[RFC] cmdline mode autocommands #7422

bfredl · 2017-10-21T09:30:06Z

Define autocommands CmdlineEnter and CmdlineLeavefor entering and leaving cmdline mode. This will be very useful for inccommand-style plugins, ref #7370.

Note, I changed my mind since #4399, ModeChange does not make much sense as nvim is not a "finite state machine" in the named modes, rather a stack machine (proof: what happens when <c-r>= in cmdline? and <cr> right after that?). But what could make sense is generic ModeEnter and ModeLeave as suggesed by @ZyX-I #7416 (comment) . But I'm not entirely sure how to structure v:event then, I like that cmdline status is readily accessible in v:event without nesting. An alternative could of course be to put detailed info in cmdline_info() function (more detailed than getcmdtype(), it would then also be accessible in timers, "cmdline changed" callbacks: only colors now but callbacks for inccommand will follow, and also cmdline window)

Error propagation in CmdlineLeave is a bit weird, it typically executes the command but aborts processing some time soon after, probably errors should be echoed but not affect further processing at all.

chrisbra · 2017-10-21T10:40:13Z

Note https://github.com/vim/vim/releases/tag/v8.0.1206

bfredl · 2017-10-21T11:02:48Z

@chrisbra @brammool

augroup TEST
  au!
  au CmdLineLeave * echoerr "FAIL!"
augroup END

map ,x :let g:x = "x"<cr>:let g:y = "y"<cr>

Is it intentional that when invoking ,x and CmdlineLeave raises an error, that command line is still executed, but any later processing is aborted, i e g:x get set but not g:y ? Even though it was the g:x cmdline that raised the error? I would have expected both or none would be executed.

ZyX-I · 2017-10-21T15:20:54Z

For “not affecting further processing” I had to design try_enter/try_leave pair currently used while doing cmdline highlighting.

brammool · 2017-10-21T15:44:27Z

The error in CmdlineLeave does not affect executing commands.
It causes the mapping to be aborted.

brammool · 2017-10-21T15:53:07Z

We can make an error in CmdlineLeave stop execution of the command line, that seems to make sense. Not sure what to do with an error in CmdlineError, let's leave it alone for now.

brammool · 2017-10-21T17:05:49Z

Hmm, after trying this out, there is a serious flaw: If the CmdlineLeave autocmds fail, there is no way to execute any command. That includes removing the autocommand. So let's NOT do this.

bfredl · 2017-10-21T17:21:52Z

I agree, but then I think the mapping shouldn't be aborted either. Because it is even less related to the autocmd than the :-command...

brammool · 2017-10-21T18:59:42Z

Björn Linse wrote:

I agree, but then I think the mapping shouldn't be aborted either. Because it is even less related to the autocmd than the :-command...

Every error aborts a mapping. Would be inconsistent not to do that here.

…

-- ARTHUR: Then who is your lord? WOMAN: We don't have a lord. ARTHUR: What? DENNIS: I told you. We're an anarcho-syndicalist commune. We take it in turns to act as a sort of executive officer for the week. The Quest for the Holy Grail (Monty Python) /// Bram Moolenaar -- Bram@Moolenaar.net -- http://www.Moolenaar.net \\\ /// sponsor Vim, vote for features -- http://www.Vim.org/sponsor/ \\\ \\\ an exciting new programming language -- http://www.Zimbu.org /// \\\ help me help AIDS victims -- http://ICCF-Holland.org ///

bfredl · 2017-10-22T09:39:27Z

I wonder whether this is because it is somehow the "right" thing to do, or just because the vim codebase makes it hard to do otherwise. If I understand @ZyX-I correctly before he introduced try_enter/try_leave there was no reliable, reusable logic to do this.

bfredl · 2017-10-31T11:40:42Z

Marking RFC. I went for color_cmdline style error handling. Errors in autocmd should not abort unrelated commands/mappings. let v:event.aborting = 1 works to do it explicitly. (but not the other way around, trying to un-abort something already aborted seems like a can of worms not to open).

osa1 · 2017-11-01T04:19:45Z

What is the status of this?

bfredl · 2017-11-01T06:55:04Z

The is currently some discussion upstream vim/vim#2263 about the semantics of these events, we should probably wait until it is resolved. It can wait until 0.2.1 is released, let's not hurry.

