Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On 31.08.13 Hilmar Preusse (hill...@web.de) wrote: Hi, I imported the 2 changed files into my texinfo 5.1 build dir. Now the test suite fails even more often. I'm afraid we have to use an svn checkout...unless upstream prepares a release ASAP and we disable the test suite until then. Norbert? http://lists.gnu.org/archive/html/bug-texinfo/2013-09/msg0.html snip On Sat, Aug 31, 2013 at 11:40:43PM +0200, Hilmar Preusse wrote: I simply tried to import the two changed files into a texinfo 5.1 build tree. The test suite fails even more severe. Are there plans to prepare a news texinfo release? It is planned soon after the addition of 5 more @inline* commands. snap So, my suggestion would be to disable the test suite (or the two single tests if possible) and then wait for the release. H. -- sigmentation fault signature.asc Description: Digital signature
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
Hi Hilmar, On Mi, 04 Sep 2013, Hilmar Preusse wrote: I imported the 2 changed files into my texinfo 5.1 build dir. Now the test suite fails even more often. I'm afraid we have to use an svn checkout...unless upstream prepares a release ASAP and we disable the test suite until then. I think we wait for a new release ... So, my suggestion would be to disable the test suite (or the two single tests if possible) and then wait for the release. is there an urgent need to fix it *now*? Norbert PREINING, Norbert http://www.preining.info JAIST, Japan TeX Live Debian Developer DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On Do, 05 Sep 2013, Norbert Preining wrote: I imported the 2 changed files into my texinfo 5.1 build dir. Now the test suite fails even more often. I'm afraid we have to use an svn Hmm, you did something wrong. I guess you used the wrong diff. I have imported the changes from the two svn revisions, the diff stat is tp/Texinfo/Convert/HTML.pm |2 tp/init/chm.pm |2 tp/tests/formatting/res_parser/unknown_nodes_renamed/unknown_nodes_renamed.2 |2 tp/tests/layout/res_parser/formatting_chm/formatting.hhk | 352 +- tp/tests/sectioning/res_parser/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2 |2 tp/tests/sectioning/res_parser_html/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2 |2 6 files changed, 181 insertions(+), 181 deletions(-) With that, I manage to come much further, all tests *but* one *new* fails: Making check in Pod-Simple-Texinfo make[2]: Entering directory `/tmp/buildd/texinfo-5.1.dfsg.1/Pod-Simple-Texinfo' /usr/bin/make check-TESTS make[3]: Entering directory `/tmp/buildd/texinfo-5.1.dfsg.1/Pod-Simple-Texinfo' make[4]: Entering directory `/tmp/buildd/texinfo-5.1.dfsg.1/Pod-Simple-Texinfo' FAIL: prove.sh make[5]: Entering directory `/tmp/buildd/texinfo-5.1.dfsg.1/Pod-Simple-Texinfo' make[5]: Nothing to be done for `all'. make[5]: Leaving directory `/tmp/buildd/texinfo-5.1.dfsg.1/Pod-Simple-Texinfo' Testsuite summary for GNU Texinfo 5.1 # TOTAL: 1 # PASS: 0 # SKIP: 0 # XFAIL: 0 # FAIL: 1 # XPASS: 0 # ERROR: 0 So I guess there are other necessary fixes, too Norbert PREINING, Norbert http://www.preining.info JAIST, Japan TeX Live Debian Developer DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On Do, 05 Sep 2013, Norbert Preining wrote: Making check in Pod-Simple-Texinfo Found the fix for that, too. Build succeeded in clean chroot now. Uploading soon. Norbert PREINING, Norbert http://www.preining.info JAIST, Japan TeX Live Debian Developer DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On 31.08.13 Hilmar Preusse (hill...@web.de) wrote:
Hi,
I reported upstream @texinfo and got the following information:
This is already solved in svn in r5308 and r5298. Thanks for the
report nevertheless!
