Re: Health update

2022-12-23 Fir de Conversatie puremo...@gmail.com
That's so great to hear. Glad you're feeling better, Bram. On Thursday, December 22, 2022 at 1:11:04 PM UTC Bram Moolenaar wrote: > > > Good evening everyone and Bram in particular. > > I just wanted to know if it was possible to have news of your health > which > > I hope has now fully recove

Re: Choices for Vim9 class implementation

2022-12-20 Fir de Conversatie puremo...@gmail.com
For what it's worth, I feel that the formal "var memberVariable : Type" syntax for member is (while more verbose) more likely to stand out. The `var` keyword is a signpost and so is likely to be syntax highlighted. Simple `memberVariable : Type` doesn't stand out anywhere and contains mostly id

Re: Zero cmdheight

2022-08-30 Fir de Conversatie puremo...@gmail.com
I tried it out and quite liked it initially. though I'm not sure if I would use it permanently. I think I would miss `showcmd` too much. That said, I did a quick GitHub search. As many users put their vimrc in GitHub we can search for it. This probably covers only users that either tried the ne

Re: def! in legacy vimscript

2022-02-18 Fir de Conversatie puremo...@gmail.com
> Wether a function is script-local or global (when not using a "s:" or "g:" prefix) depends on the script it's defined in > The rule is "at the script level, the type of script defines what the scope of the items is". I think that's easy to understand. You know, I think the other rule is argu

Re: [vim/vim] [FEATURE REQUEST] A solution for getting all the text properties in a buffer (Issue #9128)

2021-11-15 Fir de Conversatie puremo...@gmail.com
Bram, we _think_ the overhead is calling prop_list for every line, and that doing that in one call spanning the whole buffer would be more efficient. (currently we're doing a request per line, filtering the result for the types that start with Ycm, then adding them to a python list). That said,

Re: [vim/vim] [FEATURE REQUEST] A solution for getting all the text properties in a buffer (Issue #9128)

2021-11-15 Fir de Conversatie puremo...@gmail.com
We are not trying to modify them per se, in fact we're trying _not to_. YCM receives a batch of diagnostics, say: - A: an error on line 10 spanning { 10, 4 } to { 10, 8 }. - B: a warning on line 13 spanning { 13,1 } to { 14, 2 } YCM creates text props spanning those ranges with the some Ycm

Re: RFC - External debugger API for debugging vimscript

2021-06-15 Fir de Conversatie puremo...@gmail.com
Waking up this thread, just to say that this is still work in progress. I actually use it quite often for my own debugging, but I'm holding off on patches for now while vim9script develops. I think it makes logical sense for this all to happen after vim9script is mature. But as a teaser, I noti

Re: [vim/vim] proposal: allow popup col/line relative to the mouse position (#7645)

2021-01-10 Fir de Conversatie puremo...@gmail.com
Or popup_beval ? On Saturday, January 9, 2021 at 3:35:29 PM UTC Bram Moolenaar wrote: > Is there a reason you can't use getmousepos() ? > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub >

Re: [vim/vim] Compiling VIM with Huge and Python Results in Error: This version was built without iconv(). (#7642)

2021-01-09 Fir de Conversatie puremo...@gmail.com
> vim --version output is for macvim which is too old for what I need the latest MacVim is more than sufficient. On Saturday, January 9, 2021 at 10:13:44 AM UTC puremo...@gmail.com wrote: > > I am trying to install the latest version of VIM on macOS and compile it > with

Re: [vim/vim] Compiling VIM with Huge and Python Results in Error: This version was built without iconv(). (#7642)

2021-01-09 Fir de Conversatie puremo...@gmail.com
> I am trying to install the latest version of VIM on macOS and compile it with a different configuration. I want to do this because I'm trying to install vimspecto `brew install vim` is sufficient for that FYI On Saturday, January 9, 2021 at 4:03:0

Re: [vim/vim] Impossible to pass non-UTF-8 strings from vim to Python 3 (#1053)

2020-12-22 Fir de Conversatie puremo...@gmail.com
Bit off the wall, but perhaps we should provide a py3 api something like `vim.eval_bytes()` returning a `bytes` instance rather than trying to decode the evaluated bytes into a (unicode) `str` object. In the general case a vim string variable can contain any bytes (except NUL?); to read that in

Re: [vim/vim] complete_info().selected is -2 when 'menuone' is set and only 1 completion (#6945)

2020-10-26 Fir de Conversatie puremo...@gmail.com
I get the same behaviour. we get `-1` when it should be `0`. The original report was that it would return `-2` which was invalid by the spec. `-1` is just incorrect I think. On Monday, October 26, 2020 at 1:25:12 AM UTC csh wrote: > @puremourning I use vim(patc

Re: Vim9: rethinking conditions and boolean expressions

2020-10-05 Fir de Conversatie puremo...@gmail.com
> var name = Getname() ?? 'unknown' This `??` operator seems a bit unique to vim9script. If we're trying to be more like other languages how about either re-using `else` (as in Getname() else 'unknown') or using `or` (Like python): `Getname() or "unknown"` On Sunday, October 4, 2020 at 10:47:

Re: [vim/vim] termux ycm mouse openating (#7007)

2020-09-24 Fir de Conversatie puremo...@gmail.com
> I don't think that the ycm (YouCompleteMe) .. relevant here. Confirmed. YCM just calls complete() with completeopt+=noselect set On Thursday, September 24, 2020 at 12:01:03 PM UTC+1 Dominique Pellé wrote: > My understanding of this issue is that we cannot select an item in the pum > (popup

Re: [vim/vim] Request for vim9: Cast to boolean implicitly (#6903)

2020-09-09 Fir de Conversatie puremo...@gmail.com
Python's take on this is that if you want a bool, you use the bool constructor `bool( )`; we could add a vimscript function (like `string`) which takes an expr and returns either `v:true` or `v:false`, as in `return bool( 'string' )` or `return bool( 99 )`. The semantics of the type specified

Re: [vim/vim] Treat tmux as an xterm-like terminal (#6609)

2020-08-03 Fir de Conversatie puremo...@gmail.com
> logic were *configurable* It's pretty configurable now - you can set every t_* code manually. But really, isn't that literally what terminfo/termcap database and nottybuiltin is for ? As in, if you install the rxvt termcap and set 'nottybuiltin' then it's completely configurable. On Monday,