unofficial mirror of guix-patches@gnu.org 
 help / color / mirror / code / Atom feed
* [bug#69652] [PATCH gnome-team 1/1] gnu: orbit2: Mark it unsupported in 32-bits systems.
  2024-03-08 17:29 [bug#69652] [PATCH gnome-team 0/1] What should we do with orbit2? Vivien Kraus via Guix-patches via
@ 2024-03-08 17:20 ` Vivien Kraus via Guix-patches via
  2024-03-08 19:11   ` Liliana Marie Prikler
  2024-03-09  8:47 ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Vivien Kraus via Guix-patches via
  2024-03-09 15:57 ` [bug#69652] [PATCH gnome-team 0/1] What should we do with orbit2? Aaron Covrig via Guix-patches via
  2 siblings, 1 reply; 12+ messages in thread
From: Vivien Kraus via Guix-patches via @ 2024-03-08 17:20 UTC (permalink / raw)
  To: 69652; +Cc: rg, vivien, liliana.prikler, maxim.cournoyer

* gnu/packages/gnome.scm (orbit2) [supported-systems]: New field.
[arguments]: Convert to list of G-Expressions.

Change-Id: I55766c43e7abc703815c30dccf38faf640ffc07e
---
 gnu/packages/gnome.scm | 53 ++++++++++++++++++++++--------------------
 1 file changed, 28 insertions(+), 25 deletions(-)

diff --git a/gnu/packages/gnome.scm b/gnu/packages/gnome.scm
index 13cad0add2..711890d617 100644
--- a/gnu/packages/gnome.scm
+++ b/gnu/packages/gnome.scm
@@ -3798,34 +3798,37 @@ (define-public orbit2
               (sha256
                (base32
                 "0l3mhpyym9m5iz09fz0rgiqxl2ym6kpkwpsp1xrr4aa80nlh1jam"))))
+    ;; Many tests fail on 32-bit systems.
+    (supported-systems %64bit-supported-systems)
     (build-system gnu-build-system)
     (arguments
-     `(;; The "timeout-server" test hangs when run in parallel.
-       #:parallel-tests? #f
-       #:configure-flags
-       '(;; We don't need static libraries, plus they don't build reproducibly
-         ;; (non-deterministic ordering of .o files in the archive.)
-         "--disable-static"
-
-         ;; The programmer kindly gives us a hook to turn off deprecation
-         ;; warnings ...
-         "DISABLE_DEPRECATED_CFLAGS=-DGLIB_DISABLE_DEPRECATION_WARNINGS")
-       ;; ... which they then completly ignore !!
-       #:phases
-       (modify-phases %standard-phases
-         (add-after 'unpack 'fix-parallel-build
-           ;; Parallel build fails because of a failed dependency,
-           ;; https://bugzilla.gnome.org/show_bug.cgi?id=732274
-           (lambda _
-             (substitute* "src/services/name/Makefile.am"
-               (("orbit_name_server_2_DEPENDENCIES = \\$(DEPS) CosNaming.h")
-                "orbit_name_server_2_DEPENDENCIES = \
+     (list
+      ;; The "timeout-server" test hangs when run in parallel.
+      #:parallel-tests? #f
+      #:configure-flags
+      #~'(;; We don't need static libraries, plus they don't build reproducibly
+          ;; (non-deterministic ordering of .o files in the archive.)
+          "--disable-static"
+
+          ;; The programmer kindly gives us a hook to turn off deprecation
+          ;; warnings ...
+          "DISABLE_DEPRECATED_CFLAGS=-DGLIB_DISABLE_DEPRECATION_WARNINGS")
+      ;; ... which they then completly ignore !!
+      #:phases
+      #~(modify-phases %standard-phases
+          (add-after 'unpack 'fix-parallel-build
+            ;; Parallel build fails because of a failed dependency,
+            ;; https://bugzilla.gnome.org/show_bug.cgi?id=732274
+            (lambda _
+              (substitute* "src/services/name/Makefile.am"
+                (("orbit_name_server_2_DEPENDENCIES = \\$(DEPS) CosNaming.h")
+                 "orbit_name_server_2_DEPENDENCIES = \
 $(DEPS) CosNaming.h libname-server-2.a"))))
-         (add-before 'configure 'ignore-deprecations
-           (lambda _
-             (substitute* "linc2/src/Makefile.in"
-               (("-DG_DISABLE_DEPRECATED")
-                "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
+          (add-before 'configure 'ignore-deprecations
+            (lambda _
+              (substitute* "linc2/src/Makefile.in"
+                (("-DG_DISABLE_DEPRECATED")
+                 "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
     (inputs (list glib libidl))
     (native-inputs
      (list pkg-config))
-- 
2.41.0




