[READY] Stop auto hover timer if buffer changes or cursor moves

[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][cont] document.
• I have read and understood YCM's [CODE_OF_CONDUCT][code] 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

[Please explain in detail why the changes in this PR are needed.]

Lack of tests is again just so I can preserve my sanity.
This was just reported on gitter. If the user changes buffer while we are running thte hover time, it causes a traceback.
[cont]: https://github.com/ycm-core/YouCompleteMe/blob/master/CONTRIBUTING.md
[code]: https://github.com/ycm-core/YouCompleteMe/blob/master/CODE_OF_CONDUCT.md

---

This change is 

[codecov]

Codecov Report

Merging #4193 (36a1a54) into master (1212def) will increase coverage by 11.36%.
The diff coverage is 47.82%.

Additional details and impacted files

@@             Coverage Diff             @@
##           master    #4193       +/-   ##
===========================================
+ Coverage   78.22%   89.59%   +11.36%     
===========================================
  Files          29       34        +5     
  Lines        2760     4440     +1680     
===========================================
+ Hits         2159     3978     +1819     
+ Misses        601      462      -139     

[bstaletic]

To be clear, this has never actually worked. For some reason, the auto hover just ends up not working at all.

[puremourning]

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

---

autoload/youcompleteme.vim line 770 at r2 (raw file):

  if g:ycm_auto_hover ==# 'CursorHold' && s:enable_hover
    call s:StopPoller( s:pollers.command )

This stops any outstanding async command request. I’m not sure that’s the right thing to do because the request might be for something else.

We probably need to do something like set a flag and ignore the response in the hover response handler?maybe set a flag on the poller object and check that in the handler. Wdyt?

[bstaletic]

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

---

autoload/youcompleteme.vim line 770 at r2 (raw file):
You are definitely right that this idea is not the right approach. It was very late at night when I had tried to implement a quick and dirty fix for this.

maybe set a flag on the poller object and check that in the handler.

That sounds like it might work, but the trouble is I could not find a way to repro the original issue - change buffer between firing an async request and receiving the response which ends up causing... was it a backtrace?

[bstaletic]

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

---

autoload/youcompleteme.vim line 770 at r2 (raw file):

Previously, bstaletic (Boris Staletic) wrote…

You are definitely right that this idea is not the right approach. It was very late at night when I had tried to implement a quick and dirty fix for this.

maybe set a flag on the poller object and check that in the handler.

That sounds like it might work, but the trouble is I could not find a way to repro the original issue - change buffer between firing an async request and receiving the response which ends up causing... was it a backtrace?

Yes, it's a traceback because b:ycm_hover might be undefined if the buffer changes.

Maybe we could just check if b:ycm_hover still exists when we enter s:ShowHoverResult?

[puremourning]

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

---

autoload/youcompleteme.vim line 770 at r2 (raw file):

Previously, bstaletic (Boris Staletic) wrote…

Yes, it's a traceback because b:ycm_hover might be undefined if the buffer changes.

Maybe we could just check if b:ycm_hover still exists when we enter s:ShowHoverResult?

I think that just papers over the crack. The issue is that when the cursor moves, we should not display the hover popup, period. Checking for b:ycm_hover just avoids a crash but doesn't really fix the issue.

I think we should do this:

• in s:pollers.command add a key which describes the request.
• here, only StopPoller if the request is a 'hover' request.

we could add a fairly generic "CancelCommand( expected_command_type )` too if you think that would be clearer.

[bstaletic]

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

---

autoload/youcompleteme.vim line 770 at r2 (raw file):

Previously, puremourning (Ben Jackson) wrote…

I think that just papers over the crack. The issue is that when the cursor moves, we should not display the hover popup, period. Checking for b:ycm_hover just avoids a crash but doesn't really fix the issue.

I think we should do this:

• in s:pollers.command add a key which describes the request.
• here, only StopPoller if the request is a 'hover' request.

we could add a fairly generic "CancelCommand( expected_command_type )` too if you think that would be clearer.

Done, but I would appreciate some manual testing.
From a brief messing around, seems to work.

[bstaletic]

Rebased as well.

[puremourning]

Reviewable status: 1 of 2 LGTMs obtained

[puremourning]

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

[mergify]

Thanks for sending a PR!