I imported the 2 changed files into my texinfo 5.1 build dir. Now
the test suite fails even more often. I'm afraid we have to use an
svn checkout...unless upstream prepares a release ASAP and we
disable the test suite until then.
I just noticed that I missed a few files, which were part of these
two svn commits. But even after applying them the tests still fails.
Attached is the output of svn diff -r 5307:5308 svn diff -r
5297:5298. Did not test if they apply to 5.1.
No further ideas, sorry.
H.
--
sigmentation fault
Index:
tp/tests/sectioning/res_parser_html/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2
===
---
tp/tests/sectioning/res_parser_html/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2
(revision 5297)
+++
tp/tests/sectioning/res_parser_html/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2
(revision 5298)
@@ -7,7 +7,7 @@
equivalent_nodes-noderename.cnf:8: warning: no node to be renamed
equivalent_nodes-noderename.cnf:10: warning: nodes without a new name at the
end of file
equivalent_nodes.texi: warning: must specify a title with a title command or
@top
+equivalent_nodes-noderename.cnf: file empty for renamed node `@strong{}'
equivalent_nodes-noderename.cnf: old name for `ToP' is a node of the document
equivalent_nodes-noderename.cnf: old name for `Top' is a node of the document
-equivalent_nodes-noderename.cnf: file empty for renamed node `@strong{}'
equivalent_nodes-noderename.cnf: warning: target node (new name for `old non
existing') not in document: non existing file
Index:
tp/tests/sectioning/res_parser/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2
===
---
tp/tests/sectioning/res_parser/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2
(revision 5297)
+++
tp/tests/sectioning/res_parser/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2
(revision 5298)
@@ -7,7 +7,7 @@
equivalent_nodes-noderename.cnf:8: warning: no node to be renamed
equivalent_nodes-noderename.cnf:10: warning: nodes without a new name at the
end of file
equivalent_nodes.texi: warning: must specify a title with a title command or
@top
+equivalent_nodes-noderename.cnf: file empty for renamed node `@strong{}'
equivalent_nodes-noderename.cnf: old name for `ToP' is a node of the document
equivalent_nodes-noderename.cnf: old name for `Top' is a node of the document
-equivalent_nodes-noderename.cnf: file empty for renamed node `@strong{}'
equivalent_nodes-noderename.cnf: warning: target node (new name for `old non
existing') not in document: non existing file
Index:
tp/tests/formatting/res_parser/unknown_nodes_renamed/unknown_nodes_renamed.2
===
---
tp/tests/formatting/res_parser/unknown_nodes_renamed/unknown_nodes_renamed.2
(revision 5297)
+++
tp/tests/formatting/res_parser/unknown_nodes_renamed/unknown_nodes_renamed.2
(revision 5298)
@@ -1,2 +1,2 @@
unknown_nodes_renamed.texi: warning: must specify a title with a title command
or @top
-unknown_nodes_renamed-noderename.cnf:3: warning: target node (new name for
`unknown node 2') not in document: renamed as unknow too
+unknown_nodes_renamed-noderename.cnf:3: warning: target node (new name for
`Unknown node 1') not in document: renamed as unknow too
Index: tp/Texinfo/Convert/HTML.pm
===
--- tp/Texinfo/Convert/HTML.pm (revision 5297)
+++ tp/Texinfo/Convert/HTML.pm (revision 5298)
@@ -7071,7 +7071,7 @@
$parser_for_renamed_nodes = $self-{'parser'}-parser();
}
my %warned_new_node;
-foreach my $old_node_name (keys(%{$self-{'renamed_nodes'}})) {
+foreach my $old_node_name (sort(keys(%{$self-{'renamed_nodes'}}))) {
my $parsed_old_node = $self-_parse_node_and_warn_external(
$old_node_name, $parser_for_renamed_nodes,
$self-{'renamed_nodes_lines'}-{$old_node_name},
Index: ChangeLog
===
--- ChangeLog (revision 5297)
+++ ChangeLog (revision 5298)
@@ -1,3 +1,8 @@
+2013-08-11 Patrice Dumas pertu...@free.fr
+
+ * tp/Texinfo/Convert/HTML.pm(output): sort when processing renamed
+ nodes to have a reproducible error output. Report by Karl.
