Allow soft-wrapping of lines in the signature help window

[bstaletic]

PR Prelude

Thank you for working on YCM! :)

Please complete these steps and check these boxes (by putting an x inside
the brackets) before filing your PR:

• I have read and understood YCM's CONTRIBUTING document.
• I have read and understood YCM's CODE_OF_CONDUCT document.
• I have included tests for the changes in my PR. If not, I have included a
  rationale for why I haven't.
• I understand my PR may be closed if it becomes obvious I didn't
  actually perform all of these steps.

Why this change is necessary and useful

Fixes #4171

Or at least proposes a fix. Here's a short demo:
https://asciinema.org/a/UJlR80blt6OvuwUIqb4swnWUW

This looks ugly, but at least no data is lost due to lines going off screen.

---

This change is 

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 89.88%. Comparing base (f56c0cb) to head (05f688c).
Report is 2 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #4252      +/-   ##
==========================================
+ Coverage   89.80%   89.88%   +0.07%     
==========================================
  Files          37       37              
  Lines        4758     4763       +5     
==========================================
+ Hits         4273     4281       +8     
+ Misses        485      482       -3     

[puremourning]

I tried this out and I quite liked it but there were cases where it was I think net worse than what we have now.
This is a shame because it's just really hard to solve for both problems of horizontal positioning being "sane" and inducing as much info on the screen as possible. tricky tricky.

Reviewable status: 0 of 2 LGTMs obtained

[bstaletic]

Right, the shifting to the left is broken in this pull request.
Let's leave it for now and maybe I'll get back to it during the weekend.

[hippie68]

Thank you for working on this. Today I was searching for how to enable this feature only to find out that it doesn't exist yet.

[bstaletic]

I'll be getting back to this pull request, this week.
My idea is to only set wrap when the popup is all the way left in column 1. We'll see how that goes.

[bstaletic]

I think I got this working.
@hippie68 would you mind trying this branch/patch?

[hippie68]

This latest update works well for me. I tested it with those overly long Python function signatures, which now wrap properly. It works like I would have expected YouCompleteMe's signature help window to work out of the box.

[hippie68]

There's one minor cosmetic issue left: when the text wraps, there should be a space appended at the end of each wrapping line. At least that's the style all other pop-up windows have. I'm not sure if that is possible here though.

In the example screenshot, one space would be inserted after "Opt" and "flus" each to have the right pop-up border width match the left one.

[hippie68]

Never mind, I just noticed that this is normal wrapping behavior even for docstrings. Which brings me to the question if there is anything else left that is stopping this pull request from being merged. I have been trying this feature for quite some time now and I would love to see it included in YCM by default.

[puremourning]

Reviewable status: 0 of 2 LGTMs obtained (waiting on @bstaletic)

---

python/ycm/signature_help.py line 162 at r4 (raw file):

    col = int( screen_pos[ 'curscol' ] ) - 2

  # Vim stops shifting the popup to the left if we turn on soft-wrapping.

I think this is just lazy programming by the absolute waster that implemented this left-shifting in Vim.

https://github.com/vim/vim/blob/master/src/popupwin.c#L1437-L1446

vim/vim#4466

Part of me is like "we should change the Vim behaviour rather than work it out here because, well, that was the whole reason that I implemented it in Vim rather than here"....

Basically by enabling 'wrap' we have to implement that PR's logic in here instead: https://github.com/vim/vim/blob/master/src/popupwin.c#L1447-L1466

---

python/ycm/signature_help.py line 165 at r4 (raw file):

  # Instead, we want to first shift the popup to the left and then
  # and then turn on wrapping.
  max_line_length = len( max(

marginally more efficient (and simpler) to write:

max( len( item[ 'text' ] ) for item in buf_lines )

---

python/ycm/signature_help.py line 169 at r4 (raw file):

    key = lambda item: len( item[ 'text' ] ) )[ 'text' ] )
  vim_width = vimsupport.GetIntValue( '&columns' )
  line_available = vim_width - int( screen_pos[ 'col' ] )

Shouldn't this be col here rather than screen_pos[ 'col' ] ? (assuming that col > 0).

in the previous stanza we offset the col position according to some padding and positioning logic. I think that this calculation needs to account for both sides of the padding.

[bstaletic]

Sorry for the delay, everyone. I'm getting back to my pull requests.

@hippie68 I see what you mean regarding the "missing" space at the end, but that would be really awkward to work out.

Reviewable status: 0 of 2 LGTMs obtained (waiting on @puremourning)

---

python/ycm/signature_help.py line 162 at r4 (raw file):

Previously, puremourning (Ben Jackson) wrote…

I think this is just lazy programming by the absolute waster that implemented this left-shifting in Vim.

https://github.com/vim/vim/blob/master/src/popupwin.c#L1437-L1446

vim/vim#4466

Part of me is like "we should change the Vim behaviour rather than work it out here because, well, that was the whole reason that I implemented it in Vim rather than here"....

Basically by enabling 'wrap' we have to implement that PR's logic in here instead: https://github.com/vim/vim/blob/master/src/popupwin.c#L1447-L1466

I'd say we go with this implementation for now and once vim implements the behaviour that allows shifting with wrap, we can start relying on that.

---

python/ycm/signature_help.py line 165 at r4 (raw file):

Previously, puremourning (Ben Jackson) wrote…

marginally more efficient (and simpler) to write:

max( len( item[ 'text' ] ) for item in buf_lines )

Done.

---

python/ycm/signature_help.py line 169 at r4 (raw file):

Previously, puremourning (Ben Jackson) wrote…

Shouldn't this be col here rather than screen_pos[ 'col' ] ? (assuming that col > 0).

in the previous stanza we offset the col position according to some padding and positioning logic. I think that this calculation needs to account for both sides of the padding.

I honestly am not sure. What you're saying makes sense and I did make the change.
Brief testing seems to indicate that this works well.

[puremourning]

Reviewed 1 of 2 files at r3, 1 of 1 files at r5, all commit messages.
Reviewable status: 0 of 2 LGTMs obtained (waiting on @bstaletic)

---

python/ycm/signature_help.py line 169 at r4 (raw file):

Previously, bstaletic (Boris Staletic) wrote…

I honestly am not sure. What you're saying makes sense and I did make the change.
Brief testing seems to indicate that this works well.

I think at this point there is a weird case where col is 0 or negative. Maybe line_available = vim_width - max( col, 1 ) ?

[bstaletic]

Reviewable status: 0 of 2 LGTMs obtained (waiting on @puremourning)

---

python/ycm/signature_help.py line 169 at r4 (raw file):

Previously, puremourning (Ben Jackson) wrote…

I think at this point there is a weird case where col is 0 or negative. Maybe line_available = vim_width - max( col, 1 ) ?

Done.

[puremourning]

Reviewed 1 of 1 files at r6, all commit messages.
Reviewable status: 1 of 2 LGTMs obtained (waiting on @bstaletic)

[mergify]

Thanks for sending a PR!