.0 landed