^ permalink raw reply related	[flat|nested] 12+ messages in thread

* [bug#69652] [PATCH gnome-team 0/1] What should we do with orbit2?
@ 2024-03-08 17:29 Vivien Kraus via Guix-patches via
  2024-03-08 17:20 ` [bug#69652] [PATCH gnome-team 1/1] gnu: orbit2: Mark it unsupported in 32-bits systems Vivien Kraus via Guix-patches via
                   ` (2 more replies)
  0 siblings, 3 replies; 12+ messages in thread
From: Vivien Kraus via Guix-patches via @ 2024-03-08 17:29 UTC (permalink / raw)
  To: 69652; +Cc: liliana.prikler, maxim.cournoyer, rg, vivien

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1077 bytes --]

Dear Guix,

Orbit2 is an old, unmaintained package that was used for the Gtk+2 stack, as
far as I understand. We still have Gtk+2 around, mainly for OCaml lablgtk2
(and _3, yes).

On gnome-team, this package fails to build on 32-bit systems.  I fail to
understand why it would fail on gnome-team and not on master.  If I
understood, maybe a clear solution would appear.  As of now, I can see
different mitigations:
— pretend there is no problem, skip the failing tests (they are everywhere, so
  maybe skip tests entirely) and ship it;
— remove it (I can’t see it on Debian, for instance);
— fix the test crashes (I won’t, and it is unmaintained);
— mark it as unsupported.

I went with the last option because I feel it is the most neutral thing to do.

What do you think?

Best regards,

Vivien

Vivien Kraus (1):
  gnu: orbit2: Mark it unsupported in 32-bits systems.

 gnu/packages/gnome.scm | 53 ++++++++++++++++++++++--------------------
 1 file changed, 28 insertions(+), 25 deletions(-)


base-commit: 25c14c893f05019d746321285acf55d1aa65b943
-- 
2.41.0




^ permalink raw reply	[flat|nested] 12+ messages in thread

* [bug#69652] [PATCH gnome-team 1/1] gnu: orbit2: Mark it unsupported in 32-bits systems.
  2024-03-08 17:20 ` [bug#69652] [PATCH gnome-team 1/1] gnu: orbit2: Mark it unsupported in 32-bits systems Vivien Kraus via Guix-patches via
@ 2024-03-08 19:11   ` Liliana Marie Prikler
  2024-03-08 20:03     ` Vivien Kraus via Guix-patches via
  0 siblings, 1 reply; 12+ messages in thread
From: Liliana Marie Prikler @ 2024-03-08 19:11 UTC (permalink / raw)
  To: Vivien Kraus, 69652; +Cc: rg, maxim.cournoyer

