After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 91614 - back patch for enum operator overloading
back patch for enum operator overloading
Status: RESOLVED FIXED
Product: gtkmm
Classification: Bindings
Component: build
2.0
Other opensolaris
: Normal normal
: ---
Assigned To: gtkmm-forge
gtkmm-forge
Depends on:
Blocks:
 
 
Reported: 2002-08-24 18:21 UTC by Michael v. Szombathely
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
the patch file (3.42 KB, patch)
2002-08-24 18:22 UTC, Michael v. Szombathely
none Details | Review
the proposed README.sun for the Forte C++ compiler (2.57 KB, text/plain)
2002-08-26 20:35 UTC, Michael v. Szombathely
  Details
the ChangeLog entry for README.sun (312 bytes, patch)
2002-08-26 20:38 UTC, Michael v. Szombathely
none Details | Review
REDAME.sun, some correction made as proposed (2.64 KB, text/plain)
2002-08-28 19:24 UTC, Michael v. Szombathely
  Details

Description Michael v. Szombathely 2002-08-24 18:21:43 UTC
I have removed all occurences of AttachOptions constructors around
operator|. This was introduced in gtk/src/table.hg, because of Forte C++
compiler's inability to overload enum operators.

With the Forte C++ compiler patch 111685-09 this bug was fixed by Sun.
Comment 1 Michael v. Szombathely 2002-08-24 18:22:57 UTC
Created attachment 10697 [details] [review]
the patch file
Comment 2 Murray Cumming 2002-08-24 19:36:45 UTC
Maybe it would be best not to remove them, just to make life easier
for people who don't have the patch, or who can't apply it.

Maybe you should start a README.sun where you could mention things
like this.
Comment 3 Michael v. Szombathely 2002-08-26 20:34:12 UTC
> Maybe it would be best not to remove them, just to make life easier
> for people who don't have the patch, or who can't apply it.

I think that at actual portability stage of Gtk-- it would be better
to bring all Forte C++ users to the same compiler patch level. On one
hand all compiler patches are publically available and on the other
hand this makes build verification much easier. And finally you can
not build Gtk-- with the unpatched compiler installation of Forte C++
5.3 (WS6U2).

I will provide back patches of Gtk-- portability changes only if they
depend on offically available compiler patches. Actually beside of
this case, there are still two other bug reports for Forte C++ pending.


Comment 4 Michael v. Szombathely 2002-08-26 20:35:58 UTC
Created attachment 10720 [details]
the proposed README.sun for the Forte C++ compiler
Comment 5 Michael v. Szombathely 2002-08-26 20:38:01 UTC
Created attachment 10721 [details] [review]
the ChangeLog entry for README.sun
Comment 6 Murray Cumming 2002-08-28 09:06:54 UTC
README.SUN:
  - autoconf and automake should not be required for building gtkmm, 
unless building from cvs.
  - gtkmm should never be called "Gtk--".
  - Is there a URL that shows the patch(es), with descriptions of the 
problems that they fix? If not, then people might at least like to 
see URLs to your bugzilla bugs that show that they are necessary. I 
think some people will not install the patches until they see the 
errors, so it's nice to see the solution to those errors.
Comment 7 Murray Cumming 2002-08-28 11:00:23 UTC
Applied patch 10697 because people need to apply some SUN patches anyway.
Comment 8 Michael v. Szombathely 2002-08-28 19:24:25 UTC
Created attachment 10773 [details]
REDAME.sun, some correction made as proposed
Comment 9 Michael v. Szombathely 2002-08-28 19:44:42 UTC
> README.SUN:
>   - autoconf and automake should not be required for building gtkmm, 
> unless building from cvs.

Both removed from requirements.

>   - gtkmm should never be called "Gtk--".

Changed. I did'nt know about this.

>   - Is there a URL that shows the patch(es), with descriptions of the 
> problems that they fix? If not, then people might at least like to 
> see URLs to your bugzilla bugs that show that they are necessary. I 
> think some people will not install the patches until they see the 
> errors, so it's nice to see the solution to those errors.

Well, the URL I gave is the entry point to get all of the compiler
patches. All patches come with descriptions which mention the bug id's
they resolve, But the bug descriptions itself are not available.

I really disagree with you at the point to make all requirements
transparent to the Forte C++ users. I think it is enough to describe a
reasonable setup which works for gtkmm. The commercial UNIX field is
somewhat different from the Open Source OS's. Companies used to have
support from their vendors or hire staff for solving software problems.
Comment 10 Murray Cumming 2002-08-29 08:24:13 UTC
Yes, if they need to apply some patches then they might as well apply
all patches. But in general I think it is optimistic to expect people
to read our documentation. In case they don't read it, I like to make
sure that they find the answers to the problems that they will then
encounter. 

README.SUN added. Thanks.