buildfarm failure in XML code

Started by Alvaro Herreraalmost 19 years ago7 messages
#1Alvaro Herrera
alvherre@commandprompt.com

UX:acomp: ERROR: "xml.c", line 2188: undefined symbol: INT64_MAX
UX:acomp: ERROR: "xml.c", line 2188: undefined symbol: INT64_MIN

http://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=warthog&dt=2007-02-16%2009:06:01

--
Alvaro Herrera http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.

#2Peter Eisentraut
peter_e@gmx.net
In reply to: Alvaro Herrera (#1)
Re: buildfarm failure in XML code

Am Freitag, 16. Februar 2007 14:59 schrieb Alvaro Herrera:

UX:acomp: ERROR: "xml.c", line 2188: undefined symbol: INT64_MAX
UX:acomp: ERROR: "xml.c", line 2188: undefined symbol: INT64_MIN

http://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=warthog&dt=2007-02-1
6%2009:06:01

It needs a rebuild after the fix.

--
Peter Eisentraut
http://developer.postgresql.org/~petere/

#3Tom Lane
tgl@sss.pgh.pa.us
In reply to: Peter Eisentraut (#2)
Re: buildfarm failure in XML code

Peter Eisentraut <peter_e@gmx.net> writes:

Am Freitag, 16. Februar 2007 14:59 schrieb Alvaro Herrera:

UX:acomp: ERROR: "xml.c", line 2188: undefined symbol: INT64_MAX
UX:acomp: ERROR: "xml.c", line 2188: undefined symbol: INT64_MIN

http://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=warthog&amp;dt=2007-02-1
6%2009:06:01

It needs a rebuild after the fix.

This "fix" doesn't fix anything. We have always in the past managed to
avoid assuming that int64 actually is available; I don't intend to give
the xml code a free pass to break that, especially for such an utterly
marginal purpose as this code has. I'm also wondering why xml.c is the
only place anywhere in the code that uses <stdint.h>.

regards, tom lane

#4Peter Eisentraut
peter_e@gmx.net
In reply to: Tom Lane (#3)
Re: buildfarm failure in XML code

Am Freitag, 16. Februar 2007 16:09 schrieb Tom Lane:

This "fix" doesn't fix anything.

Sure, it fixes the compilation failures.

We have always in the past managed to
avoid assuming that int64 actually is available; I don't intend to give
the xml code a free pass to break that,

I don't intend that either, but a proper fix will be more elaborate. I will
work on that.

--
Peter Eisentraut
http://developer.postgresql.org/~petere/

#5Tom Lane
tgl@sss.pgh.pa.us
In reply to: Peter Eisentraut (#4)
Re: buildfarm failure in XML code

Peter Eisentraut <peter_e@gmx.net> writes:

Am Freitag, 16. Februar 2007 16:09 schrieb Tom Lane:

This "fix" doesn't fix anything.

Sure, it fixes the compilation failures.

Not here:

gcc -O1 -Wall -Wmissing-prototypes -Wpointer-arith -Winline -fno-strict-aliasing -g -I../../../../src/include -D_XOPEN_SOURCE_EXTENDED -c -o xml.o xml.c
xml.c: In function `map_sql_type_to_xmlschema_type':
xml.c:2192: `INT64_MAX' undeclared (first use in this function)
xml.c:2192: (Each undeclared identifier is reported only once
xml.c:2192: for each function it appears in.)
xml.c:2192: `INT64_MIN' undeclared (first use in this function)
make[4]: *** [xml.o] Error 1

And this is in a build WITHOUT xml enabled --- why is this code being
compiled at all? Kindly fix.

regards, tom lane

#6Peter Eisentraut
peter_e@gmx.net
In reply to: Tom Lane (#5)
Re: buildfarm failure in XML code

Tom Lane wrote:

Not here:

Fixed.

And this is in a build WITHOUT xml enabled --- why is this code being
compiled at all?

The enablement pertains to libxml, which this code doesn't use.

--
Peter Eisentraut
http://developer.postgresql.org/~petere/

#7Tom Lane
tgl@sss.pgh.pa.us
In reply to: Peter Eisentraut (#6)
Re: buildfarm failure in XML code

Peter Eisentraut <peter_e@gmx.net> writes:

Tom Lane wrote:

Not here:

Fixed.

Looks good, thanks.

regards, tom lane