Re: BUG #4397: crash in tab-complete.c - Mailing list pgsql-bugs

From Rudolf Leitgeb
Subject Re: BUG #4397: crash in tab-complete.c
Date
Msg-id B88921C7-C097-438C-8234-7B18613CF009@x-pin.com
Whole thread Raw
In response to Re: BUG #4397: crash in tab-complete.c  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-bugs
Problem solved!

The was a left over readline.h which caused the miscompile process :(
Thanks a lot for all your help !

Cheers,

Rudi

Am 08.09.2008 um 15:03 schrieb Tom Lane:

> Rudolf Leitgeb <r.leitgeb@x-pin.com> writes:
>> Sorry for the delay, here are the compile warnings
>> I got when I compiled tab-complete.c for x86_64:
>
>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D
>> tab-complete.c: In function =E2=80=98psql_completion=E2=80=99:
>> tab-complete.c:601: warning: implicit declaration of function
>> =E2=80=98completion_matches=E2=80=99
>> tab-complete.c:601: warning: assignment makes pointer from integer
>> without a cast
>
> Well, that nicely explains a crash in a 64-bit build, so the question
> is what happened to your declaration of completion_matches().  It's
> certainly there on my Mac:
>
> tgl$ grep completion_matches /usr/include/readline/readline.h
> char           **completion_matches(const char *, CPFunction *);
> $
>
> I'm still suspicious that you've got something bogus under
> /usr/local/include that is masking the correct include file.
>
>             regards, tom lane

pgsql-bugs by date:

Previous
From: Tom Lane
Date:
Subject: Re: BUG #4397: crash in tab-complete.c
Next
From: "Mike Tegtmeyer"
Date:
Subject: BUG #4411: One click install broken