2017-09-17 0:31 GMT+09:00 Kazunobu Kuriyama <[email protected]>:

> 2017-09-16 23:49 GMT+09:00 Manuel Ortega <[email protected]>:
>
>> On Sat, Sep 16, 2017 at 10:31 AM, Bram Moolenaar <[email protected]>
>> wrote:
>>
>>>
>>> So I'll graduate FEAT_WINDOWS.  That removes 626 #ifdefs !!!
>>>
>>
>> According to the docs, once this happens the only thing differentiating
>> "tiny" from "small" will be "+visual".  But then those same docs also say
>> of "+visual" that "Always enabled since 7.4.200."   And I just tested
>> building a tiny build, and it resulted in "+visual".
>>
>> This appears to mean that once FEAT_WINDOWS is graduated, there will be
>> no difference between tiny and small builds (according to the docs).  If
>> that's the case, then one of "tiny" or "small" should be removed entirely.
>>
>> -Manny
>>
>
> For macOS, the darwin feature (+clipboard plus some character set
> conversions unique to it) also gives a substantial difference between
> "tiny" and "small."  This is because, with "tiny", since the feature is
> made disabled by configure even if the default '--enable-darwin" is
> explicitly specified on the command line, the default "darwin-enabled" Vim
> is actually the non-darwin version as far as the feature is concerned.
>

Today, when I was reviewing my work on feature list items for macOS, I
found the statement above is false.  The difference exists between "small"
and "normal", not between "tiny" and "small."

Apart from my ignorance, the cause of the confusion is due to some
inconsistency between configure, feature.h and vim.h.

The implementation of the darwin feature consists of two different files:
os_mac_conv.c and os_macosx.m.  For the former to compile, FEAT_MBYTE goes
with, and for the latter, FEAT_CLIPBOARD.  Those FEAT's are for feature
sets larger than 'small."

Nonetheless, if configure finds that --disable-darwin is not given on the
command line, it blindly has Makefile include them without care of the
FEAT's.  Consequently, the files are compiled in "tiny" and "small" for no
use.  (Actually, os_macosx.m has already been excluded from "tiny" to avoid
linker errors.)  They pretend to be darwin-featured until runtime.

So, there's actually a difference between "tiny" and "small", but that's a
thing which should not be there.

Thanks to links.sh, "tiny" doesn't suffers much for that.  But "small" is
unnecessarily made linked against the Cocoa framework. As a result, "small"
has +clipboard without +multibyte.  Clipboard only supporting ASII
characters?  Sounds not much useful.  Rather, it obviously has a risk of
crash.

To resolve the inconsistency mentioned above, I'm now updating my patch.
I'll soon send it to the list.

Kazunobu.


> What should we do with that?  Needless to say, the Vim which is
> pre-installed on every macOS machine is non-darwin. Personally, I think we
> should keep having versions which don't depend on macOS-specific API, i.e.,
> those of non-darwin.
>
> As to Bram' plan, I'm for it.
>
> Kazunobu.
>
>> --
>> --
>> You received this message from the "vim_dev" maillist.
>> Do not top-post! Type your reply below the text you are replying to.
>> For more information, visit http://www.vim.org/maillist.php
>>
>> ---
>> You received this message because you are subscribed to the Google Groups
>> "vim_dev" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>

-- 
-- 
You received this message from the "vim_dev" maillist.
Do not top-post! Type your reply below the text you are replying to.
For more information, visit http://www.vim.org/maillist.php

--- 
You received this message because you are subscribed to the Google Groups 
"vim_dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Raspunde prin e-mail lui