[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Emacs Mac port

From: YAMAMOTO Mitsuharu
Subject: Re: Emacs Mac port
Date: Mon, 04 Jan 2016 09:38:34 +0900
User-agent: Wanderlust/2.14.0 (Africa) SEMI/1.14.6 (Maruoka) FLIM/1.14.8 (Shijō) APEL/10.6 Emacs/22.3 (sparc-sun-solaris2.8) MULE/5.0 (SAKAKI)

>>>>> On Sun, 3 Jan 2016 12:36:40 +0000, Bozhidar Batsov <address@hidden> said:

>> The size of the patch is not an issue; actually, the Mac specific
>> files are currently distributed as raw files rather than in the
>> patch form.  Also, most of the port-independent part has already
>> been committed to the emacs-25 branch.

> What exactly has been committed to the emacs-25 branch? I recall no major
> improvements to the cocoa port in Emacs 25.

* 30f4a89..: YAMAMOTO Mitsuharu 2015-12-31 Move variables to inner loop, 
preparing for Mac port merge
* 6ee327d..: YAMAMOTO Mitsuharu 2015-12-31 Add handle_user_signal_hook
* 47580e0..: YAMAMOTO Mitsuharu 2015-12-31 Avoid writing to purespace
* 0588be7..: YAMAMOTO Mitsuharu 2015-12-31 Remove unused variable
* 89e7483..: YAMAMOTO Mitsuharu 2015-12-31 * configure.ac: Find libxml2 headers 
in Xcode SDK dir on Darwin.
* 3b95e9c..: YAMAMOTO Mitsuharu 2015-12-31 Use posix_openpt instead of openpty 
on Darwin

These changes have been in the Mac port for a long time.  You wouldn't
notice anything unless you try to build Emacs without the Command Line
Developer Tools.  If you noticed any visible differences, then that
means I have injected a bug.

>> Maintaining a single version is much simpler especially when it
>> comes to processing bug reports, reproducing them, and trying fix
>> candidates.

> I'm not sure what would change if some of the functionality was
> moved to GNU Emacs. You still use it, as the basis for the Mac port
> and you always support just the latest version.

Do I need to elaborate this?  When I receive a bug report, I try to
reproduce the bug on my side.  If I have two variants of the Mac port,
restricted and fully-featured, then I have to try it for both of the
variants.  As is often the case, a typical bug report lacks clear
description of versions, configurations, and concrete steps to
reproduce, so it involves trial-and-error even for a single variant.
Also, I have to make sure that a fix candidate doesn't break either of
the variants.

                                     YAMAMOTO Mitsuharu

reply via email to

[Prev in Thread] Current Thread [Next in Thread]