Am Freitag, dem 08.03.2024 um 18:20 +0100 schrieb Vivien Kraus:
> * gnu/packages/gnome.scm (orbit2) [supported-systems]: New field.
> [arguments]: Convert to list of G-Expressions.
> 
> Change-Id: I55766c43e7abc703815c30dccf38faf640ffc07e
> ---
>  gnu/packages/gnome.scm | 53 ++++++++++++++++++++++------------------
> --
>  1 file changed, 28 insertions(+), 25 deletions(-)
> 
> diff --git a/gnu/packages/gnome.scm b/gnu/packages/gnome.scm
> index 13cad0add2..711890d617 100644
> --- a/gnu/packages/gnome.scm
> +++ b/gnu/packages/gnome.scm
> @@ -3798,34 +3798,37 @@ (define-public orbit2
>                (sha256
>                 (base32
>                 
> "0l3mhpyym9m5iz09fz0rgiqxl2ym6kpkwpsp1xrr4aa80nlh1jam"))))
> +    ;; Many tests fail on 32-bit systems.
> +    (supported-systems %64bit-supported-systems)
Prefer the “disable-failing-tests” pattern, but first investigate why
they fail.  Note that this would make icecat and icedove 64-bit only
(that is if Rust doesn't do so already).
>      (build-system gnu-build-system)
>      (arguments
> -     `(;; The "timeout-server" test hangs when run in parallel.
> -       #:parallel-tests? #f
> -       #:configure-flags
> -       '(;; We don't need static libraries, plus they don't build
> reproducibly
> -         ;; (non-deterministic ordering of .o files in the archive.)
> -         "--disable-static"
> -
> -         ;; The programmer kindly gives us a hook to turn off
> deprecation
> -         ;; warnings ...
> -         "DISABLE_DEPRECATED_CFLAGS=-
> DGLIB_DISABLE_DEPRECATION_WARNINGS")
> -       ;; ... which they then completly ignore !!
> -       #:phases
> -       (modify-phases %standard-phases
> -         (add-after 'unpack 'fix-parallel-build
> -           ;; Parallel build fails because of a failed dependency,
> -           ;; https://bugzilla.gnome.org/show_bug.cgi?id=732274
> -           (lambda _
> -             (substitute* "src/services/name/Makefile.am"
> -               (("orbit_name_server_2_DEPENDENCIES = \\$(DEPS)
> CosNaming.h")
> -                "orbit_name_server_2_DEPENDENCIES = \
> +     (list
> +      ;; The "timeout-server" test hangs when run in parallel.
> +      #:parallel-tests? #f
> +      #:configure-flags
> +      #~'(;; We don't need static libraries, plus they don't build
> reproducibly
> +          ;; (non-deterministic ordering of .o files in the
> archive.)
> +          "--disable-static"
> +
> +          ;; The programmer kindly gives us a hook to turn off
> deprecation
> +          ;; warnings ...
> +          "DISABLE_DEPRECATED_CFLAGS=-
> DGLIB_DISABLE_DEPRECATION_WARNINGS")
> +      ;; ... which they then completly ignore !!
> +      #:phases
> +      #~(modify-phases %standard-phases
> +          (add-after 'unpack 'fix-parallel-build
> +            ;; Parallel build fails because of a failed dependency,
> +            ;; https://bugzilla.gnome.org/show_bug.cgi?id=732274
> +            (lambda _
> +              (substitute* "src/services/name/Makefile.am"
> +                (("orbit_name_server_2_DEPENDENCIES = \\$(DEPS)
> CosNaming.h")
> +                 "orbit_name_server_2_DEPENDENCIES = \
>  $(DEPS) CosNaming.h libname-server-2.a"))))
> -         (add-before 'configure 'ignore-deprecations
> -           (lambda _
> -             (substitute* "linc2/src/Makefile.in"
> -               (("-DG_DISABLE_DEPRECATED")
> -                "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
> +          (add-before 'configure 'ignore-deprecations
> +            (lambda _
> +              (substitute* "linc2/src/Makefile.in"
> +                (("-DG_DISABLE_DEPRECATED")
> +                 "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
What changes in this hunk?

Cheers

^ permalink raw reply	[flat|nested] 12+ messages in thread

* [bug#69652] [PATCH gnome-team 1/1] gnu: orbit2: Mark it unsupported in 32-bits systems.
  2024-03-08 19:11   ` Liliana Marie Prikler
@ 2024-03-08 20:03     ` Vivien Kraus via Guix-patches via
  2024-03-09  6:36       ` Liliana Marie Prikler
  0 siblings, 1 reply; 12+ messages in thread
From: Vivien Kraus via Guix-patches via @ 2024-03-08 20:03 UTC (permalink / raw)
  To: Liliana Marie Prikler, 69652; +Cc: rg, maxim.cournoyer

Le vendredi 08 mars 2024 à 20:11 +0100, Liliana Marie Prikler a écrit :
> Prefer the “disable-failing-tests” pattern, but first investigate why
> they fail.  Note that this would make icecat and icedove 64-bit only
> (that is if Rust doesn't do so already).

Rust has already been introduced in both.

> >  $(DEPS) CosNaming.h libname-server-2.a"))))
> > -         (add-before 'configure 'ignore-deprecations
> > -           (lambda _
> > -             (substitute* "linc2/src/Makefile.in"
> > -               (("-DG_DISABLE_DEPRECATED")
> > -                "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
> > +          (add-before 'configure 'ignore-deprecations
> > +            (lambda _
> > +              (substitute* "linc2/src/Makefile.in"
> > +                (("-DG_DISABLE_DEPRECATED")
> > +                 "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
> What changes in this hunk?
> 
It’s hard to see, but the indentation changed.  Should it be ignored?




