This maintenance release fixes some issues found in v0.3.2 .
FIXES:
a597ab8d1b#9442 Merge pull request from jamessan/revert-pynvim
d7b3ac029c health/provider: Check for available pynvim when neovim module missing
edeb19d5e9 python#CheckForModule: Use the given module string instead of hard-coding pynvim
0dd89cda9c {health,provider}/python: Import the neovim, rather than pynvim, module
fc6e8a4db8#9423 TUI: Konsole DECSCUSR fixup
The neovim module is available for backwards compatibility. We should
not yet force the use of the pynvim module, since there's no other major
reason to bump the minimum supported Python client module.
Closes#9426
Besides the "visible" improvements, this release features numerous
internal improvements to the UI/screen code and test infrastructure.
Numerous patches were merged from Vim, which are not mentioned below.
FEATURES:
07ad5d71ab clipboard: Support custom VimL functions #9304725da1feeb#9401 win/TUI: Improve terminal/console support
7a8dadbedb#9077 startup: Use $XDG_CONFIG_DIRS/nvim/sysinit.vim if it exists
feec926633#9299 support <cmd> mapping in more places
0653ed63a5#9028 diff/highlight: Show underline for low-priority CursorLine
bddcbbb571 signs: Add "numhl" argument #911305f9c7c2f7 clipboard: support Wayland (#9230)
14ae394532#9052 TUI: add support for undercurl and underline color
4fa3492a6f#9023 man.vim: soft (dynamic) wrap #9023
API:
8b39e4ec79#6920 API: implement object namespaces
b1aaa0a881 API: Implement nvim_win_set_buf() #91008de87c7b1c#8180 API: virtual text annotations (nvim_buf_set_virtual_text)
2b9fc9a13f#8660 API: add nvim_buf_is_loaded()
API: buf_get_lines, buf_line_count handle unloaded buffers
88f77c28e5 API: nvim_buf_get_offset_for_line
94841e5eae API/UI: #8221 ext_newgrid, ext_hlstate
(use line-based rather than char-based updates)
UI
b5cfac0894#8806 TUI: use BCE again more often, (smoother resizes/scrolling)
77b5e9ae25#9315 screen: add missing status redraw when redraw_later(CLEAR) was used
5f15788dc3 TUI: clip invalid regions on resize (#8779), fixes#8774c936ae0f36#9193 TUI: improvements for scrolling and clearing
f20427451e#9143 UI: disable clearing almost everywhere
f4b2b66661#9079 TUI: always use safe cursor movement after resize
d36afafc8d#9211 ui_options: also send when starting or from OptionSet
67f80d485c TUI: Avoid reset_cursor_color in old VTE #9191e55ebae373#9021 don't erase screen on `:hi Normal` during startup
c5790d9189#8915 TUI: Hint wrapped lines to terminals.
FIXES:
231de72539 RPC: turn errors from async calls into notifications
907ad921bc TUI: Restore terminal title via "title stacking" (#9407)
cb76a8a95f genappimage: Unset $ARGV0 at invocation #9376b48efd9ba7#9347 TUI: FreeBSD: Improve support for BSD vt console
c16529afa5 TUI: Konsole 18.07.70 supports DECSCUSR (#9364)
aec096fc5b os/lang: use the correct LC_NUMERIC also for OS X
5fee0be915 provider: improve error message (#9344)
3c42d7a10a TUI: alacritty supports set_cursor_color #93537bff9a5de8 TUI: Alacritty supports DECSCUSR (#9048)
57acfceabe macOS: infer primary language if $LANG is empty #9345bc132ae123 runtime/syntax: Fix highlighting of augroup contents (#9328)
715fdfee1e#9297 VimL/confirm(): Show dialog even if :silent
799d9c3215 clipboard: Prefer xclip (#9302)
6dae7776ed provider/nodejs: fix npm,yarn detection
16bc1e9c17#9218 channel: avoid buffering output when only terminal and no callbacks are active
72fecad1ff#8804 Fix crash in lang_init() on macOS if lang_region = NULL
d581398779 ruby: detect rbenv shims for other versions (#8733)
e568ac7a68#9123 third-party/unibilium: Fix parsing of extended capability entries
c4c74c3883 jobstart(): Fix hang on non-executable cwd #92041cf50cbfd9 provider/nodejs: Simultaneously query npm and yarn #90546c496db4b7 undo: Fix infinite loop if undo_read_byte returns EOF #2880f8f83579ff#9034 'swapfile: always show dialog'
CHANGES:
c236e80cf3#9024 --embed: wait for UI unless --headless
180b50dddc#9248 python: 'neovim' module was renamed to 'pynvim'
2000b6a64a#8589 VimL: Remove legacy aliases "v:errmsg", "v:shell_error", "v:this_session"
deb18a050e defaults: background=dark #2894 (#9205)
c1187d4af0 defaults: win: 'shellpipe' for cmd.exe (#8827)
vim-patch:8.0.0428: git and hg see new files after running tests
vim-patch:8.0.0429: options test does not always test everything
vim-patch:8.0.0430: options test fails or hangs on MS-Windows
vim-patch:8.0.0436: running the options test sometimes resizes the terminal
vim-patch:8.0.0488: running tests leaves an "xxx" file behind
vim-patch:8.0.0585: test_options fails when run in the GUI
vim-patch:8.1.0558: some MS-Windows instructions are outdated
vim-patch:8.1.0565: asan complains about reading before allocated block
vim-patch:8.1.0566: SGR not enabled for mintty because $TERM is "xterm"
vim-patch:8.1.0567: error for NUL byte in ScreenLines goes unnoticed
vim-patch:8.1.0568: error message for NUL byte in ScreenLines breaks Travis CI
vim-patch:8.1.0577: tabpage right-click menu never shows "Close tab"
vim-patch:8.1.0578: cannot disable arabic, rightleft and farsi in configure
vim-patch:8.1.0587: GvimExt: realloc() failing is not handled properly
vim-patch:8.1.0589: compilation error in gvimext.cpp
vim-patch:8.1.0592: the libvterm tests are not run as part of Vim tests
vim-patch:8.1.0593: illegal memory access in libvterm test
vim-patch:8.1.0594: libvterm tests fail to run on Mac
vim-patch:8.1.0595: libvterm tests are not run with coverage
vim-patch:8.1.0597: cannot run test_libvterm from the top directory
vim-patch:8.1.0598: indent tests may use the wrong Vim binary
vim-patch:8.1.0605: running make in the top directory echoes a comment
vim-patch:8.1.0606: 'cryptmethod' defaults to a very old method
vim-patch:8.1.0607: proto files are not in sync with the source code
vim-patch:8.1.0608: coverals is not updating
vim-patch:8.1.0609: MS-Windows: unused variable, depending on the Ruby version
vim-patch:8.1.0610: MS-Windows ctags file list differs from Unix
vim-patch:8.1.0616: NSIS installer is outdated
vim-patch:8.1.0617: NSIS installer gets two files from the wrong directory
vim-patch:8.1.0620: overuling CONF_ARGS from the environment no longer works
vim-patch:8.1.0624: overuling CONF_ARGS from the environment still does not work
vim-patch:8.1.0628: Compiler warning on MS-Windows.
vim-patch:8.1.0635: Coverity complains about null pointer use
vim-patch:8.1.0637: nsis file no longer used
vim-patch:8.1.0646: cannot build with Ruby 2.6.0
The purpose of the {Nvim} hint was not well-defined, and its usage
inconsistent. It's also unnecessary.
Nvim-Vim differences are centralized at:
:help vim-differences
Removed things are centralized at:
:help deprecated
Developer guidelines for documentation are listed at:
:help dev-doc
We already set back_color_erase in our builtin terminfo
(terminfo_defs.h:conemu_terminfo), so we don't need to set it explicitly
in patch_terminfo_bugs().
Problem: An "after" directory of a package is appended to 'rtp', which
will be after the user's "after" directory. ()
Solution: Insert the package "after" directory before any other "after"
directory in 'rtp'. (closesvim/vim#3409)
99396d4cbf
Problem: Package directory not added to 'rtp' if prefix matches.
Solution: Check the match is a full match. (Ozaki Kiichi, closesvim/vim#2817)
Also handle different ways of spelling a path.
f98a39ca57
Problem: When package path is a symlink adding it to 'runtimepath' happens
at the end.
Solution: Do not resolve symlinks before locating the position in
'runtimepath'. (Ozaki Kiichi, closesvim/vim#2604)
2374faae11
Works in iTerm2 and xterm.
- patch_terminfo_bugs(): Add title-stacking sequences (\e[22t and
\e[23t) to the overrides for iTerm2 and xterm.
- Note: the builtins (terminfo_defs.h) for `iterm_256colour_terminfo`
and `xterm_256colour_terminfo` already include the sequences.
Test case (title should be "foo" after the final :q):
TERM=iterm2 nvim -u NONE +'set title titleold=foo'
:q
TERM=iterm2 nvim -u NONE +'set title titlestring=zub'
:q
closes#4063
Use uv_set_vterm_state() to override libuv's guess.
See https://github.com/libuv/libuv/pull/1873/ for discussion.
This commit uses a terminal-detection approach based on
GetProcessImageFileNameW(...), which will be reverted in the following
commit. The approach was intended to handle the case of running in
winpty (:terminal), but we will add $NVIM env var for that.
Also add some support for ConEmu, cygwin.
Problem: Python cannot handle function name of script-local function.
Solution: Use <SNR> instead of the special byte code. (Ozaki Kiichi, closes
vim/vim#3681)
9123c0b31a
ref #9280
Introduce the `vim.compat` module, to help environments with system Lua
5.2+ run the build/tests. Include the module implicitly in all tests.
ref #8677
legacy `vim` module:
beep
buffer
command
dict
eval
firstline
lastline
line
list
open
type
window
Problem: Cannot act on edits in the command line.
Solution: Add the CmdlineChanged autocommand event. (xtal8, closesvim/vim#2603,
closesvim/vim#2524)
153b704e20
In Vim (and some vestigial parts of Nvim) E319 was a placeholder for
ex_ni commands, i.e. commands that are only available in certain builds
of Vim. That is obviously counter to Nvim's goals: all Nvim commands
are available on all platforms and build types (the remaining ex_ni
commands are actually just missing providers).
We need an error id for "missing provider", so it makes sense to use
E319 for that purpose.
ref #9344
ref #3577
Executing `:python`, and similar commands that rely on `eval_call_provider()`,
while the accompanying provider it not available, leads to this error message:
E117: Unknown function: provider#python#Call
This doesn't say much to a user. Since we introduced `:checkhealth` for this
very reason, we now point to it for further diagnosis.