bfredl · 2017-11-19T08:47:08Z

@justinmk Actually, the only reason this got delayed was waiting for vim/vim#2263 to get a resolution, but it seems the resolution implicitly was to "do nothing" (which anyway was the proper "resolution" IMO). The v:event part is already properly tested and documented so I see little reason to break this PR in two. But I removed some superfluous keys.

justinmk · 2017-11-20T23:41:03Z

@bfredl instead of "aborting" how about "cancel" or "canceled"? Otherwise LGTM.

bfredl · 2017-11-21T08:10:04Z

"abort*" seems to be more common than "cancel*" for stopping a command/function/mode (as) if an error/interrupt occured. How is "cancel" better?

justinmk · 2017-11-21T09:45:32Z

@bfredl "cancel" is more common in UI event systems that I know of (javascript, .NET winforms). I associate "aborting" with a program exit, etc.

KillTheMule · 2017-11-21T09:56:24Z

runtime/doc/eval.txt

@@ -1527,16 +1527,21 @@ v:event		Dictionary of event data for the current |autocommand|.  Valid
 <		Keys vary by event; see the documentation for the specific
 		event, e.g. |DirChanged| or |TextYankPost|.
 			KEY		DESCRIPTION ~
+			abort		whether the event triggered during
+					an aborting contidion, i e |c_Esc| or


bfredl · 2017-11-21T10:04:37Z

@justinmk I would indeed associate cancel with specifically user/UI initiated stopping, like pressing CTRL-C in cmdline. But vim seems to use "abort" most often when the condition is generic, i e could be either internal/vimL error, user interrupt or some kind of edge case which is not necessarily an error but the command/mode/action should not be completed.

If this key was only ever used for CmdlineLeave I would be inclined to agree with you, as CTRL-C likely is the most common cause in this case. But if keys are as I recall supposed to be generic for future events, which can have very different dynamics, I think v:event.abort is better.

vim-patch:fafcf0dd59fd patch 8.0.1206: no autocmd for entering or leaving the command line Problem: No autocmd for entering or leaving the command line. Solution: Add CmdlineEnter and CmdlineLeave. vim/vim@fafcf0d

justinmk · 2017-11-22T21:39:26Z

src/nvim/globals.h

@@ -1167,6 +1167,8 @@ EXTERN char_u e_dirnotf[] INIT(= N_(
 EXTERN char_u e_unsupportedoption[] INIT(= N_("E519: Option not supported"));
 EXTERN char_u e_fnametoolong[] INIT(= N_("E856: Filename too long"));
 EXTERN char_u e_float_as_string[] INIT(= N_("E806: using Float as a String"));
+EXTERN char_u e_autocmd_err[] INIT(=N_(
+    "E920: autocmd has thrown an exception: %s"));


@bfredl This should be in the 5000+ range. 920 in Vim is channel-related.

It used to be our range. (Jobstart etc) Or did we change our channel errors to be the in the same range as vim channel errors?

Ah, if you chose it for reason, so be it. No big deal.

Or did we change our channel errors to be the in the same range as vim channel errors?

I meant that Vim uses error 920. We don't, until now.

Feel free to change it. But how do we know the other nvim specific codes in 900 won't collide? Or did vim-dev avoid individual numbers in 900 if not the entire range?

@bfredl That's why we started using the 5000+ range, to avoid collisions.

I meant the existing numbers. Do we know/assume that vim-dev won't use them, or should they be changed before nvim 1.0?

They should be changed. In most cases probably just prepending a 5 should work.