^ permalink raw reply	[flat|nested] 12+ messages in thread

* [bug#69652] [PATCH gnome-team 1/1] gnu: orbit2: Mark it unsupported in 32-bits systems.
  2024-03-08 20:03     ` Vivien Kraus via Guix-patches via
@ 2024-03-09  6:36       ` Liliana Marie Prikler
  0 siblings, 0 replies; 12+ messages in thread
From: Liliana Marie Prikler @ 2024-03-09  6:36 UTC (permalink / raw)
  To: Vivien Kraus, 69652; +Cc: rg, maxim.cournoyer

Am Freitag, dem 08.03.2024 um 21:03 +0100 schrieb Vivien Kraus:
> Le vendredi 08 mars 2024 à 20:11 +0100, Liliana Marie Prikler a
> écrit :
> > Prefer the “disable-failing-tests” pattern, but first investigate
> > why they fail.  Note that this would make icecat and icedove 64-bit
> > only (that is if Rust doesn't do so already).
> 
> Rust has already been introduced in both.
I'd still like to avoid overconstraining if possible.

> > >  $(DEPS) CosNaming.h libname-server-2.a"))))
> > > -         (add-before 'configure 'ignore-deprecations
> > > -           (lambda _
> > > -             (substitute* "linc2/src/Makefile.in"
> > > -               (("-DG_DISABLE_DEPRECATED")
> > > -                "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
> > > +          (add-before 'configure 'ignore-deprecations
> > > +            (lambda _
> > > +              (substitute* "linc2/src/Makefile.in"
> > > +                (("-DG_DISABLE_DEPRECATED")
> > > +                 "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
> > What changes in this hunk?
> > 
> It’s hard to see, but the indentation changed.  Should it be ignored?
Rather, the cosmetic change should be separated from the semantic
change.

Cheers

^ permalink raw reply	[flat|nested] 12+ messages in thread

* [bug#69652] [PATCH gnome-team v2 1/3] gnu: orbit2: Update style.
  2024-03-09  8:47 ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Vivien Kraus via Guix-patches via
@ 2024-03-09  7:48   ` Vivien Kraus via Guix-patches via
  2024-03-09  7:52   ` [bug#69652] [PATCH gnome-team v2 2/3] gnu: orbit2: Propagate inputs Vivien Kraus via Guix-patches via
                     ` (2 subsequent siblings)
  3 siblings, 0 replies; 12+ messages in thread
From: Vivien Kraus via Guix-patches via @ 2024-03-09  7:48 UTC (permalink / raw)
  To: 69652; +Cc: rg, vivien, liliana.prikler, maxim.cournoyer

* gnu/packages/gnome.scm (orbit2) [arguments]: Convert to list of G-Expressions.
[native-inputs]: Write the native-input list on one line.

Change-Id: I6a09616719c07cbb48195cf0c59d8b799c404d47
---
 gnu/packages/gnome.scm | 54 +++++++++++++++++++++---------------------
 1 file changed, 27 insertions(+), 27 deletions(-)