Fixes#3577
The macOS preferences have a section called `Language & Region`. There is always
at least one language defined, the primary language.
CFLocaleCopyPreferredLanguages() returns the languages defined in that section,
the first element being the primary language.
Use the primary language in case CFLocaleCopyCurrent() returns NULL.
In the case that the above fallback does not work either, which is very
unlikely, log the error and continue with an empty $LANG.
References #9134
Problem: 'commentstring' not used when adding fold marker in C.
Solution: Require white space before middle comment part. (mostly by
Hirohito Higashi)
4af7259b2b
Problem: Using an external diff program is slow and inflexible.
Solution: Include the xdiff library. (Christian Brabandt)
Use it by default.
e828b7621c
vim-patch:8.1.0360
vim-patch:8.1.0364
vim-patch:8.1.0366
vim-patch:8.1.0370
vim-patch:8.1.0377
vim-patch:8.1.0378
vim-patch:8.1.0381
vim-patch:8.1.0396
vim-patch:8.1.0432
Problem: 'commentstring' not used when adding fold marker. (Maxim Kim)
Solution: Only use empty 'comments' middle when leader is empty. (Christian
Brabandt, closesvim/vim#3670)
539328197c
Comparing `vimCommand` from Vim's runtime/syntax/vim.vim, one can see
that "augroup" and similar commands are conspicuously missing. They are
handled specially (`vimAugroupKey`, `vimAutoCmd`).
Excluding them from the generated `vimCommand` keyword list, fixes their
highlighting.
closes#9327
Install gperf using package manager instead of building it from source.
When building/installing gperf from source, its install step requires
`texi2pdf` which randomly goes missing on Travis:
cd doc; /usr/bin/make install
make[1]: Entering directory '/home/travis/nvim-deps/build/src/gperf/doc'
cd . && rm -f gperf.aux gperf.toc gperf.cp gperf.fn gperf.ky gperf.pg gperf.tp gperf.vr gperf.log gperf.cps
cd . && texi2pdf gperf.texi
/bin/sh: 1: texi2pdf: not found
It's nice to test the "bundled" deps on Travis, but that gets enough
exercise on Windows and macOS, which are the platforms that actually
need "bundled" gperf.
Problem: execute() does not work in completion of user command. (thinca)
Solution: Switch off redir_off and restore it. (Ozaki Kiichi, closesvim/vim#2492)
2095148277
Before this commit, if user does this:
let g:node_host_prog = '~/.nvm/versions/node/v11.3.0/bin/neovim-node-host'
the "~/" is not expanded to user's home directory.
`:help g:ruby_host_prog` suggests a path with "~/" so technically we
already claimed to support this.
closes https://github.com/neovim/node-client/issues/102
PR #9304 added support for functions in clipboard providers. As part of
the PR I meant to move two checks in the provider code out of an if
statement into separate statements and adding additional checks for
g:clipboard attributes - as it turns out the code is wrong and it does
not implement additional checks while it adds two conditions that make
very little sense
type(g:clipboard['copy']) #isnot# v:t_func
what would make sense would be something along the lines of
type(g:clipboard['copy']['+']) #isnot# v:t_func
but might not be what we want either, so I'm reverting this.
Up to now g:clipboard["copy"] only supported string values invoked as
system commands.
This commit enables the use of VimL functions instead. The function
signatures are the same as in provider/clipboard.vim. A clipboard
provider is expected to store and return a list of lines (i.e. the text)
and a register type (as seen in setreg()).
cache_enabled is ignored if "copy" is provided by a VimL function.
As of CMake 3.12, check_include_files() also link the check executable
against the libraries listed in CMAKE_REQUIRED_LIBRARIES. Therefore we
should unset the CMAKE_REQUIRED_* variables after each respective use to
avoid them unnecessarily bleeding into other checks.
The order was swapped in #4150 to prefer `xsel` but there wasn't a clear
explanation. Meanwhile, `xsel` has been neglected upstream.
Let's trying preferring `xclip` again, we've had a few reports of
problems with `xsel`.
closes#7237
ref #5853
ref #7449
Problem: Expression evaluation may repeat an error message. (Jason
Franklin)
Solution: Check for the value of did_emsg when giving an error
for the :execute command.
8ff5af9544
Problem: Expression evaluation may repeat an error message. (Jason
Franklin)
Solution: Increment did_emsg and check for the value when giving an error
for the echo command.
76a6345433
Problem: It is not easy to edit a script that was sourced.
Solution: Add a count to ":scriptnames", so that ":script 40" edits the
script with script ID 40.
07dc18ffa4
From test_timers.vim:
Found errors in Test_paused():
First run:
function RunTheTest[35]..Test_paused line 20: Expected range 0 - 100, but got 123
Second run:
function RunTheTest[35]..Test_paused line 20: Expected range 0 - 100, but got 106
previously: #9220
- Timer tests are less reliable on Travis CI macOS 10.12/10.13.
ref #6829
ref e39dade80b
ref de13113dc1
ref https://github.com/neovim/neovim/pull/9095#issuecomment-429603452
> We don't guarantee that a X ms timer is triggered during Y ms sleep
> for any X<Y, though I would expect the load to be really bad for this
> to happen with X=10ms, Y=40ms.
From test_alot.vim:
Found errors in Test_lambda_with_timer():
First run:
function RunTheTest[35]..Test_lambda_with_timer line 19: Expected True but got 0
Second run:
function RunTheTest[35]..Test_lambda_with_timer line 19: Expected True but got 0
previously: #9220
- Timer tests are less reliable on Travis CI macOS 10.12/10.13.
ref #6829
ref e39dade80b
ref de13113dc1
ref https://github.com/neovim/neovim/pull/9095#issuecomment-429603452
> We don't guarantee that a X ms timer is triggered during Y ms sleep
> for any X<Y, though I would expect the load to be really bad for this
> to happen with X=10ms, Y=40ms.
- Call test_garbagecollect_now(), as Vim does.
Problem: Segfault when pattern with \z() is very slow.
Solution: Check for NULL regprog. Add "nfa_fail" to test_override() to be
able to test this. Fix that 'searchhl' resets called_emsg.
bcf9442307closes#8788
Fixes https://github.com/neovim/neovim/issues/9270
---
Background info per egmontkob:
https://github.com/neovim/neovim/issues/9270#issuecomment-441979176
For undercurl, the newly invented escape sequence is `4:3` strictly with
a colon, as with a semicolon it means single underlined and italic.
For colored underline, the newly invented escape sequence `58:...` is
meant to follow the pattern of `38` and `48`. [ITU
T.416](https://www.itu.int/rec/T-REC-T.416-199303-I/en) § 13.1.8 clearly
specifies the colon only as the separator (and the well-known ECMA-48
§ 8.3.117 just points to this standard).
Using semicolon instead was/is a frequent misinterpretation of this
standard, and is commonly used in the wild – for 38 and 48. More and
more emulators are catching up and beginning to support colon, in
addition to semicolon. Semicolon is pretty fragile; in case an emulator
doesn't recognize a sequence (let's say doesn't recognize the new
extension of `58`), subsequent numbers are interpreted as other
attributes. E.g. if 256-color mode is chosen then the next numeric
parameter is `5` which turns on blinking.
So, luckily, the standard is the technically better solution, the
frequent practice of using semicolons is technically the worse.
Therefore the direction we should be going is clear.
I believe it's a fair requirement for anyone adopting colored underline
to support colons too, and it's a reasonable move from applications to
slightly push the world forward, force developers to catch up with the
recent changes, that is: 1) recognize and at least ignore
colon-delimited parameters even if they aren't supported, 2) recognize
and support colon wherever they support the nonstandard semicolon
instead.
Should you come across any terminal emulator that supports 58 with
semicolons but not with colons, I think the cleanest you can do is
report a bug against them and ignore the problem; they should fix it.
It's yet another common misunderstanding that the truecolor syntax is
`38`/`48`/`58` followed by `:2:rrr:ggg:bbb`. The wording of T.416 is
terrible, but if you read carefully, there's another parameter of
color-space-id preceding the three color channels. Assuming you don't
care about color-space-id, the syntax is `38`/`48`/`58` followed by
`:2::rrr:ggg:bbb` and of course the trailing `m`.
This is only for true-color, the 256-color format doesn't have such
a parameter, it's `38`/`48`/`58` followed by `:5:index` and the final
`m`.
closes#9274
ref #9028
If stdin closed then read_error_exit calls preserve_exit. Handling
SIGHUP during preserve_exit would cause a premature teardown, and
conflicts with e.g. ui_bridge_stop which waits for TUI to teardown.
