[Top][All Lists]

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

Re: Windows 64 port

From: Paul Eggert
Subject: Re: Windows 64 port
Date: Wed, 29 Feb 2012 22:28:49 -0800
User-agent: Mozilla/5.0 (X11; Linux i686; rv:10.0.2) Gecko/20120216 Thunderbird/10.0.2

On 02/29/2012 08:03 PM, Eli Zaretskii wrote:
> There's nothing bogus about using the right data type

But the code in question is already using a correct data type.
That is, for the diffs marked "Not needed for Windows 64" in
Emacs is already using data types that work,
both on mainstream platforms and on Windows 64.

>  or declare a variable or an argument `const' when it is.

Suppose someone proposed lots of patches like this:
-  size_t len = strlen (string);                                                
+  size_t const len = strlen (string);

on the grounds that the XYZ Corp. compiler warns whenever
code fails to use 'const' at every opportunity.
That would be bogus -- that's not the Emacs
programming style, and we shouldn't slavishly alter
mainstream code merely to pacify a third-party compiler
that prefers a different style.

> I asked several times to do the same for texinfo.tex when you sync with
> gnulib, and you declined.

Those patches were automatically merged from an already-debugged
and well-modularized upstream, and it's trivial to see which parts
affect texinfo.tex.

In contrast, here we're dealing with a handwritten patch that is
tickling core Emacs internals in tricky, low-level ways, a patch that
almost surely still contains bugs, partly because it is trying to do
too much at once.  It's entirely appropriate to spend more time with
such a patch, to see which parts are really needed for Windows 64, and
to isolate the changes as much as possible to the Windows-specific
code base.

reply via email to

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