FEATURES: 3cc7ebf #7234 built-in VimL expression parser 6a7c904 #4419 implement <Cmd> key to invoke command in any mode b836328 #7679 'startup: treat stdin as text instead of commands' 58b210e :digraphs : highlight with hl-SpecialKey #2690 7a13611 #8276 'startup: Let `-s -` read from stdin' 1e71978 events: VimSuspend, VimResume #8280 1e7d5e8 #6272 'stdpath()' f96d99a #8247 server: introduce --listen e8c39f7 #8226 insert-mode: interpret unmapped META as ESC 98e7112 msg: do not scroll entire screen (#8088) f72630b #8055 let negative 'writedelay' show all redraws 5d2dd2e win: has("wsl") on Windows Subsystem for Linux #7330 a4f6cec cmdline: CmdlineEnter and CmdlineLeave autocommands (#7422) 207b7ca #6844 channels: support buffered output and bytes sockets/stdio API: f85cbea #7917 API: buffer updates 418abfc #6743 API: list information about all channels/jobs. 36b2e3f #8375 API: nvim_get_commands 273d2cd #8329 API: Make nvim_set_option() update `:verbose set …` 8d40b36 #8371 API: more reliable/descriptive VimL errors ebb1acb #8353 API: nvim_call_dict_function 9f994bb #8004 API: nvim_list_uis 3405704 #7520 API/UI: forward option updates to UIs 911b1e4 #7821 API: improve nvim_command_output WINDOWS OS: 9cefd83 #8084, #8516 build/win: support MSVC ee4e1fd win: Fix reading content from stdin (#8267) TUI: ffb8904 #8309 TUI: add support for mouse release events in urxvt 8d5a46e #8081 TUI: implement "standout" attribute 6071637 TUI: support TERM=konsole-256color 67848c0 #7653 TUI: report TUI info with -V3 ('verbose' >= 3) 3d0ee17 TUI/rxvt: enable focus-reporting d109f56 #7640 TUI: 'term' option: reflect effective terminal behavior FIXES: ed6a113 #8273 'job-control: avoid kill-timer race' 4e02f1a #8107 'jobs: separate process-group' 451c48a terminal: flush vterm output buffer on pty output #8486 5d6732f :checkhealth fixes #8335 53f11dc #8218 'Fix errors reported by PVS' d05712f inccommand: pause :terminal redraws (#8307) 51af911 inccommand: do not execute trailing commands #8256 84359a4 terminal: resize to the max dimensions (#8249) d49c1dd #8228 Make vim_fgets() return the same values as in Vim 60e96a4 screen: winhl=Normal:Background should not override syntax (#8093) 0c59ac1 #5908 'shada: Also save numbered marks' ba87a2c cscope: ignore EINTR while reading the prompt (#8079) b1412dc #7971 ':terminal Enter/Leave should not increment jumplist' 3a5721e TUI: libtermkey: force CSI driver for mouse input #7948 6ff13d7 #7720 TUI: faster startup 1c6e956 #7862 TUI: fix resize-related segfaults a58c909 #7676 TUI: always hide cursor when flushing, never flush buffers during unibilium output 303e1df #7624 TUI: disable BCE almost always 249bdb0 #7761 mark: Make sure that jumplist item will not have zero lnum 6f41ce0 #7704 macOS: Set $LANG based on the system locale a043899 #7633 'Retry fgets on EINTR' CHANGES: ad60927 #8304 default to 'nofsync' f3f1970 #8035 defaults: 'fillchars' a6052c7 #7984 defaults: sidescroll=1 b69fa86 #7888 defaults: enable cscopeverbose 7c4bb23 defaults: do :filetype stuff unless explicitly "off" 2aa308c #5658 'Apply :lmap in macros' 8ce6393 terminal: Leave 'relativenumber' alone (#8360) e46534b #4486 refactor: Remove maxmem, maxmemtot options 131aad9 win: defaults: 'shellcmdflag', 'shellxquote' #7343 c57d315 #8031 jobwait(): return -2 on interrupt also with timeout 6452831 clipboard: macOS: fallback to tmux if pbcopy is broken #7940 300d365 #7919 Make 'langnoremap' apply directly after a map ada1956 #7880 'lua/executor: Remove lightuserdata' INTERNAL: de0a954 #7806 internal statistics for list impl dee78a4 #7708 rewrite internal list impl

marvim added the WIP label Oct 21, 2017

bfredl force-pushed the cmdlineenter branch from 6c3fda1 to 32f7c48 Compare October 21, 2017 10:31

bfredl force-pushed the cmdlineenter branch 2 times, most recently from bd45183 to 96575d6 Compare October 22, 2017 10:41

bfredl force-pushed the cmdlineenter branch from 96575d6 to b6925e7 Compare October 29, 2017 11:31

bfredl mentioned this pull request Oct 31, 2017

vim-patch:8.0.1238, 8.0.1250 (bugfix) #7463

Merged

bfredl force-pushed the cmdlineenter branch from b6925e7 to f27106f Compare October 31, 2017 11:31