Vim ignores SIGHUP in its prepare_to_exit and getout_preserve_modified
routines:
/* Ignore SIGHUP, because a dropped connection causes a read error, which
* makes Vim exit and then handling SIGHUP causes various reentrance
* problems. */
signal(SIGHUP, SIG_IGN);
vim-patch:8.0.0350: not enough test coverage for Perl
vim-patch:8.0.1135: W_WINCOL() is always the same
vim-patch:8.0.1280: Python None cannot be converted to a Vim type
vim-patch:8.0.1308: the "Reading from stdin" message may be undesired
vim-patch:8.0.1338: USE_IM_CONTROL is confusing and incomplete
vim-patch:8.0.1343: MS-Windows: does not show colored emojis
vim-patch:8.0.1350: cannot build with +eval and -multi_byte
vim-patch:8.0.1829: MS-Windows: script for vimdiff can't handle ! chars
vim-patch:8.0.1849: compiler warning for unused arguments, missing prototype
vim-patch:8.1.0001: the netrw plugin does not work
vim-patch:8.1.0006: syn_id2cterm_bg() may be undefined
vim-patch:8.1.0012: misplaced #endif
vim-patch:8.1.0021: clang warns for undefined behavior
vim-patch:8.1.0041: attribute "width" missing from python window attribute list
vim-patch:8.1.0051: MS-Windows: missing #endif
vim-patch:8.1.0063: Mac: NSStringPboardType is deprecated
vim-patch:8.1.0075: no Vim logo in README file
vim-patch:8.1.0077: header of README file is not nice
vim-patch:8.1.0079: superfluous space in messages
vim-patch:8.1.0102: cannot build without syntax highlighting
vim-patch:8.1.0104: can't build without the +eval feature
vim-patch:8.1.0109: new po makefile missing from distribution
vim-patch:8.1.0117: URL in install program still points to SourceForge
vim-patch:8.1.0122: translators don't always understand the maintainer message
vim-patch:8.1.0123: MS-Windows: colors are wrong after setting 'notgc'
vim-patch:8.1.0124: has('vcon') returns true even for non-win32 terminal
vim-patch:8.1.0127: build failure when disabling the session feature
vim-patch:8.1.0128: building with MinGW does not work out-of-the-box
vim-patch:8.1.0129: still some xterm-like terminals get a stray "p"
vim-patch:8.1.0137: CI does not run with TCL
vim-patch:8.1.0142: xterm and vt320 builtin termcap missing keypad keys
vim-patch:8.1.0147: compiler warning when building with Python 3.7
vim-patch:8.1.0148: memory leak when using :tcl expr command
vim-patch:8.1.0150: insufficient test coverage for Tcl
vim-patch:8.1.0152: cannot easily run individual tests on MS-Windows
vim-patch:8.1.0153: build with SHADOWDIR fails
vim-patch:8.1.0155: evim.man missing from the distribution
vim-patch:8.1.0157: old iTerm2 is not recognized, resulting in stray output
vim-patch:8.1.0160: no Danish manual translations
vim-patch:8.1.0162: Danish and German man pages are not installed
vim-patch:8.1.0163: insufficient testing for Tcl
vim-patch:8.1.0173: compiler warning on MS-Windows
vim-patch:8.1.0176: overlapping string argument for strcpy()
vim-patch:8.1.0178: warning for passing pointer to non-pointer argument
vim-patch:8.1.0179: redundant condition for boundary check
vim-patch:8.1.0180: static analysis errors in Lua interface
vim-patch:8.1.0183: Lua API changed, breaking the build
vim-patch:8.1.0185: running tests writes lua.vim even though it is not used
vim-patch:8.1.0190: Perl refcounts are wrong
vim-patch:8.1.0191: Perl test fails in 24 line terminal
vim-patch:8.1.0197: Windows GUI: title for search/replace is wrong
vim-patch:8.1.0201: newer Python uses "importlib" instead of "imp"
vim-patch:8.1.0202: :version always shows +packages
vim-patch:8.1.0203: building with Perl 5.28 fails on Windows
vim-patch:8.1.0207: need many menu translation files to cover regions
vim-patch:8.1.0209: stderr output from Ruby messes up display
vim-patch:8.1.0215: no error if configure --with-x cannot configure X
vim-patch:8.1.0217: compiler warning for variable set but not used
vim-patch:8.1.0222: errors are reported for "make install"
vim-patch:8.1.0232: Ruby error does not include backtrace
vim-patch:8.1.0234: incorrect reference counting in Perl interface
vim-patch:8.1.0236: Ruby build fails when ruby_intern is missing
vim-patch:8.1.0237: Ruby on Cygwin doesn't always work
vim-patch:8.1.0239: now Ruby build fails on other systems
vim-patch:8.1.0246: build failure without the +eval feature
vim-patch:8.1.0247: Python: error message for failing import is incorrect
vim-patch:8.1.0249: GTK: when screen DPI changes Vim does not handle it
vim-patch:8.1.0250: MS-Windows using VTP: windows size change incorrect
vim-patch:8.1.0254: cannot build on MS-Windows; unused macro HAVE_HANDLE_DROP
vim-patch:8.1.0260: no LGTM logo in README file
vim-patch:8.1.0287: MAX is not defined everywhere
vim-patch:8.1.0292: MS-Windows: the text "self-installing" confuses some users
vim-patch:8.1.0301: GTK: input method popup displayed on wrong screen.
vim-patch:8.1.0305: missing support for Lua 5.4 32 bits on Unix
vim-patch:8.1.0319: bzero() function prototype doesn't work for Android
vim-patch:8.1.0332: get Gdk-Critical error on first balloon show
vim-patch:8.1.0346: building with Aap is outdated and unused
vim-patch:8.1.0348: on Travis the slowest build is run last
vim-patch:8.1.0357: instructions for tests are outdated
vim-patch:8.1.0368: GTK code has too many #ifdefs and GTK 2.10 building fails
vim-patch:8.1.0379: build dependencies are incomplete
vim-patch:8.1.0380: "make proto" doesn't work well
vim-patch:8.1.0383: missing source file rename
vim-patch:8.1.0385: Coveralls badge doesn't update
vim-patch:8.1.0386: cannot test with non-default option value
vim-patch:8.1.0388: Coverity complains about possible NULL pointer use
vim-patch:8.1.0390: scrollbars are not tested
vim-patch:8.1.0391: building in a shadow directory fails
vim-patch:8.1.0403: header file missing from distribution
vim-patch:8.1.0405: too many #ifdefs for GTK
vim-patch:8.1.0408: MSVC: cannot use the "x64" native compiler option
vim-patch:8.1.0411: renamed file missing from distribution
vim-patch:8.1.0412: cannot build with GTK 2.4
vim-patch:8.1.0413: test output is duplicated or missing
vim-patch:8.1.0415: not actually using 16 colors with vtp
vim-patch:8.1.0418: MS-Windows: cannot separate Lua include and library dirs
vim-patch:8.1.0419: Cygwin: running cproto fails with -O2
vim-patch:8.1.0420: generating vim.lib when using ActivePerl 5.20.3 or later
vim-patch:8.1.0421: MS-Windows: Ruby path is wrong for Ruby 1.9 and later
vim-patch:8.1.0422: cannot create map file with MinGW
vim-patch:8.1.0427: MS-Windows GUI: using invalid encoded file name
vim-patch:8.1.0441: build failure without command line history
vim-patch:8.1.0467: cannot build with Mac OS X 10.5
vim-patch:8.1.0472: dosinst command has a few flaws
vim-patch:8.1.0474: directory where if_perl.c is written is inconsistent
vim-patch:8.1.0477: tiny build fails
vim-patch:8.1.0478: cannot build with perl using MinGW
vim-patch:8.1.0480: MinGW build file uses different -I flags than MVC
vim-patch:8.1.0482: MinGW "make clean" deletes all .exe files
vim-patch:8.1.0483: MinGW does not build tee.exe
vim-patch:8.1.0490: MS-Windows: doesn't handle missing glibwinpthread-1.dll
vim-patch:8.1.0492: "Edit with existing Vim" list can get long
vim-patch:8.1.0500: cleaning up in src/tee may not always work
vim-patch:8.1.0521: cannot build with +eval but without +quickfix
vim-patch:8.1.0534: MS-Windows installer uses different $HOME than Vim
vim-patch:8.1.0541: help message in dosinst.c is outdated
vim-patch:8.1.0485: term_start() does not check if directory is accessible #9204
In file included from ../src/nvim/eval/encode.c:974:
../src/nvim/eval/typval_encode.c.h:390:40: error: equality comparison with extraneous parentheses [-Werror,-Wparentheses-equality]
do { if ((tv->vval.v_special == kSpecialVarTrue)) { msgpack_pack_true(packer); } else { msgpack_pack_false(packer); } } while (0);
~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~
* Remove FUNC_ATTR_NONNULL_ALL from function without pointer arguments.
Otherwise the ASAN build would complain:
error: 'nonnull' attribute applied to function with no pointer arguments [-Werror,-Wignored-attributes]
static void do_autocmd_focusgained(_Bool gained) __attribute__((nonnull));
Travis is phasing out its support for containers, so we remove the `sudo:
false`, which will be a no-op soon.
Reference: https://blog.travis-ci.com/2018-11-19-required-linux-infrastructure-migration
Changes for Linux:
- Xenial comes with libtool installed already. It only provides "libtoolize",
though. For "libtool" we need to install libtool-bin.
Namespaces is a lightweight concept that should be used to group
objects for purposes of bulk operations and introspection. This is
initially used for highlights and virtual text in buffers, and is
planned to also be used for extended marks. There is no plan use them
for privileges or isolation, neither to introduce nanespace-level
options.
Problem: Characters deleted on completion. (Adrià Farrés)
Solution: Also check the last item for the ORIGINAL_TEXT flag. (Christian
Brabandt, closesvim/vim#1645)
e87edf3b85
By historical accident, Nvim defaults to background=light. So on a dark
background, `:colorscheme default` looks completely wrong.
The "smart" logic that Vim uses is confusing for anyone who uses Vim on
multiple platforms, so rather than mimic that, pick the (hopefully) most
common default.
- Since Neovim is dark-powered, we assume most users have dark backgrounds.
- Most of the GUIs tend to have a dark background by default.
ref #6289
Problem: No error when WaitFor() gets an invalid wrong expression.
Solution: Do not ignore errors in evaluationg the expression. Fix places
where the expression was wrong.
c20e0d5207
`:help $TERM` recommends TERM=nsterm for Terminal.app but we did not
actually support it.
NB: We don't include a builtin term for Terminal.app, presumably because
nsterm is commonly available on most systems (`infocmp nsterm`).
Problem: The getftype() test may fail for char devices if the file
disappeared in between the listing and the getftype() call.
Solution: Ignore empty result. (Ozaki Kiichi, closesvim/vim#3360)
3b3a506f57
Problem: No Danish translations.
Solution: Add Danish message translations. (closesvim/vim#3073) Move list of
languages to a common makefile.
a4a2934e59
- TUI: _never_ rely on BCE for implicit clearing, only explicit commands.
- TUI: use unibi_erase_chars when possible.
- TUI: use end-exclusive ranges for invalid and cleared areas
- screen: scrolling leaves scrolled in aree undefined. This is a
conservative change, a client assuming the old semantics will still
behave correctly.
- screen: factor out vsep handling from line drawing. This is needed
anyway for the multigrid refactor.
- screen: simplifications of win_do_lines
- window_split_tab_spec.lua: Put cursor at bottom of :terminal buffer so
that it follows output.
- inccommand_spec.lua: Increase timeout to allow 2nd retry.
- Timer tests are less reliable on Travis CI macOS 10.12/10.13.
ref #6829
ref e39dade80b
ref de13113dc1
ref https://github.com/neovim/neovim/pull/9095#issuecomment-429603452
> We don't guarantee that a X ms timer is triggered during Y ms sleep
> for any X<Y, though I would expect the load to be really bad for this
> to happen with X=10ms, Y=40ms.
* os/fs.c: add os_isdir_executable()
* eval.c: fix hang on job start caused by non-executable cwd option
* channel.c: assert cwd is an executable directory
* test: jobstart() produces error when using non-executable cwd
Problem: Filter test fails when $LANG is C.UTF-8.
Solution: Set 'helplang' to "en" for any C language. (Christian Brabandt,
closesvim/vim#3577)
dcd71cbaed
Test is unreliable on macOS 10.13. The lower-bound isn't central to the
purpose of the test, so just relax it.
ref https://github.com/neovim/neovim/pull/9095#issuecomment-429603452
> We don't guarantee that a X ms timer is triggered during Y ms sleep
> for any X<Y, though I would expect the load to be really bad for this
> to happen with X=10ms, Y=40ms.
related: #6829
Generate helptags before :help.
