Adjust tests to gdbm-1.15 using an upstream fix

perl-5.28.1-ext-GDBM_File-t-fatal.t-handle-non-fatality.patch

@@ -0,0 +1,80 @@
+From 72cc38bc65d4a675d9134acb085d2e0c3dcd5383 Mon Sep 17 00:00:00 2001
+From: David Mitchell <davem@iabyn.com>
+Date: Fri, 14 Dec 2018 16:54:42 +0000
+Subject: [PATCH] ext/GDBM_File/t/fatal.t: handle non-fatality
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+This script is supposed to exercise the error handling callback
+mechanism in gdbm, by triggering an error by surreptitiously closing
+the file handle which gdbm has opened.
+
+However, this doesn't trigger an error in newer releases of the gdbm
+library, which uses mmap() rather than write() etc. In fact I can't see
+any way of triggering an error: so just skip the relevant tests if we
+can't trigger a failure.
+
+Petr Písař: Ported to 5.28.1 from
+upstream's d33f9fbdb3bb27a3b32a2ffa9aa035617c07f7a1.
+
+Signed-off-by: Petr Písař <ppisar@redhat.com>
+---
+ ext/GDBM_File/t/fatal.t | 35 ++++++++++++++++++++++++++---------
+ 1 file changed, 26 insertions(+), 9 deletions(-)
+
+diff --git a/ext/GDBM_File/t/fatal.t b/ext/GDBM_File/t/fatal.t
+index 0e426d4..6945653 100644
+--- a/ext/GDBM_File/t/fatal.t
++++ b/ext/GDBM_File/t/fatal.t
+@@ -1,4 +1,12 @@
+ #!./perl -w
++#
++# Exercise the error handling callback mechanism in gdbm.
++#
++# Try to trigger an error by surreptitiously closing the file handle which
++# gdbm has opened.  Note that this won't trigger an error in newer
++# releases of the gdbm library, which uses mmap() rather than write() etc:
++# so skip in that case.
++
+ use strict;
+ 
+ use Test::More;
+@@ -34,16 +42,25 @@ isnt((open $fh, "<&=$fileno"), undef, "dup fileno $fileno")
+     or diag("\$! = $!");
+ isnt(close $fh, undef,
+      "close fileno $fileno, out from underneath the GDBM_File");
+-is(eval {
++
++# store some data to a closed file handle
++
++my $res = eval {
+     $h{Perl} = 'Rules';
+     untie %h;
+-    1;
+-}, undef, 'Trapped error when attempting to write to knobbled GDBM_File');
+-
+-# Observed "File write error" and "lseek error" from two different systems.
+-# So there might be more variants. Important part was that we trapped the error
+-# via croak.
+-like($@, qr/ at .*\bfatal\.t line \d+\.\n\z/,
+-     'expected error message from GDBM_File');
++    99;
++};
++
++SKIP: {
++    skip "Can't tigger failure", 2 if $res == 99;
++
++    is $res, undef, "eval should return undef";
++
++    # Observed "File write error" and "lseek error" from two different
++    # systems.  So there might be more variants. Important part was that
++    # we trapped the error # via croak.
++    like($@, qr/ at .*\bfatal\.t line \d+\.\n\z/,
++         'expected error message from GDBM_File');
++}
+ 
+ unlink <Op_dbmx*>;
+-- 
+2.17.2
+

