ecpg regression broken on mingw

On Thu, Jun 14, 2007 at 07:47:58AM -0400, Andrew Dunstan wrote:

The ECPG regression tests appear to be broken on MinGW (see buildfarm
trout and vaquita). Please tell me that this isn't a line ending problem
like it appears on cursory inspection.

It certainly looks like that, but yak is also a mingw machine, no? And it
shows green (and yes, it has pulled the changes in the regression tests
already)

//Magnus

Magnus Hagander wrote:

On Thu, Jun 14, 2007 at 07:47:58AM -0400, Andrew Dunstan wrote:

The ECPG regression tests appear to be broken on MinGW (see buildfarm
trout and vaquita). Please tell me that this isn't a line ending problem
like it appears on cursory inspection.

It certainly looks like that, but yak is also a mingw machine, no? And it
shows green (and yes, it has pulled the changes in the regression tests
already)

Maybe it uses a different CVS or something. Who knows? Aren't our diffs
supposed to be done ignoring whitespace?

cheers

andrew

On Thu, Jun 14, 2007 at 08:17:05AM -0400, Andrew Dunstan wrote:

Maybe it uses a different CVS or something. Who knows? Aren't our diffs
supposed to be done ignoring whitespace?

pg_init() in pg_regress_ecpg.c has:

/* no reason to set -w for ecpg checks, except for when on windows */
if (strstr(host_platform, "-win32"))
basic_diff_opts = "-w";
else
basic_diff_opts = "";

What value does host_platform have on MinGW?

Joachim

On Thu, Jun 14, 2007 at 02:28:26PM +0200, Joachim Wieland wrote:

On Thu, Jun 14, 2007 at 08:17:05AM -0400, Andrew Dunstan wrote:

Maybe it uses a different CVS or something. Who knows? Aren't our diffs
supposed to be done ignoring whitespace?

pg_init() in pg_regress_ecpg.c has:

/* no reason to set -w for ecpg checks, except for when on windows */
if (strstr(host_platform, "-win32"))
basic_diff_opts = "-w";
else
basic_diff_opts = "";

What value does host_platform have on MinGW?

That could certainly be it - on the failing machine, it's:
i686-pc-mingw32

Dave, any chance you can change that check to look for both -win32 and
-mingw32 and give it a test run on vaquita?

(I'll see if I have a vmware around that can run the build meanwhile, but
if it's easy for you to do..)

//Magnus

On Thu, Jun 14, 2007 at 02:35:29PM +0200, Magnus Hagander wrote:

On Thu, Jun 14, 2007 at 02:28:26PM +0200, Joachim Wieland wrote:

On Thu, Jun 14, 2007 at 08:17:05AM -0400, Andrew Dunstan wrote:

Maybe it uses a different CVS or something. Who knows? Aren't our diffs
supposed to be done ignoring whitespace?

pg_init() in pg_regress_ecpg.c has:

/* no reason to set -w for ecpg checks, except for when on windows */
if (strstr(host_platform, "-win32"))
basic_diff_opts = "-w";
else
basic_diff_opts = "";

What value does host_platform have on MinGW?

That could certainly be it - on the failing machine, it's:
i686-pc-mingw32

Dave, any chance you can change that check to look for both -win32 and
-mingw32 and give it a test run on vaquita?

(I'll see if I have a vmware around that can run the build meanwhile, but
if it's easy for you to do..)

This seems to fix most of the problems - I still see 3 tests failing. I'll
commit this change for now.

The interesting thing is that the failed ones seem to *still* be because of
line endings...

//Magnus

On Thu, Jun 14, 2007 at 03:08:00PM +0200, Magnus Hagander wrote:

On Thu, Jun 14, 2007 at 02:35:29PM +0200, Magnus Hagander wrote:

On Thu, Jun 14, 2007 at 02:28:26PM +0200, Joachim Wieland wrote:

On Thu, Jun 14, 2007 at 08:17:05AM -0400, Andrew Dunstan wrote:

Maybe it uses a different CVS or something. Who knows? Aren't our diffs
supposed to be done ignoring whitespace?

pg_init() in pg_regress_ecpg.c has:

/* no reason to set -w for ecpg checks, except for when on windows */
if (strstr(host_platform, "-win32"))
basic_diff_opts = "-w";
else
basic_diff_opts = "";

What value does host_platform have on MinGW?

That could certainly be it - on the failing machine, it's:
i686-pc-mingw32

Dave, any chance you can change that check to look for both -win32 and
-mingw32 and give it a test run on vaquita?

(I'll see if I have a vmware around that can run the build meanwhile, but
if it's easy for you to do..)

This seems to fix most of the problems - I still see 3 tests failing. I'll
commit this change for now.

The interesting thing is that the failed ones seem to *still* be because of
line endings...

Yeah yeah, spammer here, 'eh ;-)

It's not line-ending problem anymore. It's the different float formats on
win32. I looked at the wrong diff file.

I think what's left now is just that mingw needs a different output file
with a different float format in it, just like for the other regression
tests. I'll try to make that happen.

//Magnus