MIME-Version: 1.0
References: <CAOYmi+nGbv3pGgGeWG7xTAnYo=gPXkeJ-sBT+kvZS1kfRF7H1Q@mail.gmail.com>
 <7ead3d4d-ddc6-2499-0a04-6b29c693ea4a@dunslane.net> <CAOYmi+mQvywAyFZTfujatvY3u2PbzVeqotaWSdB4ipRndB3FbA@mail.gmail.com>
 <8447d168-b981-2601-8ad0-53827fe18e5a@dunslane.net> <158c19a9-f092-4f37-b555-2b5c5f85ffd8@dunslane.net>
 <549100.1712242815@sss.pgh.pa.us> <562157.1712243788@sss.pgh.pa.us>
 <606171.1712246685@sss.pgh.pa.us> <40e7a22a-426f-4c53-9720-b51a81d81202@dunslane.net>
 <CAGECzQT=De58bxcVcoUTmOUEJZh1J8_xrZgyQ=+T+W2mBGud+A@mail.gmail.com> <20240404180602.GA3878309@nathanxps13>
In-Reply-To: <20240404180602.GA3878309@nathanxps13>
From: Jacob Champion <jacob.champion@enterprisedb.com>
Date: Thu, 4 Apr 2024 12:28:59 -0700
Message-ID: <CAOYmi+nMw+cfOpGiwPBU3p4ykKGenL9andOaosPF8_CUidXYzg@mail.gmail.com>
Subject: Re: WIP Incremental JSON Parser
To: Nathan Bossart <nathandbossart@gmail.com>, Andrew Dunstan <andrew@dunslane.net>
Cc: Jelte Fennema-Nio <postgres@jeltef.nl>, Tom Lane <tgl@sss.pgh.pa.us>, 
	Amit Langote <amitlangote09@gmail.com>, Robert Haas <robertmhaas@gmail.com>, 
	Peter Smith <smithpb2250@gmail.com>, 
	PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: <https://www.postgresql.org/message-id/CAOYmi%2BnMw%2BcfOpGiwPBU3p4ykKGenL9andOaosPF8_CUidXYzg%40mail.gmail.com>
Precedence: bulk

On Thu, Apr 4, 2024 at 11:06=E2=80=AFAM Nathan Bossart <nathandbossart@gmai=
l.com> wrote:
> ../postgresql/src/common/jsonapi.c: In function =E2=80=98IsValidJsonNumbe=
r=E2=80=99:
> ../postgresql/src/common/jsonapi.c:2016:30: error: =E2=80=98dummy_lex.inc=
_state=E2=80=99 may be used uninitialized in this function [-Werror=3Dmaybe=
-uninitialized]
>  2016 |  if (lex->incremental && !lex->inc_state->is_last_chunk &&
>       |                           ~~~^~~~~~~~~~~
> ../postgresql/src/common/jsonapi.c:2020:36: error: =E2=80=98dummy_lex.tok=
en_start=E2=80=99 may be used uninitialized in this function [-Werror=3Dmay=
be-uninitialized]
>  2020 |           lex->token_start, s - lex->token_start);
>       |                                 ~~~^~~~~~~~~~~~~

Ah, it hasn't figured out that the incremental path is disabled.
Zero-initializing the dummy_lex would be good regardless.

> ../postgresql/src/common/jsonapi.c:302:26: error: =E2=80=98numeric_error=
=E2=80=99 may be used uninitialized in this function [-Werror=3Dmaybe-unini=
tialized]
>   302 |  return (!numeric_error) && (total_len =3D=3D dummy_lex.input_len=
gth);
>       |         ~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

For this case, though, I think it'd be good for API consistency if
*numeric_error were always set on return, instead of relying on the
caller to initialize.

--Jacob