perl-5.29.0-Remove-ext-GDBM_File-t-fatal.t.patch

@@ -1,94 +0,0 @@
-From 892e8b006aa99ac2c880cdc2a81fd16f06c1a0f3 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Petr=20P=C3=ADsa=C5=99?= <ppisar@redhat.com>
-Date: Mon, 9 Jul 2018 16:18:36 +0200
-Subject: [PATCH] Remove ext/GDBM_File/t/fatal.t
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-gdbm-1.15 defaults to a memory-mapped I/O and does not report any I/O
-errors on store and close operations. Thus ext/GDBM_File/t/fatal.t
-test that expects these fatal error reports fails. Because there is
-no other way to provoke a fatal error in gdbm-1.15 this patch
-removes the test. Future gdbm version promisses reporting a regular
-error on closing a database.
-
-RT#133295
-
-Signed-off-by: Petr Písař <ppisar@redhat.com>
----
- MANIFEST                |  1 -
- ext/GDBM_File/t/fatal.t | 49 -------------------------------------------------
- 2 files changed, 50 deletions(-)
- delete mode 100644 ext/GDBM_File/t/fatal.t
-
-diff --git a/MANIFEST b/MANIFEST
-index 95fa539095..b07fed1f54 100644
---- a/MANIFEST
-+++ b/MANIFEST
-@@ -4100,7 +4100,6 @@ ext/GDBM_File/GDBM_File.pm	GDBM extension Perl module
- ext/GDBM_File/GDBM_File.xs	GDBM extension external subroutines
- ext/GDBM_File/hints/sco.pl	Hint for GDBM_File for named architecture
- ext/GDBM_File/Makefile.PL	GDBM extension makefile writer
--ext/GDBM_File/t/fatal.t		Test the fatal_func argument to gdbm_open
- ext/GDBM_File/t/gdbm.t		See if GDBM_File works
- ext/GDBM_File/typemap		GDBM extension interface types
- ext/Hash-Util/Changes		Change history of Hash::Util
-diff --git a/ext/GDBM_File/t/fatal.t b/ext/GDBM_File/t/fatal.t
-deleted file mode 100644
-index 0e426d4dbc..0000000000
---- a/ext/GDBM_File/t/fatal.t
-+++ /dev/null
-@@ -1,49 +0,0 @@
--#!./perl -w
--use strict;
--
--use Test::More;
--use Config;
--
--BEGIN {
--    plan(skip_all => "GDBM_File was not built")
--	unless $Config{extensions} =~ /\bGDBM_File\b/;
--
--    # https://rt.perl.org/Public/Bug/Display.html?id=117967
--    plan(skip_all => "GDBM_File is flaky in $^O")
--        if $^O =~ /darwin/;
--
--    plan(tests => 8);
--    use_ok('GDBM_File');
--}
--
--unlink <Op_dbmx*>;
--
--open my $fh, '<', $^X or die "Can't open $^X: $!";
--my $fileno = fileno $fh;
--isnt($fileno, undef, "Can find next available file descriptor");
--close $fh or die $!;
--
--is((open $fh, "<&=$fileno"), undef,
--   "Check that we cannot open fileno $fileno. \$! is $!");
--
--umask(0);
--my %h;
--isa_ok(tie(%h, 'GDBM_File', 'Op_dbmx', GDBM_WRCREAT, 0640), 'GDBM_File');
--
--isnt((open $fh, "<&=$fileno"), undef, "dup fileno $fileno")
--    or diag("\$! = $!");
--isnt(close $fh, undef,
--     "close fileno $fileno, out from underneath the GDBM_File");
--is(eval {
--    $h{Perl} = 'Rules';
--    untie %h;
--    1;
--}, undef, 'Trapped error when attempting to write to knobbled GDBM_File');
--
--# Observed "File write error" and "lseek error" from two different systems.
--# So there might be more variants. Important part was that we trapped the error
--# via croak.
--like($@, qr/ at .*\bfatal\.t line \d+\.\n\z/,
--     'expected error message from GDBM_File');
--
--unlink <Op_dbmx*>;
--- 
-2.14.4
-

perl-5.29.5-Avoid-Use-of-uninitialized-value-res-in-numeric-eq-w.patch

@@ -0,0 +1,35 @@
+From c950d6fa306a1a0a1e28ece3646aec9490a7ea0e Mon Sep 17 00:00:00 2001
+From: James E Keenan <jkeenan@cpan.org>
+Date: Sat, 15 Dec 2018 10:29:12 -0500
+Subject: [PATCH] Avoid "Use of uninitialized value $res in numeric eq (==)"
+ warning
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+The test within the SKIP block expects $res to be undef, but the 'skip'
+condition itself expects it to be defined and numeric.  So we were
+getting an uninitialized value warning.  In 'skip' condition, test for
+definedness before numeric comparison.
+
+Signed-off-by: Petr Písař <ppisar@redhat.com>
+---
+ ext/GDBM_File/t/fatal.t | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/ext/GDBM_File/t/fatal.t b/ext/GDBM_File/t/fatal.t
+index 01068f3cf4..1cbfdc6018 100644
+--- a/ext/GDBM_File/t/fatal.t
++++ b/ext/GDBM_File/t/fatal.t
+@@ -52,7 +52,7 @@ my $res = eval {
+ };
+ 
+ SKIP: {
+-    skip "Can't trigger failure", 2 if $res == 99;
++    skip "Can't trigger failure", 2 if (defined $res and $res == 99);
+ 
+     is $res, undef, "eval should return undef";
+ 
+-- 
+2.17.2
+

perl-5.29.5-Correct-spelling-error-in-skip-message.patch

@@ -0,0 +1,29 @@
+From c4df8c74c98b6ecac7f95d0184638f24f0a13bcc Mon Sep 17 00:00:00 2001
+From: James E Keenan <jkeenan@cpan.org>
+Date: Sat, 15 Dec 2018 09:01:55 -0500
+Subject: [PATCH] Correct spelling error in 'skip' message
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+Signed-off-by: Petr Písař <ppisar@redhat.com>
+---
+ ext/GDBM_File/t/fatal.t | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/ext/GDBM_File/t/fatal.t b/ext/GDBM_File/t/fatal.t
+index 159916901a..01068f3cf4 100644
+--- a/ext/GDBM_File/t/fatal.t
++++ b/ext/GDBM_File/t/fatal.t
+@@ -52,7 +52,7 @@ my $res = eval {
+ };
+ 
+ SKIP: {
+-    skip "Can't tigger failure", 2 if $res == 99;
++    skip "Can't trigger failure", 2 if $res == 99;
+ 
+     is $res, undef, "eval should return undef";
+ 
+-- 
+2.17.2
+