diff --git a/gnu/packages/gnome.scm b/gnu/packages/gnome.scm
index 13cad0add2..fa911e1e59 100644
--- a/gnu/packages/gnome.scm
+++ b/gnu/packages/gnome.scm
@@ -3800,35 +3800,35 @@ (define-public orbit2
                 "0l3mhpyym9m5iz09fz0rgiqxl2ym6kpkwpsp1xrr4aa80nlh1jam"))))
     (build-system gnu-build-system)
     (arguments
-     `(;; The "timeout-server" test hangs when run in parallel.
-       #:parallel-tests? #f
-       #:configure-flags
-       '(;; We don't need static libraries, plus they don't build reproducibly
-         ;; (non-deterministic ordering of .o files in the archive.)
-         "--disable-static"
-
-         ;; The programmer kindly gives us a hook to turn off deprecation
-         ;; warnings ...
-         "DISABLE_DEPRECATED_CFLAGS=-DGLIB_DISABLE_DEPRECATION_WARNINGS")
-       ;; ... which they then completly ignore !!
-       #:phases
-       (modify-phases %standard-phases
-         (add-after 'unpack 'fix-parallel-build
-           ;; Parallel build fails because of a failed dependency,
-           ;; https://bugzilla.gnome.org/show_bug.cgi?id=732274
-           (lambda _
-             (substitute* "src/services/name/Makefile.am"
-               (("orbit_name_server_2_DEPENDENCIES = \\$(DEPS) CosNaming.h")
-                "orbit_name_server_2_DEPENDENCIES = \
+     (list
+      ;; The "timeout-server" test hangs when run in parallel.
+      #:parallel-tests? #f
+      #:configure-flags
+      #~'(;; We don't need static libraries, plus they don't build reproducibly
+          ;; (non-deterministic ordering of .o files in the archive.)
+          "--disable-static"
+
+          ;; The programmer kindly gives us a hook to turn off deprecation
+          ;; warnings ...
+          "DISABLE_DEPRECATED_CFLAGS=-DGLIB_DISABLE_DEPRECATION_WARNINGS")
+      ;; ... which they then completly ignore !!
+      #:phases
+      #~(modify-phases %standard-phases
+          (add-after 'unpack 'fix-parallel-build
+            ;; Parallel build fails because of a failed dependency,
+            ;; https://bugzilla.gnome.org/show_bug.cgi?id=732274
+            (lambda _
+              (substitute* "src/services/name/Makefile.am"
+                (("orbit_name_server_2_DEPENDENCIES = \\$(DEPS) CosNaming.h")
+                 "orbit_name_server_2_DEPENDENCIES = \
 $(DEPS) CosNaming.h libname-server-2.a"))))
-         (add-before 'configure 'ignore-deprecations
-           (lambda _
-             (substitute* "linc2/src/Makefile.in"
-               (("-DG_DISABLE_DEPRECATED")
-                "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
+          (add-before 'configure 'ignore-deprecations
+            (lambda _
+              (substitute* "linc2/src/Makefile.in"
+                (("-DG_DISABLE_DEPRECATED")
+                 "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
     (inputs (list glib libidl))
-    (native-inputs
-     (list pkg-config))
+    (native-inputs (list pkg-config))
     (home-page "https://projects.gnome.org/orbit2/")
     (synopsis "CORBA 2.4-compliant Object Request Broker")
     (description  "ORBit2 is a CORBA 2.4-compliant Object Request Broker (orb)
-- 
2.41.0




^ permalink raw reply related	[flat|nested] 12+ messages in thread

* [bug#69652] [PATCH gnome-team v2 2/3] gnu: orbit2: Propagate inputs.
  2024-03-09  8:47 ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Vivien Kraus via Guix-patches via
  2024-03-09  7:48   ` [bug#69652] [PATCH gnome-team v2 1/3] gnu: orbit2: Update style Vivien Kraus via Guix-patches via
@ 2024-03-09  7:52   ` Vivien Kraus via Guix-patches via
  2024-03-09  8:43   ` [bug#69652] [PATCH gnome-team v2 3/3] gnu: orbit2: Fix bug on 32-bit systems Vivien Kraus via Guix-patches via
  2024-03-09 17:50   ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Maxim Cournoyer
  3 siblings, 0 replies; 12+ messages in thread
From: Vivien Kraus via Guix-patches via @ 2024-03-09  7:52 UTC (permalink / raw)
  To: 69652; +Cc: rg, vivien, liliana.prikler, maxim.cournoyer

* gnu/packages/gnome.scm (orbit2) [inputs]: Move to 'propagated-inputs'.

Change-Id: I09e613cb4fee7962c50f62459ad6272bb7ffaea4
---
 gnu/packages/gnome.scm | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/gnu/packages/gnome.scm b/gnu/packages/gnome.scm
index fa911e1e59..8b69a231af 100644
--- a/gnu/packages/gnome.scm
+++ b/gnu/packages/gnome.scm
@@ -3827,7 +3827,8 @@ (define-public orbit2
               (substitute* "linc2/src/Makefile.in"
                 (("-DG_DISABLE_DEPRECATED")
                  "-DGLIB_DISABLE_DEPRECATION_WARNINGS")))))))
-    (inputs (list glib libidl))
+    ;; These are required in the installed pkg-config files.
+    (propagated-inputs (list glib libidl))
     (native-inputs (list pkg-config))
     (home-page "https://projects.gnome.org/orbit2/")
     (synopsis "CORBA 2.4-compliant Object Request Broker")
-- 
2.41.0




