Thread: pgsql: Prevent redeclaration of typedef TocEntry.
Prevent redeclaration of typedef TocEntry. Commit 9c02e3a986 added a forward declaration for this typedef that caused redeclarations, which is not valid in C99. To fix, add some preprocessor guards to avoid a redefinition, as is done elsewhere (e.g., commit 382092a0cd). Per buildfarm. Branch ------ master Details ------- https://git.postgresql.org/pg/commitdiff/8ec0aaeae09482925d2d15ce4a91f6953bdb1566 Modified Files -------------- src/bin/pg_dump/pg_backup.h | 3 +++ src/bin/pg_dump/pg_backup_archiver.h | 3 +++ 2 files changed, 6 insertions(+)
On 2025-Apr-04, Nathan Bossart wrote: > Prevent redeclaration of typedef TocEntry. > > Commit 9c02e3a986 added a forward declaration for this typedef that > caused redeclarations, which is not valid in C99. To fix, add some > preprocessor guards to avoid a redefinition, as is done elsewhere > (e.g., commit 382092a0cd). Hmm, I have the impression that this results from some very old untidyness in the pg_dump headers in general. The fact that DataDumperPtr is in pg_backup.h seems now more of an accident than something we would really choose to do nowadays -- it's been that way since pg_dump was introduced in commit 500b62b0570f apparently. I'm sure there were good reasons to have it there then. But we've restructured the pg_dump headers at least once, and as far as I can see there aren't anymore. AFAICS we could move both DefnDumperPtr and DataDumperPtr typedefs to pg_backup_archiver.h, together with struct _tocEntry (the only place where they are used), and we'd have less of a mess here. Then we don't need struct TocEntry in pg_backup.h anymore. Plus, these function typedefs aren't really "public interface" for pg_dump anyway (which is what pg_backup.h claims to be), so I think it would be a sensible cleanup. Patch attached. -- Álvaro Herrera PostgreSQL Developer — https://www.EnterpriseDB.com/ Thou shalt check the array bounds of all strings (indeed, all arrays), for surely where thou typest "foo" someone someday shall type "supercalifragilisticexpialidocious" (5th Commandment for C programmers)
Attachment
=?utf-8?Q?=C3=81lvaro?= Herrera <alvherre@alvh.no-ip.org> writes:
> ... AFAICS we could move both DefnDumperPtr and
> DataDumperPtr typedefs to pg_backup_archiver.h, together with struct
> _tocEntry (the only place where they are used), and we'd have less of a
> mess here. Then we don't need struct TocEntry in pg_backup.h anymore.
+1 for this simplification.
regards, tom lane
Sorry, I missed this one last night. On Sat, Apr 05, 2025 at 02:10:10AM -0400, Tom Lane wrote: > =?utf-8?Q?=C3=81lvaro?= Herrera <alvherre@alvh.no-ip.org> writes: >> ... AFAICS we could move both DefnDumperPtr and >> DataDumperPtr typedefs to pg_backup_archiver.h, together with struct >> _tocEntry (the only place where they are used), and we'd have less of a >> mess here. Then we don't need struct TocEntry in pg_backup.h anymore. > > +1 for this simplification. This is a better fix. Thanks for taking care of it. -- nathan