On 12/05/08 09:17, Bill McCarthy wrote:
> On Sun 11-May-08 9:14pm -0600, Tony Mechelynck wrote:
>> On 11/05/08 22:38, Bram Moolenaar wrote:
>>> Bill McCarthy wrote:
>
>>>> My source is now patched through 298 but I haven't rebuilt
>>>> yet. The separate source with the floating point patch is
>>>> only at 293.
>>>>
>>>> I see the latest official runtime now has floating point in
>>>> the docs, yet the official patches don't appear to include
>>>> floating point yet.
>>>>
>>>> What's the plan?
>
>>> There are a few fixes for floating point in the pipeline. I'll send out
>>> an updated patch when I have included them. But I'm very busy the
>>> coming week, don't hold your breath.
>
>> In the meantime, you can apply the "official" patches even after the
>> "unofficial" patch for floating point, they will succeed (with a slight
>> fuzz in the case of version.c). My current Vim displays a ":version"
>> text as follows:
>
> Thanks for your note, Tony. I've updated my separate
> floating point tree of \src with 298 and applied Bram's
> patch (after removing the eval.txt portion), then John
> Beckett's patch to fix a few problems.
>
> The patch went fine - I believe the only adjustments made
> were to the Unix stuff (files with 'config' in the names)
> not needed for WinXP.
These "config" files are part of the "configure" process which, on
Unix-like systems, determines which compile-time options will be
included considering what you want to include (i.e., user-set
parameters) and what you can afford to include (i.e., which software can
be found on your machine). The same process checks which variant of
which compiler (and other software) you're using, which functions are
available or buggy, etc., and sets the appropriate #DEFINEs in function
of that.
On Windows, a different Makefile is used (one for each compiler,
actually), bypassing the configure step, which means that there is less
automatic detection (e.g. if you include Ruby you'll have to pass the
Ruby version two different ways to the make process).
One file is modified with every patch, namely version.c which includes
the list of included features and patches. You'll see that from now on,
its patch step will succeed with a discrepancy of five lines
corresponding to
#ifdef FEAT_FLOAT
"+float",
#else
"-float",
#endif
(220-224) which are of course not present in the "standard" sources.
That discrepancy is normal and no cause for alarm.
Best regards,
Tony.
--
Of all possible committee reactions to any given agenda item, the
reaction that will occur is the one which will liberate the greatest
amount of hot air.
-- Thomas L. Martin
--~--~---------~--~----~------------~-------~--~----~
You received this message from the "vim_dev" maillist.
For more information, visit http://www.vim.org/maillist.php
-~----------~----~----~----~------~----~------~--~---
