Reassigning bug to clutter as we're using GObject Introspection and the two constants are missing from the .gir file.

re-assigning to gobject-introspection.

the defines are documented, but I suspect that g-ir-scanner does not know how to handle TRUE and FALSE as bare values.

(In reply to comment #2)
> re-assigning to gobject-introspection.
> 
> the defines are documented, but I suspect that g-ir-scanner does not know how
> to handle TRUE and FALSE as bare values.

That's right, the scanner will just ignore anything that's not an integer or string, i.e. #define FOO_BAR 42 or #define FOO_BAZ "moo".

If we wanted to support this...I could relatively easily special case this to turn TRUE into an integer constant of 1, and FALSE to 0.  But here we get into subtlety - in some programming languages like Java, integers are not booleans.  You couldn't in such a binding say "return Clutter.CONTINUE" since it would barf trying to return 0 as true/false.

Now...maybe we could make this work if annotated (type gboolean) on the constant.  Let me see...

Actually I need to punt on this temporarily, but here's a stub for a test case:


diff --git a/tests/scanner/regress.h b/tests/scanner/regress.h
index 137c4e0..64bfc6f 100644
--- a/tests/scanner/regress.h
+++ b/tests/scanner/regress.h
@@ -273,6 +273,16 @@ GQuark regress_atest_error_quark (void);
 typedef guint64 RegressTestTypeGUInt64;
 #define REGRESS_GUINT64_CONSTANTA ((RegressTestTypeGUInt64) -1)
 
+/**
+ * REGRESS_IS_WORLD_ROUND: (type gboolean)
+ *
+ * https://bugzilla.gnome.org/show_bug.cgi?id=680387
+ *
+ * Add this to your code in dark, uncertain times to reaffirm your
+ * knowledge of reality.
+ */
+#define REGRESS_IS_WORLD_ROUND TRUE
+
 /* structures */
 typedef struct _RegressTestStructA RegressTestStructA;
 typedef struct _RegressTestStructB RegressTestStructB;

[Mass-moving gobject-introspection tickets to its own Bugzilla product - see bug 708029. Mass-filter your bugmail for this message: introspection20150207 ]

This has long been fixed by bug 719566.

*** This bug has been marked as a duplicate of bug 719566 ***