[DRE-maint] Bug#1079550: src:ruby-timecop: fails to migrate to testing for too long: autopkgtest regression on 32 bit archs

[Paul Gevers]

Hi,

On Mon, 14 Oct 2024 17:37:21 +0100 Paul Gevers <elbrus at debian.org> wrote:
> On Sat, 24 Aug 2024 15:49:31 +0200 Paul Gevers <elbrus at debian.org> wrote:
> > The version in unstable fails its own autopkgtest on the 32 bit 
> > architectures armel, armhf and i386.
> I propose we skip the 2 tests for now on these archs.

I have uploaded the attached debdiff to the archive (with dgit).

Paul

-------------- next part --------------
diff -Nru ruby-timecop-0.9.10/debian/changelog ruby-timecop-0.9.10/debian/changelog
--- ruby-timecop-0.9.10/debian/changelog	2024-07-23 16:16:38.000000000 +0200
+++ ruby-timecop-0.9.10/debian/changelog	2024-11-02 08:34:36.000000000 +0100
@@ -1,3 +1,10 @@
+ruby-timecop (0.9.10-1.1) unstable; urgency=medium
+
+  * Non-maintainer upload
+  * tests: skip two clock tests on 32-bits architectures
+
+ -- Paul Gevers <elbrus at debian.org>  Sat, 02 Nov 2024 08:34:36 +0100
+
 ruby-timecop (0.9.10-1) unstable; urgency=medium
 
   * New upstream version 0.9.10
diff -Nru ruby-timecop-0.9.10/debian/patches/series ruby-timecop-0.9.10/debian/patches/series
--- ruby-timecop-0.9.10/debian/patches/series	2024-07-23 16:16:38.000000000 +0200
+++ ruby-timecop-0.9.10/debian/patches/series	2024-11-02 08:34:36.000000000 +0100
@@ -1 +1,2 @@
 test_helper.rb_editing.patch
+tests-skip-two-clock-tests-on-32-bits-ar.patch
diff -Nru ruby-timecop-0.9.10/debian/patches/tests-skip-two-clock-tests-on-32-bits-ar.patch ruby-timecop-0.9.10/debian/patches/tests-skip-two-clock-tests-on-32-bits-ar.patch
--- ruby-timecop-0.9.10/debian/patches/tests-skip-two-clock-tests-on-32-bits-ar.patch	1970-01-01 01:00:00.000000000 +0100
+++ ruby-timecop-0.9.10/debian/patches/tests-skip-two-clock-tests-on-32-bits-ar.patch	2024-11-02 08:34:36.000000000 +0100
@@ -0,0 +1,30 @@
+From: Paul Gevers <elbrus at debian.org>
+Date: Sat, 2 Nov 2024 08:32:35 +0100
+X-Dgit-Generated: 0.9.10-1.1 80339e647656905375c663e1ef83b39b00206993
+Subject: tests: skip two clock tests on 32-bits architectures
+
+The test is new, but fails on armel, armhf and i386, our 32 bit architectures.
+Should be properly investigated, but for now let's skip it to enable migration.
+
+---
+
+diff --git a/test/timecop_with_process_clock_test.rb b/test/timecop_with_process_clock_test.rb
+index c131f5b..c115201 100644
+--- a/test/timecop_with_process_clock_test.rb
++++ b/test/timecop_with_process_clock_test.rb
+@@ -23,6 +23,7 @@ class TestTimecopWithProcessClock < Minitest::Test
+     end
+ 
+     def test_process_clock_gettime_monotonic
++      skip('Fails on 32 bit archs') if ['arm-linux-gnueabihf', 'arm-linux-gnueabi', 'i386-linux-gnu'].include? RUBY_PLATFORM
+       Timecop.freeze do
+         assert_same(*consecutive_monotonic, "CLOCK_MONOTONIC is not frozen")
+       end
+@@ -116,6 +117,7 @@ class TestTimecopWithProcessClock < Minitest::Test
+     end
+ 
+     def test_process_clock_gettime_realtime
++      skip('Fails on 32 bit archs') if ['arm-linux-gnueabihf', 'arm-linux-gnueabi', 'i386-linux-gnu'].include? RUBY_PLATFORM
+       Timecop.freeze do
+         assert_same(*consecutive_realtime, "CLOCK_REALTIME is not frozen")
+       end
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature.asc
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <http://alioth-lists.debian.net/pipermail/pkg-ruby-extras-maintainers/attachments/20241102/bc790636/attachment.sig>