This only failed in the TSAN build because other builds run the Lua
tests first, which generate the helptags.
closes#8824
Give variables a default value to pass strict mode.
$ErrorActionPreference defines the default behavior
if a powershell command fails.
If it's set to 'stop', then it aborts the script
on the first unresolved error.
This behavior extends to native programs like cmake
but do not depend on it.
https://github.com/PowerShell/PowerShell/issues/3996
Remove libuv-overlapped.patch since UV_OVERLAPPED_PIPE was included in
libuv v1.21.0:
62a0f763a7
Notable changes since v1.12:
- 1.16.0
- uv_os_getppid(): get parent PID
- "win,tty: improve SIGWINCH support" (v1.15.0)
- 1.18.0
- uv_os_getpid()
- 1.19.0
- Windows: uv_kill() pid 0 now means "current process group", like unix.
890eedaf59
- 1.20.0
- unix,spawn: respect user stdio flags for new pipe
c409b3fcff
Before this change, some tests fail if the user has a local
shada/viminfo in the default $HOME location. For example:
TEST_FILE=test_marks.res make oldtest
Ostensibly this change should not be necessary, because of this
statement in runtest.vim:
set viminfo+=nviminfo
Problem: Cannot see what digraph is used to insert a character.
Solution: Show the digraph with the "ga" command. (Christian Brabandt)
5f73ef8d20close#8190
Update runtime files.
fc65cabb15
---
vim-patch:8.0.1279: initializing menus can be slow
Problem: Initializing menus can be slow, especially when there are many
keymaps, color schemes, etc.
Solution: Do the globbing for runtime files lazlily. (Ken Takata)
Update runtime files.
b5b7562475
---
NA patch (Nvim does not ship with spell files):
vim-patch:3ad8772ef02e
Include Serbian spell input files
3ad8772ef0
close#8474
vim-patch:8.0.0264: memory error reported by ubsan
vim-patch:8.0.0340: not checking return valud of dict_add()
vim-patch:8.0.0346: Vim relies on limits.h to be included indirectly
vim-patch:8.0.0361: GUI initialisation is not sufficiently tested
vim-patch:8.0.0365: might free a dict item that wasn't allocated
vim-patch:8.0.0367: types in include files may be inconsistent
vim-patch:8.0.0369: a few options are not defined, depending on features
vim-patch:8.0.0372: more options are not always defined
vim-patch:8.0.0386: tiny build has a problem with generating the options test
vim-patch:8.0.0392: GUI test fails with Athena and Motif
vim-patch:8.0.0395: testing the + register fails with Motif
vim-patch:8.0.0422: Python test fails with Python 3.6
vim-patch:8.0.0425: build errors when building without folding
vim-patch:8.0.0432: "make shadow" creates an invalid link
vim-patch:8.0.0476: missing change to main.c
vim-patch:8.0.0526: Coverity complains about possible negative value
vim-patch:8.0.0540: building unit tests fails
vim-patch:8.0.0556: getting window position fails if GUI and term possible
vim-patch:8.0.0563: crash when getting the window position in tmux
vim-patch:8.0.0567: call for requesting color and ambiwidth is too early
vim-patch:8.0.0569: bracketed paste is still enabled in a shell command
vim-patch:8.0.0587: configure check for return value of tgetent skipped
vim-patch:8.0.0588: job_stop() often assumes the channel will be closed
vim-patch:8.0.0610: the screen is redrawn when default 'background' is detected
vim-patch:8.0.0611: the screen is not redrawn after sending t_u7
vim-patch:8.0.0620: checking for HAVE_GTK_MULTIHEAD is not needed
vim-patch:8.0.0626: in the GUI the cursor may flicker
vim-patch:8.0.0628: cursor disappears after silent mapping
vim-patch:8.0.0631: can't build with Perl 5.26
vim-patch:8.0.0637: crash when using some version of GTK 3
vim-patch:8.0.0639: the cursor position is set when editing a new commit message
vim-patch:8.0.0649: when opening a help file the filetype is set several times
vim-patch:8.0.0653: the default highlight for QuickFixLine is not good
vim-patch:8.0.0690: compiler warning on non-Unix system
vim-patch:8.0.0691: compiler warning without the linebreak feature
vim-patch:8.0.0696: .inc files missing in git
vim-patch:8.0.0719: build failure without +terminal feature
vim-patch:8.0.0729: the help for the terminal configure option is wrong
vim-patch:8.0.0731: cannot build the terminal feature on MS-Windows
vim-patch:8.0.0741: cannot build with HPUX
vim-patch:8.0.0750: OpenPTY missing in non-GUI build
vim-patch:8.0.0751: OpenPTY missing with some combination of features
vim-patch:8.0.0752: build fails on MS-Windows
vim-patch:8.0.0763: libvterm can be improved
vim-patch:8.0.0770: compiler warning for missing field initializer
vim-patch:8.0.0772: other stdbool.h dependencies in libvterm
vim-patch:8.0.0773: mixing 32 and 64 bit libvterm builds fails
vim-patch:8.0.0774: build failure without the multi-byte feature
vim-patch:8.0.0790: MSVC compiler warning for strncpy in libvterm
vim-patch:8.0.0795: terminal feature does not build with older MSVC
vim-patch:8.0.0808: cannot build with terminal feature and DEBUG defined
vim-patch:8.0.0809: MS-Windows: tests hang
vim-patch:8.0.0811: MS-Windows: test_expand_dllpath fails
vim-patch:8.0.0822: Test_with_partial_callback is a tiny bit flaky
vim-patch:8.0.0834: can't build without the client-server feature
vim-patch:8.0.0865: cannot build with channel but without terminal feature
vim-patch:8.0.0866: Solaris also doesn't have MIN and MAX
vim-patch:8.0.0875: crash with weird command sequence
vim-patch:8.0.0880: Travis uses an old Ubuntu version
vim-patch:8.0.0901: asan suppress file missing from distribution
vim-patch:8.0.0919: cursor color isn't set on startup
vim-patch:8.0.0921: terminal window cursor shape not supported in the GUI
vim-patch:8.0.0946: using PATH_MAX does not work well on some systems
vim-patch:8.0.0954: /proc/self/exe might be a relative path
vim-patch:8.0.0966: build failure without terminal feature
vim-patch:8.0.0967: using a terminal may cause the cursor to blink
vim-patch:8.0.0972: compiler warnings for unused variables
vim-patch:8.0.0980: Coverity warning for failing to open /dev/null
vim-patch:8.0.0986: terminal feature always requires multi-byte feature
vim-patch:8.0.0988: warning from Covscan about using NULL pointer
vim-patch:8.0.0993: sometimes an xterm sends an extra CTRL-X
vim-patch:8.0.0996: Mac: t_RS is echoed on the screne in Terminal.app
vim-patch:8.0.0997: Libvterm and Terminal.app not regognized from termresponse.
vim-patch:8.0.1008: slow updating of terminal window in Motif
vim-patch:8.0.1010: build failure without termresponse feature
vim-patch:8.0.1014: old compiler doesn't know uint32_t
vim-patch:8.0.1016: gnome terminal echoes t_RC
vim-patch:8.0.1021: older Gnome terminal still echoes t_RC
vim-patch:8.0.1026: GTK on-the-spot input has problems
vim-patch:8.0.1027: more terminals can't handle requesting cursor mode
vim-patch:8.0.1030: MS-Windows: wrong size computation in is_cygpty()
vim-patch:8.0.1033: detecting background color does not work in screen
vim-patch:8.0.1042: without the syntax feature highlighting doesn't work
vim-patch:8.0.1047: buffer overflow in Ruby
vim-patch:8.0.1059: older Gnome terminal returns smaller version number
vim-patch:8.0.1062: Coverity warnings in libvterm
vim-patch:8.0.1065: not all macro examples are included
vim-patch:8.0.1066: some terminals can't handle requesting cursor mode
vim-patch:8.0.1068: vandyke SecureCRT terminal can't handle cursor mode request
vim-patch:8.0.1069: still get CTRL-X sometimes
vim-patch:8.0.1071: putty-color and cygwin-color are not recognized
vim-patch:8.0.1094: using ssh from Terminal.app runs into xterm incompatibility
vim-patch:8.0.1110: FORTIFY_SOURCE from Perl causes problems
vim-patch:8.0.1111: syntax error in configure when using Perl
vim-patch:8.0.1118: FEAT_WINDOWS adds a lot of #ifdefs
vim-patch:8.0.1136: W_WIDTH() is always the same
vim-patch:8.0.1152: encoding of error message wrong in Cygwin terminal
vim-patch:8.0.1155: Ruby command triggers a warning
vim-patch:8.0.1157: compiler warning on MS-Windows
vim-patch:8.0.1159: typo in #ifdef
vim-patch:8.0.1166: :terminal doesn't work on Mac High Sierra
vim-patch:8.0.1175: build failure without +termresponse
vim-patch:8.0.1192: MS-Windows: terminal feature not enabled by default
vim-patch:8.0.1195: can't build on MS-Windows
vim-patch:8.0.1198: older compilers don't know uint8_t
vim-patch:8.0.1211: cannot reorder tab pages with drag & drop
vim-patch:8.0.1247: not easy to find Debian build info
vim-patch:8.0.1248: stray + in README file
vim-patch:8.0.1255: duplicate badge README file
vim-patch:8.0.1256: typo in configure variable vim_cv_tgent
vim-patch:8.0.1258: 'ttymouse' is set to "sgr" even though it's not supported
vim-patch:8.0.1269: effect of autocommands on marks is not tested
vim-patch:8.0.1272: warnings for unused variables in tiny build
vim-patch:8.0.1273: old test file remaining
vim-patch:8.0.1303: 'ttymouse' is not set to "sgr" for Terminal.app and Iterm2
vim-patch:8.0.1306: ASAN error stack trace is not useful
vim-patch:8.0.1310: cproto generates errors because of missing type
vim-patch:8.0.1322: textformat test isn't run
vim-patch:8.0.1324: some xterm sends different mouse move codes
vim-patch:8.0.1325: more tests are not run
vim-patch:8.0.1327: new proto file missing from distribution
vim-patch:8.0.1337: typo in #ifdef
vim-patch:8.0.1369: MS-Windows: drawing underline slow, mFallbackDC not updated
vim-patch:8.0.1380: using "vim -r swapfile" the hit-enter prompt is misplaced.