perl.spec

@@ -83,7 +83,7 @@ License:        GPL+ or Artistic
 Epoch:          %{perl_epoch}
 Version:        %{perl_version}
 # release number must be even higher, because dual-lived modules will be broken otherwise
-Release:        429%{?dist}
+Release:        430%{?dist}
 Summary:        Practical Extraction and Report Language
 Url:            https://www.perl.org/
 Source0:        https://www.cpan.org/src/5.0/perl-%{perl_version}.tar.xz
@@ -152,9 +152,6 @@ Patch13:        perl-5.26.0-perl-131588-be-a-little-more-careful-in-arybase-_tie
 # <https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/3RHZEHLRUHJFF2XGHI5RB6YPDNLDR4HG/>
 Patch14:        perl-5.27.8-hints-linux-Add-lphtread-to-lddlflags.patch
 
-# Adjust tests to gdbm-1.15, RT#133295
-Patch15:        perl-5.29.0-Remove-ext-GDBM_File-t-fatal.t.patch
-
 # Fix printing a warning about a wide character when matching a regular
 # expression while ISO-8859-1 locale is in effect, in upstream after 5.29.0
 Patch17:        perl-5.29.0-regexec.c-Call-macro-with-correct-args.patch
@@ -217,6 +214,11 @@ Patch37:        perl-5.29.5-perl-133659-make-an-in-place-edit-successful-if-the-
 # compiled code blocks, RT#133687, in upstream after 5.29.5
 Patch38:        perl-5.29.5-handle-code-mixed-compile-and-runtime.patch
 
+# Adjust tests to gdbm-1.15, RT#133295, in upstream after 5.29.5
+Patch39:        perl-5.28.1-ext-GDBM_File-t-fatal.t-handle-non-fatality.patch
+Patch40:        perl-5.29.5-Correct-spelling-error-in-skip-message.patch
+Patch41:        perl-5.29.5-Avoid-Use-of-uninitialized-value-res-in-numeric-eq-w.patch
+
 # Link XS modules to libperl.so with EU::CBuilder on Linux, bug #960048
 Patch200:       perl-5.16.3-Link-XS-modules-to-libperl.so-with-EU-CBuilder-on-Li.patch
 
@@ -2790,7 +2792,6 @@ Perl extension for Version Objects
 %patch12 -p1
 %patch13 -p1
 %patch14 -p1
-%patch15 -p1
 %patch17 -p1
 %patch18 -p1
 %patch20 -p1
@@ -2810,6 +2811,9 @@ Perl extension for Version Objects
 %patch36 -p1
 %patch37 -p1
 %patch38 -p1
+%patch39 -p1
+%patch40 -p1
+%patch41 -p1
 %patch200 -p1
 %patch201 -p1
 
@@ -2831,7 +2835,6 @@ perl -x patchlevel.h \
     'Fedora Patch12: Replace EU::MakeMaker dependency with EU::MM::Utils in IPC::Cmd (bug #1129443)' \
     'Fedora Patch13: Fix executing arybase::_tie_it() in Safe compartement (RT#131588)' \
     'Fedora Patch14: Link XS modules to pthread library to fix linking with -z defs' \
-    'Fedora Patch15: Adjust tests to gdbm-1.15 (RT#133295)' \
     'Fedora Patch17: Fix printing a warning about a wide character when matching a regular expression while ISO-8859-1 locale is in effect' \
     'Fedora Patch18: Fix invoking a check for wide characters while ISO-8859-1 locale is in effect' \
     'Fedora Patch20: Fix build conditions in locale.c' \
@@ -2847,6 +2850,7 @@ perl -x patchlevel.h \
     'Fedora Patch34: Fix an undefined behaviour in S_hv_delete_common()' \
     'Fedora Patch35: Fix in-place edit to replace files on a successful perl exit status (bug #1650041)' \
     'Fedora Patch38: Fix compiling regular expressions that contain both compile- and run-time compiled code blocks (RT#133687)' \
+    'Fedora Patch39: Adjust tests to gdbm-1.15 (RT#133295)' \
     'Fedora Patch200: Link XS modules to libperl.so with EU::CBuilder on Linux' \
     'Fedora Patch201: Link XS modules to libperl.so with EU::MM on Linux' \
     %{nil}
@@ -5135,6 +5139,9 @@ popd
 
 # Old changelog entries are preserved in CVS.
 %changelog
+* Mon Jan 14 2019 Petr Pisar <ppisar@redhat.com> - 4:5.28.1-430
+- Adjust tests to gdbm-1.15 using an upstream fix (RT#133295)
+
 * Mon Jan 14 2019 Björn Esser <besser82@fedoraproject.org> - 4:5.28.1-429
 - Rebuilt for libcrypt.so.2 (#1666033)