^ permalink raw reply related	[flat|nested] 12+ messages in thread

* [bug#69652] [PATCH gnome-team v2 3/3] gnu: orbit2: Fix bug on 32-bit systems.
  2024-03-09  8:47 ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Vivien Kraus via Guix-patches via
  2024-03-09  7:48   ` [bug#69652] [PATCH gnome-team v2 1/3] gnu: orbit2: Update style Vivien Kraus via Guix-patches via
  2024-03-09  7:52   ` [bug#69652] [PATCH gnome-team v2 2/3] gnu: orbit2: Propagate inputs Vivien Kraus via Guix-patches via
@ 2024-03-09  8:43   ` Vivien Kraus via Guix-patches via
  2024-03-09 17:50   ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Maxim Cournoyer
  3 siblings, 0 replies; 12+ messages in thread
From: Vivien Kraus via Guix-patches via @ 2024-03-09  8:43 UTC (permalink / raw)
  To: 69652; +Cc: rg, vivien, liliana.prikler, maxim.cournoyer

* gnu/packages/patches/orbit2-fix-array-allocation-32bit.patch: New file.
* gnu/local.mk (dist_patch_DATA): Register it here.
* gnu/packages/gnome.scm (orbit2): Use it here.

Change-Id: I72e10fc9d8a5421a7ec6b4ccf1425008b322ea3f
---
 gnu/local.mk                                  |  1 +
 gnu/packages/gnome.scm                        |  4 +-
 .../orbit2-fix-array-allocation-32bit.patch   | 40 +++++++++++++++++++
 3 files changed, 44 insertions(+), 1 deletion(-)
 create mode 100644 gnu/packages/patches/orbit2-fix-array-allocation-32bit.patch

diff --git a/gnu/local.mk b/gnu/local.mk
index dcd320fec4..9fbfc754a0 100644
--- a/gnu/local.mk
+++ b/gnu/local.mk
@@ -1781,6 +1781,7 @@ dist_patch_DATA =						\
   %D%/packages/patches/opentaxsolver-file-browser-fix.patch     \
   %D%/packages/patches/open-zwave-hidapi.patch			\
   %D%/packages/patches/orangeduck-mpc-fix-pkg-config.patch	\
+  %D%/packages/patches/orbit2-fix-array-allocation-32bit.patch \
   %D%/packages/patches/orpheus-cast-errors-and-includes.patch	\
   %D%/packages/patches/osip-CVE-2017-7853.patch			\
   %D%/packages/patches/ots-no-include-missing-file.patch	\
diff --git a/gnu/packages/gnome.scm b/gnu/packages/gnome.scm
index 8b69a231af..103065141e 100644
--- a/gnu/packages/gnome.scm
+++ b/gnu/packages/gnome.scm
@@ -3797,7 +3797,9 @@ (define-public orbit2
                                     upstream-name "-" version ".tar.bz2")))
               (sha256
                (base32
-                "0l3mhpyym9m5iz09fz0rgiqxl2ym6kpkwpsp1xrr4aa80nlh1jam"))))
+                "0l3mhpyym9m5iz09fz0rgiqxl2ym6kpkwpsp1xrr4aa80nlh1jam"))
+              (patches
+               (search-patches "orbit2-fix-array-allocation-32bit.patch"))))
     (build-system gnu-build-system)
     (arguments
      (list
diff --git a/gnu/packages/patches/orbit2-fix-array-allocation-32bit.patch b/gnu/packages/patches/orbit2-fix-array-allocation-32bit.patch
new file mode 100644
index 0000000000..377c09d0a5
--- /dev/null
+++ b/gnu/packages/patches/orbit2-fix-array-allocation-32bit.patch
@@ -0,0 +1,40 @@
+From 28067b88ad91e794675734bc92bd9c8653b4a46a Mon Sep 17 00:00:00 2001
+Message-ID: <28067b88ad91e794675734bc92bd9c8653b4a46a.1709973702.git.vivien@planete-kraus.eu>
+From: Vivien Kraus <vivien@planete-kraus.eu>
+Date: Sat, 9 Mar 2024 09:36:56 +0100
+Subject: [PATCH] corba-loc: Do not allocate more tokens than necessary.
+
+To split the list of locations, the code calls g_strsplit with the
+last argument set to G_MAXINT.  It means that g_strsplit will try to
+allocate a huge array (G_MAXINT + 1, for the final NULL), mostly
+filled with NULL.  Unfortunately, on 32-bit systems, this is one past
+the authorized length for an array.  Previous versions of glib would
+not care, but the new version now raises an error if this happens.
+
+To get an array of the appropriate size, we can just pass -1 to the
+last argument.
+
+* src/orb/orb-core/corba-loc.c (ORBit_corbaloc_parse): Replace
+G_MAXINT with -1.
+---
+ src/orb/orb-core/corba-loc.c | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/src/orb/orb-core/corba-loc.c b/src/orb/orb-core/corba-loc.c
+index abfcaa29..6492d392 100644
+--- a/src/orb/orb-core/corba-loc.c
++++ b/src/orb/orb-core/corba-loc.c
+@@ -309,7 +309,7 @@ ORBit_corbaloc_parse (const gchar       *corbaloc)
+ 	if (!(objkey = orbit_url_decode (okey)))
+ 		goto ret_error;
+ 	
+-	if (!(token = g_strsplit (loc, ",", G_MAXINT))) 
++	if (!(token = g_strsplit (loc, ",", -1)))
+ 		goto ret_error;
+ 
+ 	/* [ 'iiop' ]   ':' [ '//' ] [ version '@' ] host [ ':' port ] */
+
+base-commit: 144be2e9860286c83f009e7689250e0af977cc5e
+-- 
+2.41.0
+
-- 
2.41.0




