Re: vim-7.4.389 UI extremely sluggish with custom syntax highlighting plugins

Am 2014-08-14 01:30, schrieb Bram Moolenaar:
> Christian Brabandt wrote:
>
>> Hi Bram!
>>
>> On Mi, 13 Aug 2014, Bram Moolenaar wrote:
>>
>> >
>> > Christian Brabandt wrote:
>> >
>> > > > > On Sa, 09 Aug 2014, Bram Moolenaar wrote:
>> > > > > > David Barnett wrote:
>> > > > > >
>> > > > > > > Looks like it's patch 7.4.362:
>> > > > > > >
>> > > > > > > Problem: When matchaddpos() uses a length smaller than the number of
>> > > > > > > bytes
>> > > > > > > in the (last) character the highlight continues until the end of
>> > > > > > > the line.
>> > > > > > > Solution: Change condition from equal to larger-or-equal.
>> > > > > > > Files: src/screen.c
>> > > > > > >
>> > > > > > >
>> > > > > > > It's a very small change but apparently problematic. Can it be reverted or
>> > > > > > > rethought?
>> > > > > >
>> > > > > > I'm glad you could pinpoint it. I assumed that when the condition
>> > > > > > evaluates to true the match information would be updated. But perhaps
>> > > > > > that doesn't happen in this case and it searches for a match every time.
>> > > > > >
>> > > > > > Needs some debugging to figure out what happens. What matches does this
>> > > > > > plugin add when it's slow?
>> > > > >
>> > > > > Perhaps it is better to revert that patch and make sure, highlighting
>> > > > > gets only applied, if there is a match at the actual position?
>> > > > >
>> > > > > diff --git a/src/screen.c b/src/screen.c
>> > > [...]
>> > > >
>> > > > I haven't tried it, but it looks like this only avoids highlighting
>> > > > positions after the endcol when "v" and shl->endcol aren't exactly the
>> > > > same (which can happen for multi-byte characters), but it doesn't update
>> > > > the state, thus a next match might be missed.
>> > >
>> > > Yes, that is what patch 7.3.362 tried to fix, isn't it?
>> >
>> > So your patch doesn't work. Try this:
>> >
>> > highlight MyGroup ctermbg=green guibg=green
>> > let m = matchaddpos("MyGroup", [[4, 24, 2], [4, 30, 2]])
>> >
>> > " asdfasdfasdfasdfasdfas℅fasdfasdfasdfasdf
>> >
>> > Before pach 362 it would highlight everything from column 24.
>> > After the fix it looks OK, but we get complaints about slowness.
>> > After your fix only the first match shows up.
>>
>> I don't see that here (screenshot attached)
>
> Did you put a multi-byte character at the end of the first match?

Ah, now I can reproduce it.

>> But anyway, this raises the question again, why matchaddpos() accepts
>> a
>> list of positions (but only 8 items), while all other functions
>> usually
>> only accept 1 position.
>
> It's used for the matchparen plugin. The idea is that it's useful for
> other plugins as well.

Yes, but for consistency with the other match() functions and the VimL
API
it might still be a good idea to only accept a single position per
matchaddpos() call. It shouldn't be too hard, to adjust the matchparen
plugin
to call matchaddpos() twice, right?

Best,
Christian

--
--
You received this message from the "vim_use" 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_use" group.
To unsubscribe from this group and stop receiving emails from it, send an email to vim_use+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.