Re: BUG #18708: regex problem: (?:[^\d\D]){0} asserts with "lp->nouts == 0 && rp->nins == 0"

Aleksander Alekseev <aleksander@timescale.com>

From: Aleksander Alekseev <aleksander@timescale.com>

To: pgsql-bugs@lists.postgresql.org

Cc: Tom Lane <tgl@sss.pgh.pa.us>, dhyan@nataraj.su

Date: 2024-11-15T12:51:08Z

Lists: pgsql-bugs

Hi,

I reviewed, applied and tested patch v2 on Linux x64 and MacOS x64. LGTM.

> I'm not especially in love with the CANTMATCH arc type name, but
> other possibilities such as DUMMY or NOMATCH felt too generic.
> Better ideas anyone?

Well... at least it's consistent with the current naming e.g.
HASCANTMATCH. I suggest keeping it as is.

--
Best regards,
Aleksander Alekseev

Commits

Fix recently-exposed portability issue in regex optimization.
- df1a2633b11a 14.15 landed
- adb6dbc7f535 13.18 landed
- 6ab39c02747c 15.10 landed
Avoid assertion due to disconnected NFA sub-graphs in regex parsing.
- b69bdcee9c9c 18.0 landed
- b6312becc819 16.6 landed
- 5f28e6ba7fe1 17.2 landed
- 2bdd3b248924 14.15 landed
- 2496c3f6f1bf 15.10 landed
Fix recovery conflict SIGUSR1 handling.
- 0da096d78e1e 17.0 cited
Redesign interrupt/cancel API for regex engine.
- db4f21e4a34b 16.0 cited
Use MemoryContext API for regex memory management.
- bea3d7e3831f 16.0 cited
Invent "rainbow" arcs within the regex engine.
- 08c0d6ad65f7 14.0 cited