^ permalink raw reply related	[flat|nested] 12+ messages in thread

* [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2
  2024-03-08 17:29 [bug#69652] [PATCH gnome-team 0/1] What should we do with orbit2? Vivien Kraus via Guix-patches via
  2024-03-08 17:20 ` [bug#69652] [PATCH gnome-team 1/1] gnu: orbit2: Mark it unsupported in 32-bits systems Vivien Kraus via Guix-patches via
@ 2024-03-09  8:47 ` Vivien Kraus via Guix-patches via
  2024-03-09  7:48   ` [bug#69652] [PATCH gnome-team v2 1/3] gnu: orbit2: Update style Vivien Kraus via Guix-patches via
                     ` (3 more replies)
  2024-03-09 15:57 ` [bug#69652] [PATCH gnome-team 0/1] What should we do with orbit2? Aaron Covrig via Guix-patches via
  2 siblings, 4 replies; 12+ messages in thread
From: Vivien Kraus via Guix-patches via @ 2024-03-09  8:47 UTC (permalink / raw)
  To: 69652; +Cc: rg, vivien, liliana.prikler, maxim.cournoyer

Dear Guix,

Our new glib was the first to not let orbit2 get away with terrible bad
practice.  I explained the situation in the patch.

Best regards,

Vivien

Vivien Kraus (3):
  gnu: orbit2: Update style.
  gnu: orbit2: Propagate inputs.
  gnu: orbit2: Fix bug on 32-bit systems.

 gnu/local.mk                                  |  1 +
 gnu/packages/gnome.scm                        | 61 ++++++++++---------
 .../orbit2-fix-array-allocation-32bit.patch   | 40 ++++++++++++
 3 files changed, 73 insertions(+), 29 deletions(-)
 create mode 100644 gnu/packages/patches/orbit2-fix-array-allocation-32bit.patch


base-commit: 25c14c893f05019d746321285acf55d1aa65b943
-- 
2.41.0




^ permalink raw reply	[flat|nested] 12+ messages in thread

* [bug#69652] [PATCH gnome-team 0/1] What should we do with orbit2?
  2024-03-08 17:29 [bug#69652] [PATCH gnome-team 0/1] What should we do with orbit2? Vivien Kraus via Guix-patches via
  2024-03-08 17:20 ` [bug#69652] [PATCH gnome-team 1/1] gnu: orbit2: Mark it unsupported in 32-bits systems Vivien Kraus via Guix-patches via
  2024-03-09  8:47 ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Vivien Kraus via Guix-patches via
@ 2024-03-09 15:57 ` Aaron Covrig via Guix-patches via
  2 siblings, 0 replies; 12+ messages in thread
From: Aaron Covrig via Guix-patches via @ 2024-03-09 15:57 UTC (permalink / raw)
  To: 69652

[-- Attachment #1: Type: text/plain, Size: 149 bytes --]

Is the version of glib on the gnome branch different from that on the
master branch? If so, what happens if you pin to the version on master?