+
2013-08-10 Patrice Dumas pertu...@free.fr
* tp/Texinfo/Convert/Line.pm, Texinfo/Convert/Paragraph.pm: preserve
Index: tp/tests/layout/res_parser/formatting_chm/formatting.hhk
===
--- tp/tests/layout/res_parser/formatting_chm/formatting.hhk(revision 5307)
+++
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On 25.08.13 Dominic Hargreaves (d...@earth.li) wrote: Hi, Please ask debian-p...@lists.debian.org if you need help with preparing or testing a change; if you have a suitable test i386 system you can install perl 5.18 by adding experimental to your sources as well as the repository at http://people.debian.org/~dom/perl/test/perl-5.18.0/ I reported upstream @texinfo and got the following information: This is already solved in svn in r5308 and r5298. Thanks for the report nevertheless! I imported the 2 changed files into my texinfo 5.1 build dir. Now the test suite fails even more often. I'm afraid we have to use an svn checkout...unless upstream prepares a release ASAP and we disable the test suite until then. Norbert? H. -- sigmentation fault signature.asc Description: Digital signature
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On 30.08.13 Norbert Preining (prein...@logic.at) wrote:
On So, 25 Aug 2013, Dominic Hargreaves wrote:
Hi,
This package FTBFS with perl 5.18:
FAIL: test_scripts/sectioning_equivalent_nodes_test_renamed_nodes.sh
FAIL: test_scripts/layout_formatting_chm.sh
Unreproducible in a clean chroot/pbuilder as of today.
I used a chroot (not pbuilder). I see exactly these tests above
failing (log is attached).
H.
--
sigmentation fault
==
GNU Texinfo 5.1: tp/tests/test-suite.log
==
# TOTAL: 186
# PASS: 171
# SKIP: 13
# XFAIL: 0
# FAIL: 2
# XPASS: 0
# ERROR: 0
.. contents:: :depth: 2
FAIL: test_scripts/sectioning_equivalent_nodes_test_renamed_nodes.sh
D: sectioning/diffs/equivalent_nodes_test_renamed_nodes.diff
D: sectioning/diffs/equivalent_nodes_test_renamed_nodes_html.diff
base_result_dir .././sectioning, driving_file .././sectioning/tests-parser.txt
texi2any.pl equivalent_nodes_test_renamed_nodes
/usr/bin/perl -w -I .././sectioning/../../ -I
.././sectioning/../../maintain/lib/Unicode-EastAsianWidth/lib/ -I
.././sectioning/../../maintain/lib/libintl-perl/lib/ -I
.././sectioning/../../maintain/lib/Text-Unidecode/lib/
.././sectioning/../..//texi2any.pl --set-customization-variable=TEXI2HTML
--force --conf-dir .././sectioning/../../t/init/ --conf-dir
.././sectioning/../../init -I .././sectioning/ -I .././sectioning/../
--set-customization-variable L2H_FILE=.././sectioning/../../t/init/l2h.init
--error-limit=1000 --set-customization-variable TEST=1
--set-customization-variable L2H_CLEAN=0 --output
out_parser/equivalent_nodes_test_renamed_nodes/ --init test_renamed_nodes.init
.././sectioning/equivalent_nodes.texi
out_parser/equivalent_nodes_test_renamed_nodes/equivalent_nodes.1
2out_parser/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2
texi2any.pl equivalent_nodes_test_renamed_nodes
/usr/bin/perl -w -I .././sectioning/../../ -I
.././sectioning/../../maintain/lib/Unicode-EastAsianWidth/lib/ -I
.././sectioning/../../maintain/lib/libintl-perl/lib/ -I
.././sectioning/../../maintain/lib/Text-Unidecode/lib/
.././sectioning/../..//texi2any.pl --html --force --conf-dir
.././sectioning/../../t/init/ --conf-dir .././sectioning/../../init -I
.././sectioning/ -I .././sectioning/../ --set-customization-variable
L2H_FILE=.././sectioning/../../t/init/l2h.init --error-limit=1000
--set-customization-variable TEST=1 --set-customization-variable L2H_CLEAN=0
--output out_parser_html/equivalent_nodes_test_renamed_nodes/ --init
test_renamed_nodes.init .././sectioning/equivalent_nodes.texi