vim-patch:8.0.1394: cannot intercept a yank command
vim-patch:8.0.1415: warning for unused function without timers feature
vim-patch:8.0.1424: the timer_pause test is flaky on Travis
vim-patch:8.0.1437: pkg-config doesn't work with cross compiling
vim-patch:8.0.1456: timer test on travis Mac is still flaky
vim-patch:8.0.1467: libvterm doesn't handle illegal byte sequence correctly
vim-patch:8.0.1478: unnecessary condition
vim-patch:8.0.1504: Win32: the screen may be cleared on startup
vim-patch:8.0.1506: new version of HP NonStop (Tandem) doesn't like a header
vim-patch:8.0.1532: compiler warnings without termguicolors feature
vim-patch:8.0.1556: may not parse the t_RS response correctly
vim-patch:8.0.1564: too many #ifdefs
vim-patch:8.0.1565: can't build Mac version without GUI
vim-patch:8.0.1566: too many #ifdefs
vim-patch:8.0.1567: cannot build Win32 GUI without IME
vim-patch:8.0.1571: can't build without GUI
vim-patch:8.0.1573: getwinpos(1) may cause response to be handled as command
vim-patch:8.0.1580: FEAT_CURSORBIND and FEAT_SCROLLBIND are unused
vim-patch:8.0.1581: cannot build Win32 GUI without +eval
vim-patch:8.0.1583: using C99 comment
vim-patch:8.0.1584: using C99 in Mac file gives compiler warning messages
vim-patch:8.0.1585: enabling beval_term feature in Win32 GUI
vim-patch:8.0.1586: imactivatefunc does not work on non-GUI Mac
vim-patch:8.0.1597: autocommand events are not sorted
vim-patch:8.0.1603: cannot build with +terminal but without +menu
vim-patch:8.0.1606: singular/plural variants not translated
vim-patch:8.0.1608: Win32: directx not enabled by default
vim-patch:8.0.1610: cannot build without GUI
vim-patch:8.0.1613: warning for unused variable in tiny build
vim-patch:8.0.1614: "make tags" doesn't include libvterm
vim-patch:8.0.1618: color Grey50 is missing in the compiled-in table
vim-patch:8.0.1625: test_quotestar is flaky when run in GTK GUI
vim-patch:8.0.1627: compiler warning for visibility attribute not supported
vim-patch:8.0.1635: undefining _POSIX_THREADS causes problems with Python 3
vim-patch:8.0.1639: libvterm code lags behind master
vim-patch:8.0.1646: MS-Windows: executable contains unreferenced functions
vim-patch:8.0.1650: too many #ifdefs
vim-patch:8.0.1659: scroll events not recognized for some xterm emulators
vim-patch:8.0.1663: cannot build without multi-byte feature
vim-patch:8.0.1674: libvterm can't handle an OSC string split
vim-patch:8.0.1675: unused macro argument in libvterm
vim-patch:8.0.1679: compiler warning for printf format
vim-patch:8.0.1683: Python upgrade breaks Vim when defining PYTHON_HOME
vim-patch:8.0.1686: Python does not work when configuring with specific dir
vim-patch:8.0.1692: Python may not work when using statically linked library
vim-patch:8.0.1696: coverage statistics don't work
vim-patch:8.0.1698: coverage statistics don't work on coveralls
vim-patch:8.0.1699: leftover stuff for Python 1.4
vim-patch:8.0.1700: coverage statistics still don't work on coveralls
vim-patch:8.0.1701: can disable COLOR_EMOJI with MSVC but not MinGW
vim-patch:8.0.1710: building with Ruby fails
vim-patch:8.0.1717: C89 check causes too much trouble
vim-patch:8.0.1719: cannot specify which Python executable configure should use
vim-patch:8.0.1730: no configure check for the used C99 features
vim-patch:8.0.1736: check for C99 features is incomplete
vim-patch:8.0.1737: fchown() used when it is not supported
vim-patch:8.0.1739: MS-Windows with msys2 cannot build Ruby statically
vim-patch:8.0.1741: MS-Windows with msys2 cannot build Ruby statically
vim-patch:8.0.1749: VMS: 100% CPU use, redefining mch_open() fails
vim-patch:8.0.1751: #ifdef causes bad highlighting
vim-patch:8.0.1755: MS-Windows: high unicode char received as two utf-16 words
vim-patch:8.0.1757: unnecessary changes in libvterm
vim-patch:8.0.1760: wrong number of arguments to vms_read()
vim-patch:8.0.1764: lgtm considers tutor.es to be EcmaScript
vim-patch:8.0.1785: missing symbol in Win32 small build
vim-patch:8.0.1788: tool to check a color scheme is not installed
vim-patch:8.0.1791: using uint8_t does not work everywhere
vim-patch:8.0.1813: Windows installer doesn't install terminal debugger
vim-patch:8.0.1825: might use NULL pointer when out of memory
vim-patch:8.0.1840: getwinpos() is not tested
vim-patch:8.0.0216: decoding js style json may fail
vim-patch:8.0.1391: encoding empty string to JSON sometimes gives "null"
vim-patch:8.0.1602: crash in parsing JSON
vim-patch:8.1.0343: #9041
Problem: Python2 and python3 detection not tested. (Matej Cepl)
Solution: Add test for detecting python2 and python3. Also detect a script
using "js" as javascript.
4bc0bed536
man#init_pager() guesses the ref by the heading, which is usually
uppercase, so we don't know the correct casing. But lowercase is more
common, so use that for the buffer name instead of uppercase.
ref #9156
Before this commit, man#init_pager() always tries to scrape the manpage
name and set the buffer name. That's much less important than avoiding
duplicate buffers and E95. And it doesn't seem to be necessary, usually.
Steps to reproduce:
$ export MANPAGER="nvim -c 'set ft=man' -"
$ man sleep
:Man sleep
Error detected while processing function man#init_pager:
line 15:
E95: Buffer with this name already exists
:ls!
1 h- "man://SLEEP(1)" line 4
2 %a- "man://sleep(1)" line 1
Despite #9095, `brew upgrade python` broke again, somehow.
We should not bother attempting to force a python version. Instead use
whatever python Travis provides on the macOS image.
Avoid clearing the screen in most situations. NOT_VALID should be
equivalent to CLEAR unless some external force messed up the terminal,
for these situations <c-l> and :mode will still clear the screen.
Also eliminate some obsolete code in screen.c, that dealt with that in
vim drawing window 1 can mess up window 2, but this never happens in
nvim.
But what about slow terminals? There is two common meanings in which
a terminal is said to be "slow":
Most commonly (and in the sense of vim:s nottyfast) it means low
bandwidth for sending bytes from nvim to the terminal. If the screen is
very similar before and after the update_screen(CLEAR) this change
should reduce bandwidth. If the screen is quite different, but there is
no new regions of contiguous whitespace, clearing doesn't reduce
bandwidth significantly. If the new screen contains a lot of whitespace,
it will depend of if vsplits are used or not: as long as there is no
vsplits, ce is used to cheaply clear the rest of the line, so
full-screen clear is not needed to reduce bandwith. However a left
vsplit currently needs to be padded with whitespace all the way to the
separator. It is possible ec (clear N chars) can be used to reduce
bandwidth here if this is a problem. (All of this assumes that one
doesn't set Normal guibg=... on a non-BCE terminal, if you do you are
doomed regardless of this change).
Slow can also mean that drawing pixels on the screen is slow. E-ink
screens is a recent example. Avoiding clearing and redrawing the
unchanged part of the screen will always improve performance in these
cases.
For each package manager,
1. Check if they are executable before starting a job.
2. Check if the job opts are set before checking the job status
TODO: Validate the job id.
Prioritize npm because it's faster than yarn.
Reindent the code to 2-space indent.
PR #9087 changed the error string by removing 'Running', breaking the
MSBuild hack detecting failure for functional tests. If stdout or stderr
has a line with 'functional tests failed with error', fail the build.
clang scan-build noticed that find_command() may bitmask `eap->flags`.
cmd_can_preview() only uses `ea.cmdidx`, but let's fix the warning...
Found by clang scan-build 5.0
mandoc may not handle quoted MANPAGER arguments correctly. E.g. with
export MANPAGER='nvim -u NORC -c "set ft=man"'
mandoc treats `"set` and `ft=man"'` as separate tokens.
To workaround that, provide :Man! so that MANPAGER can avoid quoting.
closes#9120
fix#9050
Q: (from From #7891)
> It turns out there's a difference between executing
> /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/clang
> and /usr/bin/clang. CMake was picking up the former and with that it left off
> a key include path. Why?
A: /usr/bin/clang is a shim that uses libxcselect/libxcrun under the hood.
$ otool -L /usr/bin/clang
/usr/bin/clang:
/usr/lib/libxcselect.dylib (compatibility version 1.0.0, current version 1.0.0)
/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version 1252.200.5)
From https://macops.ca/developer-binaries-on-os-x-xcode-select-and-xcrun/ :
> this shim binary loads functions in libxcselect.dylib that can locate the
> path to the real binary, depending on how the system has been configured. One
> part of this process is to check whether this path contains
> usr/lib/libxcrun.dylib, and the xcrun tool, in which case it will invoke
> xcrun to run the binary.
> ...
> (If all of this isn’t yet enough indirection for you, /usr/bin/xcrun itself
> is a shim, and so libxcselect.dylib contains code to detect whether the
> executed xcrun is a shim. Look for the __xcrun_shim segment in the __DATA
> section output by the command: pagestuff /usr/bin/xcrun -a.)
Sometimes 008_autocommands_spec fails like this:
[ RUN ] autocommands that delete and unload buffers: BufUnload, VimLeave: -- Output to stderr:
CMake Error at /home/travis/build/neovim/neovim/cmake/RunTests.cmake:53 (message):
functional tests failed with error: 1
The final :quit + wait() is a race. Use command() instead, which is
synchronous. Use command('silent! ...') everywhere else too, because
it's clearer instead of feeding input and clearing the expected errors
with CTRL-L.
Problem: Using try/catch in timer does not prevent it from being stopped.
Solution: Reset the exception context and use did_emsg instead of
called_emsg.
e723c42836
Problem: Can't use input() in a timer callback. (Cosmin Popescu)
Solution: Reset vgetc_busy and set timer_busy. (Ozaki Kiichi, closesvim/vim#1790,
closesvim/vim#1129)
1e8e14552e
Problem: Cursorline not removed when using 'cursorbind'. (Justin Keyes)
Solution: Store the last cursor line per window. (closesvim/vim#3488)
4a5abbd613
In the 20180331 release, the format was slightly changed:
> 20180331
> + improve terminfo write/read by modifying the fourth item of the
> extended header to denote the number of valid strings in the extended
> string table (prompted by a comment in unibilium's sources).
Since the number of valid string capabilities is not necessarily the
same as extstrslen, it's not possible to sanity check the total number
of items up front anymore.