bfredl changed the title ~~[WIP] cmdline mode autocommands~~ [RFC] cmdline mode autocommands Oct 31, 2017

marvim added RFC and removed WIP labels Oct 31, 2017

justinmk added this to the 0.2.2 milestone Oct 31, 2017

bfredl force-pushed the cmdlineenter branch 3 times, most recently from 8f2cc68 to 9c24ed6 Compare October 31, 2017 20:51

bfredl force-pushed the cmdlineenter branch from 9c24ed6 to 1e2e02d Compare November 10, 2017 20:43

justinmk modified the milestones: 0.2.2, 0.2.3 Nov 13, 2017

bfredl force-pushed the cmdlineenter branch from 1e2e02d to 24611e0 Compare November 19, 2017 08:38

bfredl force-pushed the cmdlineenter branch from 24611e0 to 0592593 Compare November 19, 2017 08:41

bfredl force-pushed the cmdlineenter branch 2 times, most recently from 7ab2767 to 9ccc42a Compare November 20, 2017 19:27

bfredl force-pushed the cmdlineenter branch from 9ccc42a to 9aa8421 Compare November 21, 2017 09:08

KillTheMule reviewed Nov 21, 2017

View reviewed changes

bfredl force-pushed the cmdlineenter branch 3 times, most recently from cf730fa to 0293f2f Compare November 22, 2017 07:23

cmdline: CmdlineEnter and CmdlineLeave autocommands

b461eb5

vim-patch:fafcf0dd59fd patch 8.0.1206: no autocmd for entering or leaving the command line Problem: No autocmd for entering or leaving the command line. Solution: Add CmdlineEnter and CmdlineLeave. vim/vim@fafcf0d

bfredl force-pushed the cmdlineenter branch from 0293f2f to b461eb5 Compare November 22, 2017 20:33

bfredl merged commit a4f6cec into neovim:master Nov 22, 2017

jszakmeister removed the RFC label Nov 22, 2017

justinmk reviewed Nov 22, 2017

View reviewed changes

bfredl added a commit that referenced this pull request Nov 23, 2017

FIXUP: duplicate error number in #7422

d9b3ebf

h-youhei mentioned this pull request Feb 2, 2018

Use CmdLineEnter/Leave to controll search mode h-youhei/vim-fcitx#2

Closed

Uh oh!

[RFC] cmdline mode autocommands #7422

[RFC] cmdline mode autocommands #7422

Uh oh!

Conversation

bfredl commented Oct 21, 2017

Uh oh!

chrisbra commented Oct 21, 2017

Uh oh!

bfredl commented Oct 21, 2017

Uh oh!

ZyX-I commented Oct 21, 2017

Uh oh!

brammool commented Oct 21, 2017

Uh oh!

brammool commented Oct 21, 2017

Uh oh!

brammool commented Oct 21, 2017

Uh oh!

bfredl commented Oct 21, 2017

Uh oh!

brammool commented Oct 21, 2017 via email

Uh oh!

bfredl commented Oct 22, 2017

Uh oh!

bfredl commented Oct 31, 2017

Uh oh!

osa1 commented Nov 1, 2017

Uh oh!

bfredl commented Nov 1, 2017

Uh oh!

bfredl commented Nov 19, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

justinmk commented Nov 20, 2017

Uh oh!

bfredl commented Nov 21, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

justinmk commented Nov 21, 2017

Uh oh!

KillTheMule Nov 21, 2017

Choose a reason for hiding this comment

Uh oh!

bfredl commented Nov 21, 2017

Uh oh!

justinmk Nov 22, 2017

Choose a reason for hiding this comment

Uh oh!

bfredl Nov 22, 2017

Choose a reason for hiding this comment

Uh oh!

justinmk Nov 22, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bfredl Nov 22, 2017

Choose a reason for hiding this comment

Uh oh!

justinmk Nov 22, 2017

Choose a reason for hiding this comment

Uh oh!

bfredl Nov 22, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

justinmk Nov 22, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

bfredl commented Nov 19, 2017 •

edited

Loading

bfredl commented Nov 21, 2017 •

edited

Loading

justinmk Nov 22, 2017 •

edited

Loading

bfredl Nov 22, 2017 •

edited

Loading

justinmk Nov 22, 2017 •

edited

Loading