diff options
| author | Martin Pitt <martinpitt@gnome.org> | 2012-05-02 12:49:25 +0200 |
|---|---|---|
| committer | Martin Pitt <martinpitt@gnome.org> | 2012-05-02 12:49:25 +0200 |
| commit | be867f9658b64ee0a93f7730929b421e37b911f6 (patch) | |
| tree | 00ace6da3056cb5de0e61c16f8d55a96dc9f1869 /tests/scanner/regress.h | |
| parent | d7ff7296d4bf661b165dc5975754c70ec8610eec (diff) | |
| download | gobject-introspection-be867f9658b64ee0a93f7730929b421e37b911f6.tar.gz | |
Fix GVariant handling in regress_test_closure_variant
Commit 7c4fbbd6 introduced some low-level g_memdup()ing of the passed GVariant
in regress_test_closure_variant(). This causes a race condition with GI clients
like pygobject's test suite, which sometimes fail with
(runtests.py:15653): GLib-CRITICAL **: g_variant_ref_sink: assertion `value->ref_count > 0' failed
Drop the memduping again, and instead make the argument non-const (which is
true, as the method temporarily increases the refcount).
Diffstat (limited to 'tests/scanner/regress.h')
| -rw-r--r-- | tests/scanner/regress.h | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/tests/scanner/regress.h b/tests/scanner/regress.h index c2a25c13..15b8ce2e 100644 --- a/tests/scanner/regress.h +++ b/tests/scanner/regress.h @@ -122,7 +122,7 @@ GPtrArray *regress_test_garray_full_return (void); /* closure */ int regress_test_closure (GClosure *closure); int regress_test_closure_one_arg (GClosure *closure, int arg); -GVariant *regress_test_closure_variant (GClosure *closure, const GVariant* arg); +GVariant *regress_test_closure_variant (GClosure *closure, GVariant* arg); /* value */ int regress_test_int_value_arg(const GValue *v); |