After bumping Travis macOS to 10.13, it now hangs at:
+ check_core_dumps --delete quiet
+ local del=
+ test --delete = --delete
+ del=1
+ shift
+ local app=quiet
+ test osx = osx
++ find /cores/ -type f -print
+ local 'cores=/cores//core.554
/cores//core.641
/cores//core.801'
+ test -z '/cores//core.554
/cores//core.641
/cores//core.801'
+ local core
+ for core in '$cores'
+ test 1 = 1
+ print_core quiet /cores//core.554
+ local app=quiet
+ local core=/cores//core.554
+ test quiet = quiet
+ echo 'Found core /cores//core.554'
Found core /cores//core.554
+ return 0
+ rm /cores//core.554
override r-------- root/admin for /cores//core.554?
The cores are always present on the Travis macOS 10.13 image! Hilarious.
Problem: User doesn't notice file does not exist when swap file does.
Solution: Add a note that the file cannot be found. Make the "still
running" notice stand out.
d6105cb408
homebrew or Travis changed something, now `pip3` isn't in $PATH.
`ls /usr/local/opt/python/libexec/bin` confirmed this, no matter what
brew reinstall/relink/upgrade are used.
Bumping the macOS image to 10.12 or 10.13 makes the problem go away.
==> Processing gcc49 formula rename to gcc@4.9
==> Unlinking gcc49
==> Moving gcc49 versions to /usr/local/Cellar/gcc@4.9
==> Relinking gcc@4.9
Warning: gcc@4.9 is outdated!
To avoid broken installations, as soon as possible please run:
brew upgrade
Or, if you're OK with a less reliable fix:
brew upgrade gcc@4.9
python info:
Python 2.7.12
Python 2.7.12
ci/before_install.sh: line 18: python3: command not found
pip 8.1.2 from /usr/local/lib/python2.7/site-packages (python 2.7)
pip 8.1.2 from /usr/local/lib/python2.7/site-packages (python 2.7)
ci/before_install.sh: line 21: pip3: command not found
pyenv versions:
* system (set by /Users/travis/.pyenv/version)
Upgrade Python 3.
To restore the stashed changes to /usr/local/Homebrew run:
'cd /usr/local/Homebrew && git stash pop'
==> Caveats
Python has been installed as
/usr/local/bin/python3
Unversioned symlinks `python`, `python-config`, `pip` etc. pointing to
`python3`, `python3-config`, `pip3` etc., respectively, have been installed into
/usr/local/opt/python/libexec/bin
If you need Homebrew's Python 2.7 run
brew install python@2
Pip, setuptools, and wheel have been installed. To update them run
pip3 install --upgrade pip setuptools wheel
You can install Python packages with
pip3 install <package>
They will install into the site-package directory
/usr/local/lib/python3.7/site-packages
See: https://docs.brew.sh/Homebrew-and-Python
==> Summary
º /usr/local/Cellar/python/3.7.0: 8,864 files, 153.8MB, built in 6 minutes 32 seconds
...
Upgrade Python 3 pip.
ci/before_install.sh: line 30: pip3: command not found
travis_time🔚0d23f522:start=1538818824750644000,finish=1538819451424021000,duration=626673377000
The command "ci/before_install.sh" failed and exited with 127 during .
Your build has been stopped.
/Users/travis/.travis/job_stages: line 373: shell_session_update: command not found
==> Processing gcc49 formula rename to gcc@4.9
==> Unlinking gcc49
==> Moving gcc49 versions to /usr/local/Cellar/gcc@4.9
==> Relinking gcc@4.9
Warning: gcc@4.9 is outdated!
To avoid broken installations, as soon as possible please run:
brew upgrade
Or, if you're OK with a less reliable fix:
brew upgrade gcc@4.9
python info:
Python 2.7.12
Python 2.7.12
ci/before_install.sh: line 18: python3: command not found
pip 8.1.2 from /usr/local/lib/python2.7/site-packages (python 2.7)
pip 8.1.2 from /usr/local/lib/python2.7/site-packages (python 2.7)
ci/before_install.sh: line 21: pip3: command not found
pyenv versions:
* system (set by /Users/travis/.pyenv/version)
Upgrade Python 3.
To restore the stashed changes to /usr/local/Homebrew run:
'cd /usr/local/Homebrew && git stash pop'
==> Caveats
Python has been installed as
/usr/local/bin/python3
Unversioned symlinks `python`, `python-config`, `pip` etc. pointing to
`python3`, `python3-config`, `pip3` etc., respectively, have been installed into
/usr/local/opt/python/libexec/bin
If you need Homebrew's Python 2.7 run
brew install python@2
Pip, setuptools, and wheel have been installed. To update them run
pip3 install --upgrade pip setuptools wheel
You can install Python packages with
pip3 install <package>
They will install into the site-package directory
/usr/local/lib/python3.7/site-packages
See: https://docs.brew.sh/Homebrew-and-Python
==> Summary
º /usr/local/Cellar/python/3.7.0: 8,864 files, 153.8MB, built in 6 minutes 32 seconds
...
Upgrade Python 3 pip.
ci/before_install.sh: line 30: pip3: command not found
travis_time🔚0d23f522:start=1538818824750644000,finish=1538819451424021000,duration=626673377000
The command "ci/before_install.sh" failed and exited with 127 during .
Your build has been stopped.
/Users/travis/.travis/job_stages: line 373: shell_session_update: command not found
Problem: Undo in the options window makes it empty.
Solution: Set 'undolevels' while filling the buffer. (Yasuhiro Matthew,
closesvim/vim#2645)
9c474b2773
Problem: When making backups is disabled and conversion with iconv fails
the written file is truncated. (Luo Chen)
Solution: First try converting the file and write the file only when it did
not fail. (partly by Christian Brabandt)
e6bf655bc4
Problem: There is no easy way to get the global directory, esp. if some
windows have a local directory.
Solution: Make getcwd(-1) return the global directory. (Andy Massimino,
closesvim/vim#2606)
5459129af2
- Much of the mouse chatter in gui.txt and term.txt is redundant; other
parts are outdated or irreleveant to Nvim.
- Fix default value of 'mousemodel' for Windows
design-flexible and design-multi-platform sections are "obvious" by now
(and covered in much of our other resources) and I've never seen it
referenced in any discussion.
Problem: One character cmdline abbreviation not triggered after '<,'>.
Solution: Skip over the special range. (Christian Brabandt, closesvim/vim#2320)
5e3423d192
Problem: MS-Windows: Filter command with pipe character fails. (Johannes
Riecken)
Solution: Find the pipe character outside of quotes. (Yasuhiro Matsumoto,
closesvim/vim#1743, closesvim/vim#3523)
0664089ecc
Problem: Deleting in a block selection causes problems.
Solution: Check the length of the line before adding bd.textcol and
bd.textlen. (Christian Brabandt, closesvim/vim#2825)
35e802e713
Windows has "Read and execute" permission via ACL
but nvim and libuv do not support ACL.
Windows does not support the executable bit in chmod-style permissions
but it is safe to assume that if the file exists and is readable,
then it is most likely executable.
This means that win.ini and shell32.dll are "executable"
because they exist, are readable, and are in PATH.
PATHEXT does not affect the executable permission of a file;
it exists to run files on the shell while omitting the file extension.
Assume that PATHEXT is intended for cmd.exe only
because powershell can execute powershell files (ie. *.ps1)
without changing PATHEXT or related cmd.exe environment variable.
In the future, nvim should check the outputs of 'assoc' and 'ftype',
cmd.exe internal commands, or check the registry.
Powershell can be used for ACL if C++/C# API is too difficult to use.
The old code could lead to a memory error in the following situation:
0. The previous cursor position was row 50 since before, on a grid
larger than 50 rows.
1. grid_resize changes the grid height to 40, and invalidly assumes the
resize moved the physical cursor to row 0
2. Some event used a operation that could move the cursor (such as clear), and
then reset the cursor to the "true" position row 50 (pointless after #8221, but
I forgot to remove it)
3. raw_line/cheap_to_print is invoked, and tries to inspect the grid at
row 50 (memory error)
4. grid_cursor_goto would have been called at this point, and set a
valid cursor position 0-39.
Fixes handing of "-" on Python 3:
Traceback (most recent call last):
File "…/Vcs/neovim/src/clint.py", line 3625, in <module>
main()
File "…/Vcs/neovim/src/clint.py", line 3618, in main
ProcessFile(filename, _cpplint_state.verbose_level)
File "…/Vcs/neovim/src/clint.py", line 3464, in ProcessFile
'replace').read().split('\n')
File "/usr/lib/python3.7/codecs.py", line 701, in read
return self.reader.read(size)
File "/usr/lib/python3.7/codecs.py", line 500, in read
data = self.bytebuffer + newdata
TypeError: can't concat str to bytes
Problem: Cannot use :unlet for an environment variable.
Solution: Make it work. Use unsetenv() if available.
(Yasuhiro Matsumoto, closesvim/vim#2855)
137374fd65
I have `merge.ff = no` in my Git config to not use fast-forward merges
by default, but when updating the Vim sources it should not cause a
merge commit.
[ci skip]
It fails with `nvim -u NONE -c 'set modified' -c 'confirm q'`.
Introduced in 3dffc842f (vim-patch:8.0.0983), but the Vim patch [1] does not
have this assertion.
NULL gets handled in `dialog_msg` [2].
1: 3f9a1ff141
2: c6d36b97ba/src/nvim/ex_docmd.c (L9704-L9705)
Problem: Nasty autocommand causes using freed memory. (Dominique Pelle)
Solution: Do not force executing autocommands if the value of 'syntax' or
'filetype' did not change.
c3ffc9b8d3
Problem: Weird autocmd may cause arglist to be changed recursively.
Solution: Prevent recursively changing the argument list. (Christian
Brabandt, closesvim/vim#2472)
9e33efd152
tmpdir_get() may be an absolute path, but we invoke glob() with
a relative `initial_path`.
That can lead to this error:
[ ERROR ] test/functional/helpers.lua @ 752: after_each
test/helpers.lua:95: cannot open ./Xtest-tmpdir/nvim8jKCjR: No such file or directory
stack traceback:
test/helpers.lua:95: in function 'glob'
test/helpers.lua:273: in function 'check_cores'
test/functional/helpers.lua:757: in function <test/functional/helpers.lua:752>
Problem: Cursorline highlight not removed in some situation. (Vitaly
Yashin)
Solution: Reset last_cursorline when resetting 'cursorline'. (Christian
Brabandt, closesvim/vim#3481)
8c63e0ec31
The scrolling region is always local to a single grid_scroll event, use
local variables and parameters instead.
