|
From: Thomas W. <tho...@gm...> - 2006-04-28 14:50:27
Attachments:
awgn.patch
|
Hi,
there's a bracket missing in awgn.m. Strangely enough,
make check
fails with Octave 2.9.5 without it.
But even with it, it fails with the following error:
=======================================================================
extra/testfun [tests 3 of 7 files]
warning: warning: setting state with structure not implemented
warning: warning: setting state with structure not implemented
error: regexp: missing terminating ] for character class at position 57
of expression
error: evaluating argument list element number 1
error: if: error evaluating conditional expression
error: evaluating if command near line 628, column 11
error: evaluating if command near line 604, column 7
error: evaluating if command near line 491, column 5
error: evaluating for command near line 461, column 3
error: called from `test' in file
`/tmp/octave-forge-2006.03.17/extra/testfun/test.m'
error: near line 373 of file `fntests.m'
make: *** [check] Error 1
=======================================================================
This means it fails with
[p,n] = test('extra/testfun/assert.m','quiet',fid);
However, if I start Octave myself and run the tests in assert.m,
everything works. In case it's important, I have compiled both Octave
and Octave-Forge with pcre support.
Thomas
|
|
From: David B. <Dav...@mo...> - 2006-04-28 16:29:17
|
Thomas Weber wrote:
>Hi,
>
>there's a bracket missing in awgn.m. Strangely enough,
> make check
>fails with Octave 2.9.5 without it.
>
>But even with it, it fails with the following error:
>=======================================================================
>extra/testfun [tests 3 of 7 files]
>warning: warning: setting state with structure not implemented
>warning: warning: setting state with structure not implemented
>error: regexp: missing terminating ] for character class at position 57
>of expression
>error: evaluating argument list element number 1
>error: if: error evaluating conditional expression
>error: evaluating if command near line 628, column 11
>error: evaluating if command near line 604, column 7
>error: evaluating if command near line 491, column 5
>error: evaluating for command near line 461, column 3
>error: called from `test' in file
>`/tmp/octave-forge-2006.03.17/extra/testfun/test.m'
>error: near line 373 of file `fntests.m'
>
>make: *** [check] Error 1
>=======================================================================
>
>This means it fails with
> [p,n] = test('extra/testfun/assert.m','quiet',fid);
>
>However, if I start Octave myself and run the tests in assert.m,
>everything works. In case it's important, I have compiled both Octave
>and Octave-Forge with pcre support.
>
> Thomas
>
>
This is a buglet certainly, but might need more checking as I'm unsure
what regexp, etc you are using (the one in octave-forge or in octave)..
D.
--
David Bateman Dav...@mo...
Motorola Labs - Paris +33 1 69 35 48 04 (Ph)
Parc Les Algorithmes, Commune de St Aubin +33 6 72 01 06 33 (Mob)
91193 Gif-Sur-Yvette FRANCE +33 1 69 35 77 01 (Fax)
The information contained in this communication has been classified as:
[x] General Business Information
[ ] Motorola Internal Use Only
[ ] Motorola Confidential Proprietary
|
|
From: Thomas W. <tho...@gm...> - 2006-04-30 11:00:00
|
Hi,
Am Freitag, den 28.04.2006, 18:29 +0200 schrieb David Bateman:
> This is a buglet certainly, but might need more checking as I'm unsure
> what regexp, etc you are using (the one in octave-forge or in octave)..
That question is not too difficult to anser: the one from Octave. regexp
from octave-forge is not built, due to the following snippet from the
Makefile in main/strings/:
t2.9.5=regexp.oct str2double.m strmatch.m strcmpi.m
DEPRECIATED_TARGETS=$($(word 2, $(sort t$(OCTAVE_VERSION) t2.9.5)))
PROGS=$(DEPRECIATED_TARGETS)
'sort' filters double entries, so for Octave 2.9.5 DEPRECIATED_TARGETS
is empty.
(The Makefile is from the octave-forge release. Quentin has since
cleaned it up, but regexp won't be built neither).
Is this behaviour intentional (i.e. should Octave's regexp be used for
versions >= 2.9.5)?
Anyway, after building regexp.oct manually, the fntests.m part finishes
(though a small warning regarding [main/sparse] might be in order; I
thought the test failed when it justs takes some time (and > 1 GB
memory) to finish).
The 'batch_test.m' of 'make check' however crashes in the
"comms('test')" part; more precisely, the line
tmp = rand (n, m)
results in
panic: Segmentation fault -- stopping myself...
admin/run_forge: line 56: 19134 Segmentation fault (core dumped) $*
make: *** [check] Error 139
According to strace, rand.oct from octave-forge is used. However, my
debugging knowledge is very limited, so if you need more information,
just ask.
Thomas
|