Horton
Menu

Re: off-by-one in pg_repack index loop

Xiaopeng Wang <wxp_728@163.com>

View thread
From: Xiaopeng Wang <wxp_728@163.com>
To: Lakshmi N <lakshmin.jhs@gmail.com>, "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>
Date: 2026-04-15T02:07:51Z
Lists: pgsql-hackers 
在 2026/4/14 15:34, Lakshmi N 写道:
> Hi hackers,
>
> foreach_current_index returns a 0-based index so
> list_length also takes a 0-based index. Perhaps the
> check should be list_length(ind_oids_new) <= pos?
> Please find the attached patch.
>
> int pos = foreach_current_index(ind_old);
>
> if (unlikely(list_length(ind_oids_new) < pos))
>      elog(ERROR, "list of new indexes too short");
> ident_idx_new = list_nth_oid(ind_oids_new, pos);
> break;
>
> Regards,
> Lakshmi

This fix looks correct to me. Both foreach_current_index(), so 0<=pos<list_length is expected.

Regard,
Xiaopeng Wang

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Fix off-by-one in repack index loop

  2. Improve compiler code layout in elog/ereport ERROR calls