The invocation of reset_scroll_region in grid_resize is cargo-culted to use
margin reset under exactly the same circumstances, not sure if it is necessary
though.
The statusline may incorporate b:term_title, so redraw it when that
title changes.
Introduce a new function status_redraw_buf to redraw windows associated
with the current buffer.
When building with -DEXITFREE, the ILOG call would result in a crash
trying to access VV_PROGPATH, which had already been released:
(gdb) bt
#0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51
#1 0x00007f8f761082f1 in __GI_abort () at abort.c:79
#2 0x00007f8f760ffa8a in __assert_fail_base (fmt=0x7f8f76253ec8 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n",
assertion=assertion@entry=0x1c74280 <.str.8> " ,\t\n", file=file@entry=0x1c73fe2 "256]'", line=line@entry=610,
function=function@entry=0x1c742e0 <.str.9+32> "") at assert.c:92
#3 0x00007f8f760ffb02 in __GI___assert_fail (assertion=0x1c74280 <.str.8> " ,\t\n", file=0x1c73fe2 "256]'", line=610,
function=0x1c742e0 <.str.9+32> "") at assert.c:101
#4 0x00000000012d87c1 in vim_getenv (name=0x2f5a460 <get_special_key_name.string+64> "NVIM_LOG_FILE") at ../src/nvim/os/env.c:608
#5 0x00000000012d6538 in expand_env_esc (srcp=0x1c2f4e0 <.str.10+32> "",
dst=0x2f5a460 <get_special_key_name.string+64> "NVIM_LOG_FILE", dstlen=4095, esc=false, one=false, prefix=0x0)
at ../src/nvim/os/env.c:351
#6 0x00000000012d85af in expand_env_esc (srcp=0x625000000100 "\004", dst=0x7ffeed88cf40 "", dstlen=32766, esc=237, one=136,
prefix=0x60200401c8b4 <error: Cannot access memory at address 0x60200401c8b4>) at ../src/nvim/os/env.c:472
#7 0x0000000000eb4274 in do_log_to_file (log_file=0x0, log_level=0, context=0x0, func_name=0x0, line_num=0, eol=false, fmt=0x0)
at ../src/nvim/log.c:254
#8 0x0000000000eb305b in open_log_file () at ../src/nvim/log.c:164
#9 0x0000000000eb2cc6 in logmsg (log_level=<error reading variable: Cannot access memory at address 0x268>,
context=<error reading variable: Cannot access memory at address 0x260>,
func_name=<error reading variable: Cannot access memory at address 0x258>,
line_num=<error reading variable: Cannot access memory at address 0x254>,
eol=<error reading variable: Cannot access memory at address 0x253>,
fmt=<error reading variable: Cannot access memory at address 0x248>) at ../src/nvim/log.c:109
#10 0x00000000013022c7 in mch_free_acl (aclent=0x4f59100) at ../src/nvim/os_unix.c:132
#11 0x0000000000efddac in getout (exitval=0) at ../src/nvim/main.c:681
#12 0x0000000000c1bb3e in ex_quit (eap=0x7ffeed88cd00) at ../src/nvim/ex_docmd.c:6067
#13 0x0000000000bab781 in do_one_cmd (cmdlinep=0x7ffeed88f180, flags=10, cstack=0x7ffeed88f1a0, fgetline=0x0, cookie=0x0)
at ../src/nvim/ex_docmd.c:2228
#14 0x0000000000b8de6d in do_cmdline (cmdline=0x7ffeed891ae2 "quit", fgetline=0x0, cookie=0x0, flags=10)
at ../src/nvim/ex_docmd.c:592
#15 0x0000000000b94036 in do_cmdline_cmd (cmd=0x7ffeed891ae2 "quit") at ../src/nvim/ex_docmd.c:268
#16 0x0000000000efb900 in exe_commands (parmp=0x7ffeed890900) at ../src/nvim/main.c:1699
#17 0x0000000000ee96b2 in main (argc=11, argv=0x7ffeed890fa8) at ../src/nvim/main.c:524
- Since the jemalloc upgrade to 5.1.0, I'm seeing weird behavior such as
infinite loops inside jemalloc routines.
- VimR maintainer reported major performance regression correlated with
jemalloc 5.1.0.
ref https://github.com/neovim/neovim/pull/7808
reverts 765515010f
Problem: printf() does not work with only one argument. (Daniel Hahler)
Solution: Allow using just the format. (Ken Takata, closesvim/vim#2687)
c71807db9c
Problem: Invalid memory access when setting wildchar empty.
Solution: Avoid going over the end of the option value. (Dominique Pelle,
closesvim/vim#1509) Make option test check all number options with
empty value.
a12e40351d
Problem: v:option_old and v:option_new are cleared when using :set in
OptionSet autocmd. (Gary Johnson)
Solution: Don't trigger OptionSet recursively.
3f3fb0b147
Resets the TUI cursor color if:
- current 'guicursor' mode does not specify a highlight group
- cursor highlight group has "inverse" or "reverse" flag
- on Nvim exit
We interpret, "inverse" to mean "default cursor".
Example:
hi Cursor guifg=bg guibg=fg
set termguicolors
set guicursor=n-v-c-sm:block,i-ci-ve:ver25-Cursor,r-cr-o:hor20
* When the cursor shape is block, its color will be "inverse"
* When the cursor shape is I-beam, its color will be `hi Cursor`.
This is useful e.g. to prevent `set listchars=eol:¬` causing your cursor
color to a low contrast color in insert mode because you cursor are
often at EOL in insert mode.
close#8572
Problem: File info message not always suppressed with 'F' in 'shortmess'.
(Asheq Imran)
Solution: Save and restore msg_silent. (Christian Brabandt, closesvim/vim#3221)
2f0f871159
ref #8840
ref #9027
When nroff justifies a line, it fills the line with whitespace to meet
$MANWIDTH. With $MANWIDTH=9999, that of course results in nonsense (and
behaves poorly with 'cursorline' option).
To work around that, instead of trying to hard-justify the lines, just
replace the mega-whitespace with a fixed size of 10 spaces.
Perhaps N/Vim needs a "soft justify" feature?
NB: existing `color default` test was actually enough to trigger the bug,
when ext_newgrid=false is used. I created the `:hi Normal` test as
I thought the builtin colors wouldn't set Normal (unless 'bg' is changed)
But as the root cause actually comes from `:hi Normal`, it makes sense
to still add the separate test (if `color default` here gets optimized to
become a no-op, or something).
- Checks for ECHOE, ICANON were left over from Vim code. We already
reference the symbols elsewhere without checking.
- newline_on_exit, intr_char: Both are vestigial remnants of Vim 4.x,
not implemented in Nvim. intr_char is a termios/stty feature, it's
probably not useful because users have other ways to configure their
terminals.
Give embeders a chance to set up nvim, by processing a request before
startup. This allows an external UI to show messages and prompts from
--cmd and buffer loading (e.g. swap files)
Problem: Test doesn't search for a sentence. Still fails when searching for
start of sentence. (Dominique Pelle)
Solution: Add paren. Check for MAXCOL in dec().
1bd999f982
Problem: It is not easy to see if a colorscheme is well written.
Solution: Add a script that checks for common mistakes. (Christian Brabandt)
200d0e36bc
loop_schedule_deferred may leak its allocated Event if the Event does
not get processed (when Nvim exits).
It only happens on exit, so we tried to ignore it in .asan-blacklist,
but for that isn't working for unknown reasons.
= ==22007==ERROR: LeakSanitizer: detected memory leaks
= Direct leak of 88 byte(s) in 1 object(s) allocated from:
= 0 0x4ef7a3 in malloc /local/mnt/workspace/tmp/ubuntu_rel/llvm/utils/release/final/llvm.src/projects/compiler-rt/lib/asan/asan_malloc_linux.cc:67:3
= 1 0xfcafc1 in try_malloc /home/travis/build/neovim/neovim/build/../src/nvim/memory.c:88:15
= 2 0xfcb1e9 in xmalloc /home/travis/build/neovim/neovim/build/../src/nvim/memory.c:122:15
= 3 0xa6e75f in loop_schedule_deferred /home/travis/build/neovim/neovim/build/../src/nvim/event/loop.c:91:19
= 4 0x18ab048 in tui_main /home/travis/build/neovim/neovim/build/../src/nvim/tui/tui.c:377:5
= 5 0x18fe120 in ui_thread_run /home/travis/build/neovim/neovim/build/../src/nvim/ui_bridge.c:102:3
= 6 0x7f09e5e58183 in start_thread /build/eglibc-ripdx6/eglibc-2.19/nptl/pthread_create.c:312
= SUMMARY: AddressSanitizer: 88 byte(s) leaked in 1 allocation(s).
Problem: Incorrect adjusting the popup menu for the preview window.
Solution: Compute position and height properl. (Ronan Pigott) Also show at
least ten items. (closesvim/vim#3414)
Problem: Recovering does not work when swap file ends in .stz.
Solution: Check for all possible swap file names. (Elfling, closesvim/vim#2395,
closesvim/vim#2396)
af903e5d49
Problem: Marks test fails in very wide window. (Vladimir Lomov)
Solution: Extend the text to match 'columns'. (closesvim/vim#3180, closesvim/vim#3181)
bde14d8e24
Vim :! may "mess up the screen" because of e.g. switching between cooked
mode, but Nvim just uses pipes. So maybe we can avoid these
redraw_later_clear() CYA calls.
Problem: maparg() and mapcheck() confuse empty and non-existing.
Solution: Return <Nop> for an existing non-empty mapping. (closesvim/vim#2940)
f88a5bc102
Problem: File names in quickfix window are not always shortened.
Solution: Shorten the file name when opening the quickfix window. (Yegappan
Lakshmanan, closesvim/vim#2851, closesvim/vim#2846)
a796d46f29
It fails PRs which otherwise would succeed (e.g. #8971), and the Codecov
status page doesn't load, so we can't see what it's complaining about.
Reverts 5c7237640f
ref #8910
Problem: Largefile test fails on CI, glob test on MS-Windows.
Solution: Remove largefile test from list of all tests. Don't run
Test_glob() on non-unix systems. More cleanup. (Yegappan
Lakshmanan, closesvim/vim#2354)
6e77df2d85
Problem: Cannot parse text with 'erroformat' without changing a quickfix
list.
Solution: Add the "text" argument to getqflist(). (Yegappan Lakshmanan)
7adf06f4e2
Problem: Quickfix list always added after current one.
Solution: Make it possible to add a quickfix list after the last one.
(Yegappan Lakshmanan)
55b6926450
This is intended to be used with source introspection tools like
clangd, where it would useful to regenerate headers and source files,
which introspection results depend on, without making a full
rebuild which takes much longer time than just generating headers.