out_parser_html/equivalent_nodes_test_renamed_nodes/equivalent_nodes.1
2out_parser_html/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2
diff -a -r
diffs/staging_res/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2
out_parser/equivalent_nodes_test_renamed_nodes/equivalent_nodes.2
9a10,11
equivalent_nodes-noderename.cnf: warning: target node (new name for `old non
existing') not in document: non existing file
equivalent_nodes-noderename.cnf: file empty for renamed node `@strong{}'
12,13d13
equivalent_nodes-noderename.cnf: file empty for renamed node `@strong{}'
equivalent_nodes-noderename.cnf: warning: target node (new name for `old non
existing') not in document: non existing file
FAIL: test_scripts/layout_formatting_chm.sh
===
D: layout/diffs/formatting_chm.diff
base_result_dir .././layout, driving_file .././layout/tests-parser.txt
texi2any.pl formatting_chm
/usr/bin/perl -w -I .././layout/../../ -I
.././layout/../../maintain/lib/Unicode-EastAsianWidth/lib/ -I
.././layout/../../maintain/lib/libintl-perl/lib/ -I
.././layout/../../maintain/lib/Text-Unidecode/lib/
.././layout/../..//texi2any.pl --set-customization-variable=TEXI2HTML --force
--conf-dir .././layout/../../t/init/ --conf-dir .././layout/../../init -I
.././layout/ -I .././layout/../ --set-customization-variable
L2H_FILE=.././layout/../../t/init/l2h.init --error-limit=1000
--set-customization-variable TEST=1 --set-customization-variable L2H_CLEAN=0
--output out_parser/formatting_chm/ --init chm.pm
.././layout/../coverage/formatting.texi
out_parser/formatting_chm/formatting.1 2out_parser/formatting_chm/formatting.2
diff -a -r diffs/staging_res/formatting_chm/formatting.hhk
out_parser/formatting_chm/formatting.hhk
188,303d187
param name=Name value=quot;
param name=Local value=formatting.html#index-copying-_0060_0060
/OBJECT /LI
LI OBJECT type=text/sitemap
param name=Name value=quot;
param name=Local value=formatting.html#index-titlepage-_0060_0060
/OBJECT /LI
LI OBJECT type=text/sitemap
param name=Name value=quot;
param name=Local value=formatting.html#index-_0060_0060
/OBJECT /LI
LI OBJECT type=text/sitemap
param name=Name value=quot;
param name=Local
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
severity 720866 serious thanks Hi Hilmar, On Fr, 30 Aug 2013, Hilmar Preusse wrote: I used a chroot (not pbuilder). I see exactly these tests above failing (log is attached). Indeed, I retried again and it failed. I guess this is again (see biber postings) one of the perl 5.18 introduces randomization of hasj key return problem. I have to investigate. If you find time (I am in the mountains the next days) please run the tests with PERL_HASH_SEED=0 and see if it succeeds. Thanks Norbert PREINING, Norbert http://www.preining.info JAIST, Japan TeX Live Debian Developer DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On 30.08.13 Norbert Preining (prein...@logic.at) wrote: Hi, If you find time (I am in the mountains the next days) please run the tests with PERL_HASH_SEED=0 and see if it succeeds. This is an env var I have to set before building? H. -- sigmentation fault -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On 30.08.13 Norbert Preining (prein...@logic.at) wrote: On Fr, 30 Aug 2013, Hilmar Preusse wrote: Hi, This is an env var I have to set before building? Yes, and export it, too, otherwise it will not be seen by the sub-process. Sorry, no improvement: Testsuite summary for GNU Texinfo 5.1 # TOTAL: 186 # PASS: 170 # SKIP: 13 # XFAIL: 0 # FAIL: 3 # XPASS: 0 # ERROR: 0 H. -- sigmentation fault signature.asc Description: Digital signature