Aaron

[-- Attachment #2: Type: text/html, Size: 195 bytes --]

^ permalink raw reply	[flat|nested] 12+ messages in thread

* [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2
  2024-03-09  8:47 ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Vivien Kraus via Guix-patches via
                     ` (2 preceding siblings ...)
  2024-03-09  8:43   ` [bug#69652] [PATCH gnome-team v2 3/3] gnu: orbit2: Fix bug on 32-bit systems Vivien Kraus via Guix-patches via
@ 2024-03-09 17:50   ` Maxim Cournoyer
  2024-03-09 18:40     ` bug#69652: " Liliana Marie Prikler
  3 siblings, 1 reply; 12+ messages in thread
From: Maxim Cournoyer @ 2024-03-09 17:50 UTC (permalink / raw)
  To: Vivien Kraus; +Cc: rg, 69652, liliana.prikler

Hi!

Vivien Kraus <vivien@planete-kraus.eu> writes:

> Dear Guix,
>
> Our new glib was the first to not let orbit2 get away with terrible bad
> practice.  I explained the situation in the patch.
>
> Best regards,
>
> Vivien
>
> Vivien Kraus (3):
>   gnu: orbit2: Update style.
>   gnu: orbit2: Propagate inputs.
>   gnu: orbit2: Fix bug on 32-bit systems.
>
>  gnu/local.mk                                  |  1 +

Reviewed-by: Maxim Cournoyer <maxim.cournoyer@gmail>

-- 
Thanks,
Maxim




^ permalink raw reply	[flat|nested] 12+ messages in thread

* bug#69652: [PATCH gnome-team v2 0/3] FIX that damn orbit2
  2024-03-09 17:50   ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Maxim Cournoyer
@ 2024-03-09 18:40     ` Liliana Marie Prikler
  0 siblings, 0 replies; 12+ messages in thread
From: Liliana Marie Prikler @ 2024-03-09 18:40 UTC (permalink / raw)
  To: Maxim Cournoyer, Vivien Kraus; +Cc: rg, 69652-done

Am Samstag, dem 09.03.2024 um 12:50 -0500 schrieb Maxim Cournoyer:
> Hi!
> 
> Vivien Kraus <vivien@planete-kraus.eu> writes:
> 
> > Dear Guix,
> > 
> > Our new glib was the first to not let orbit2 get away with terrible
> > bad practice.  I explained the situation in the patch.
> > 
> > Best regards,
> > 
> > Vivien
> > 
> > Vivien Kraus (3):
> >   gnu: orbit2: Update style.
> >   gnu: orbit2: Propagate inputs.
> >   gnu: orbit2: Fix bug on 32-bit systems.
> > 
> >  gnu/local.mk                                  |  1 +
> 
> Reviewed-by: Maxim Cournoyer <maxim.cournoyer@gmail>
And pushed.




^ permalink raw reply	[flat|nested] 12+ messages in thread

end of thread, other threads:[~2024-03-09 18:42 UTC | newest]

Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-03-08 17:29 [bug#69652] [PATCH gnome-team 0/1] What should we do with orbit2? Vivien Kraus via Guix-patches via
2024-03-08 17:20 ` [bug#69652] [PATCH gnome-team 1/1] gnu: orbit2: Mark it unsupported in 32-bits systems Vivien Kraus via Guix-patches via
2024-03-08 19:11   ` Liliana Marie Prikler
2024-03-08 20:03     ` Vivien Kraus via Guix-patches via
2024-03-09  6:36       ` Liliana Marie Prikler
2024-03-09  8:47 ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Vivien Kraus via Guix-patches via
2024-03-09  7:48   ` [bug#69652] [PATCH gnome-team v2 1/3] gnu: orbit2: Update style Vivien Kraus via Guix-patches via
2024-03-09  7:52   ` [bug#69652] [PATCH gnome-team v2 2/3] gnu: orbit2: Propagate inputs Vivien Kraus via Guix-patches via
2024-03-09  8:43   ` [bug#69652] [PATCH gnome-team v2 3/3] gnu: orbit2: Fix bug on 32-bit systems Vivien Kraus via Guix-patches via
2024-03-09 17:50   ` [bug#69652] [PATCH gnome-team v2 0/3] FIX that damn orbit2 Maxim Cournoyer
2024-03-09 18:40     ` bug#69652: " Liliana Marie Prikler
2024-03-09 15:57 ` [bug#69652] [PATCH gnome-team 0/1] What should we do with orbit2? Aaron Covrig via Guix-patches via

Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/guix.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).