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On Fr, 30 Aug 2013, Hilmar Preusse wrote: This is an env var I have to set before building? Yes, and export it, too, otherwise it will not be seen by the sub-process. Thanks a lot Norbert PREINING, Norbert http://www.preining.info JAIST, Japan TeX Live Debian Developer DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On Fr, 30 Aug 2013, Hilmar Preusse wrote: Sorry, no improvement: Umpf, so we have to investigate. Can you send a short report with the failed items to bug-texi...@gnu.org (AFAIR) so that one of the upstream guys can look at it. Thanks Norbert PREINING, Norbert http://www.preining.info JAIST, Japan TeX Live Debian Developer DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
tags 720866 - moreinfo tags 720866 - unreproducible forwarded 720866 https://savannah.gnu.org/bugs/?39904 stop On 30.08.13 Norbert Preining (prein...@logic.at) wrote: On Fr, 30 Aug 2013, Hilmar Preusse wrote: Hi, Sorry, no improvement: Umpf, so we have to investigate. Can you send a short report with the failed items to bug-texi...@gnu.org (AFAIR) so that one of the upstream guys can look at it. Had a look into the ML archive found no trace of a known problem - bug #39904: Test-suite fails Hilmar -- sigmentation fault signature.asc Description: Digital signature
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
severity 720866 normal tags 720866 + unreproducible moreinfo thanks On So, 25 Aug 2013, Dominic Hargreaves wrote: This package FTBFS with perl 5.18: FAIL: test_scripts/sectioning_equivalent_nodes_test_renamed_nodes.sh FAIL: test_scripts/layout_formatting_chm.sh Unreproducible in a clean chroot/pbuilder as of today. Are you sure? Do you have logs? Norbert PREINING, Norbert http://www.preining.info JAIST, Japan TeX Live Debian Developer DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
On 25.08.13 Dominic Hargreaves (d...@earth.li) wrote: Hi, This package FTBFS with perl 5.18: FAIL: test_scripts/sectioning_equivalent_nodes_test_renamed_nodes.sh FAIL: test_scripts/layout_formatting_chm.sh This doesn't obviously look like a perl 5.18 issue for me, but it failed on two separate occasions when the same version passed on a plain sid chroot. Sorry I can't do any more investigation at the moment. Well, I see that issue w/ perl 5.18, but not w/ perl 5.14. The test suite compares the generated results and they don't seem to match to the expected results. H. -- sigmentation fault signature.asc Description: Digital signature
Bug#720866: texinfo: FTBFS with perl 5.18: test failures
Source: texinfo Version: 5.1.dfsg.1-4 Severity: serious Justification: transition imminent User: debian-p...@lists.debian.org Usertags: perl-5.18-transition Tags: sid jessie This package FTBFS with perl 5.18: FAIL: test_scripts/sectioning_equivalent_nodes_test_renamed_nodes.sh FAIL: test_scripts/layout_formatting_chm.sh This doesn't obviously look like a perl 5.18 issue for me, but it failed on two separate occasions when the same version passed on a plain sid chroot. Sorry I can't do any more investigation at the moment. Please ask debian-p...@lists.debian.org if you need help with preparing or testing a change; if you have a suitable test i386 system you can install perl 5.18 by adding experimental to your sources as well as the repository at http://people.debian.org/~dom/perl/test/perl-5.18.0/ You can also find more about the perl 5.18 transition, which we hope to start tomorrow[1], at http://bugs.debian.org/712615. Thanks for your work! Cheers, Dominic. [1] apologies for the late appearance of this bug report; earlier test coverage did not cover this package, possibly because of issues in build-dependencies -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