Problem: Can only add entries to one list in the quickfix stack.
Solution: Move state variables from qf_list_T to qf_list_T. (Yegappan
Lakshmanan)
a7df8c70c8
msgpack_rpc_to_object (called by handle_request .. msgpack_rpc_to_array)
always NUL-terminates API Strings.
But handle_request .. msgpack_rpc_get_handler_for operates on a raw
msgpack_object, before preparation.
Problem: 'autowrite' takes effect when buffer is not to be written.
Solution: Don't write buffers that are not supposed to be written. (Even Q
Jones, closesvim/vim#3391) Add tests for 'autowrite'.
8c9e7b00f6
Problem: Vim becomes unusable after opening new window in BufWritePre
event.
Solution: Call not_exiting(). (Martin Tournoij, closesvim/vim#2205)
Also for "2q" when a help window is open. Add a test.
2c33d7bb69
Problem: MS-Windows users expect -? to work like --help.
Solution: Add -?. (Christian Brabandt, closesvim/vim#2867)
c3e81694fc
Include runtime/ changes from 85eee130f4 to pass 8.1.0231 tests.
Currently the "gcov" build always fails on AppVeyor. It makes the builds
very slow, so disable it for PRs until the problem is fixed.
closes#8911closes#8912
Problem: When a mapping to <Nop> times out the next mapping is skipped.
Solution: Reset "timedout" when waiting for a character. (Christian
Brabandt, closesvim/vim#2921)
83f4cbd973
Previously, when neovim would wrap a line across multiple lines,
terminal emulators could not detect that the lines represent a single
wrapped line as opposed to several separate lines. As a result, many
terminals' selection/copying functionality would treat a wrapped line as
several newline-delimited lines.
Fix this by reenabling a "special trick" from Vim. When a line is
wrapped, write the last character of that line followed by the first
character of the next line to the terminal. This hints to the terminal
that the next line is a continuation of the current line.
Extends the raw_line event with a "wrap" parameter which controls when
to do wrap hinting.
Problem: Unnecessary check for NULL pointer.
Solution: Remove the NULL check in dialog_changed(), it already happens in
dialog_msg(). (Ken Takata)
3f9a1ff141
Problem: Setting 'encoding' makes 'printheader' invalid.
Solution: Do not translate the default value of 'printheader'. (Yasuhiro
Matsumoto, closesvim/vim#2026)
0903d56f5c
Make sure cmdline updates will receive highlight specifications the same
way as screen cells. This is controlled by the ext_newgrid option so
nothing is changed by default (as screen cells are also not changed by
default). This was already done for the cmdline itself in #8221, this
extends it to cmdline_block. Which currently doesn't store highlights,
but the placeholder should be one that makes sense for future use.
This avoids having to maintain this manually.
Tried to bring this upstream to Vim, but it does not work with non-GNU
make apparently in an easy way (https://github.com/vim/vim/pull/3376).
For Neovim GNU make can be used for this apparently.
diff (generated via):
$(shell printf '%s\n' $(NEW_TESTS) > 1)
$(shell printf '%s\n' $(NEW_TESTS2) > 2)
$(error $(shell echo $(NEW_TESTS) | wc -c) $(shell echo $(NEW_TESTS2) | wc -c))
Changes:
- Removed, already included in test_alot:
- test_changedtick.res
- test_filter_map.res
- test_findfile.res
- test_ga.res
- test_lambda.res
- test_put.res
- test_recover.res
- test_tabpage.res
- Added (apparently forgotten to be added to NEW_TESTS):
- test_breakindent.res: added in 49b671f8f
- test_display.res: added in a0c7e35ee
- test_help.res, added in 41bffeacf
- test_lispwords.res, added in 54b9510e0
- test_textformat.res, added in ef39f854d
Problem: Accessing freed memory when EXITFREE is set and there is more than
one tab and window. (Dominique Pelle)
Solution: Free options later. Skip redraw when exiting.
4f1982800f
This note in runtime/doc/Makefile explains the special-case:
index.html is the starting point for HTML, but for the help files it
is help.txt. Therefore use vimindex.html for index.txt.
fix#8907
Problem: Invalid memory access on exit when autocommands wipe out a buffer.
(gy741, Dominique Pelle)
Solution: Check if the buffer is still valid. (closesvim/vim#2449)
606d45ccd8
dir (param) is of type Direction (enum).
cpo_J, found_dot, noskip (local var) are bool. cpo_J is const.
startlnum (local var) is const int. Declare it in same scope as cpo_J.
Problem: Auto indenting breaks inserting a block.
Solution: Do not check for cursor movement if indent was changed. (Christian
Brabandt, closesvim/vim#2778)
8c87a2b1fe
Problem: Bogus characters appear when indenting kicks in while doing a
visual-block append.
Solution: Recompute when indenting is done. (Christian Brabandt)
e2e69e4813
Problem: 'winfixwidth' is not always respected by :close.
Solution: Prefer a frame without 'winfixwidth' or 'winfixheight'. (Jason
Franklin)
c136af29c0
Problem: When 'wfh' is set ":bel 10new" scrolls window. (Andrew Pyatkov)
Solution: Set the fraction before changing the window height. (closesvim/vim#2798)
98da6ecab9
Problem: "gf" and <cfile> don't accept ? and & in URL. (Dmitrii Tcyganok)
Solution: Check for a URL and allow for extra characters. (closesvim/vim#2493)
9e3dfc6501
Problem: Possible crash when window can be zero lines high. (Joseph
Dornisch)
Solution: Only set w_fraction if the window is at least two lines high.
3679c17917
Problem: File name not displayed with ":file" when 'F' is in 'shortmess'.
Solution: Always display the file name when there is no argument (Christian
Brabandt, closesvim/vim#3070)
fc0896093ccloses#8817closes#8873
In the case of nested repos, such as when "neovim/" repo contains
"neovim/.vim-src/" repo,
git rev-parse --show-toplevel
returns the fullpath to the "neovim/" repo, which failed the condition.
ref #8875
ref c05d7153d3
Problem: Using ":undo 0" leaves undo in wrong state.
Solution: Instead of searching for state 1 and go above, just use the start.
(Ozaki Kiichi, closesvim/vim#2595)
ce46d934af
Problem: "cit" on an empty HTML tag changes the whole tag.
Solution: Only adjust the area in Visual mode. (Andy Massimino,
closesvim/vim#3332)
b476cb7d8d
Problem: Accessing invalid memory with "it". (Dominique Pelle)
Solution: Avoid going over the end of the line. (Christian Brabandt,
closesvim/vim#2532)
82846a00ac
Problem: C indent wrong when * immediately follows comment. (John Bowler)
Solution: Do not see "/*" after "*" as a comment start. (closesvim/vim#2321)
f8c53d3d26
Problem: ml_get error with :psearch in buffer without a name. (Dominique
Pelle)
Solution: Use the buffer number instead of the file name. Check the cursor
position.
c31f9ae4f1
Vim supports multiple locations for the user's vimrc, so it will use the
first one that is found, ignoring the rest. Nvim follows the XDG spec,
so there is only one place to look for the user's vimrc, thus making the
statement unnecessary and confusing for nvim users.
Ref #8871
On macOS the output from `ps -o comm` might contain the full path of the
executable. The `ucomm` would be the basename only but is less portable
(see previous commit).
Problem: Completion for user names does not work if a prefix is also a full
matching name. (Nazri Ramliy)
Solution: Accept both full and partial matches. (Dominique Pelle)
6c5d104302
Problem: A TextChanged autocmd triggers when it is defined after creating a
buffer.
Solution: Set b_last_changedtick when opening a buffer. (Hirohito Highlight,
closesvim/vim#2742)
8c64a36e40
a369385009 fixed this for "~/.cache/nvim-deps/", but strangely not for
"~/.cache/nvim-deps-downloads/".
ref a369385009
ref #8316
ref #8281
Seen in https://travis-ci.org/neovim/neovim/jobs/414982972 :
Using third-party dependencies from Travis cache (last update: Aug 11 23:00:15 2018).
cp: /Users/travis/build/neovim/neovim/deps-downloads/nvim-deps-downloads/…/nvim-deps-downloads/libvterm/a9c7c6fd20fa35e0ad3e0e98901ca12dfca9c25c.tar.gz:
name too long (not copied)
Problem: Integer overflow when using regexp pattern. (geeknik)
Solution: Use a long instead of int. (Christian Brabandt, closesvim/vim#2251)
2c7b906afb
The arrays are put in their own file: terminfo_defs.h. This makes it simple to
update all terminfo entries in one go using scripts/update_terminfo.sh.
The entries were compiled with tic 6.1 and some use the extended number format.
Problem: Using setline() in TextChangedI splits undo.
Solution: Use another solution for undo not working properly.
9fa9506853
This deviates from Vim in the handling of the CursorHoldI event. In
Vim, any buffer changes are merged into the insert. In Neovim,
CursorHoldI is handled via the multiqueue, and the point at which
the cursor hold is implemented (in input.c) doesn't know enough about it.
Making all queued events merge into the insert seems more wronger since
changes by other asynchronous events really should be separately
undoable.
Problem: Popup menu displayed wrong when using autocmd.
Solution: Use aucmd_prepbuf(). Force updating status line if the popup menu
is going to be redrawn anyway. (Christian Brabandt, closesvim/vim#3009)
6ba3ec1bac
Problem: Swap file warning for a file in a non-existing directory, if there
is another with the same file name. (Juergen Weigert)
Solution: When expanding the file name fails compare the file names.
8c3169c58e
Problem: seq_cur of undotree() wrong after undo.
Solution: Get the actual sequence number instead of decrementing the current
one. (Ozaki Kiichi, closesvim/vim#2319)
80eaddd3a0
Problem: Matchstrpos() without a match returns too many items.
Solution: Also remove the second item when the position is beyond the end of
the string. (Hirohito Higashi) Use an enum for the type.
8d9f0ef5c6
When rbenv is used for managing Ruby installations, just checking for an
executable called "neovim-ruby-host" is not enough. It has to be run as well.
If it does not return 0, then neovim-ruby-host is merely a shim for another Ruby
installation.
$ rbenv versions
2.5.0
* 2.5.1 (set by /Users/mhi/.rbenv/version)
$ rbenv whence neovim-ruby-host
2.5.0
$ which neovim-ruby-host
/Users/mhi/.rbenv/shims/neovim-ruby-host
$ neovim-ruby-host
rbenv: neovim-ruby-host: command not found
The `neovim-ruby-host' command exists in these Ruby versions:
2.5.0
$ echo $?
127
Additionally, the detection logic was moved from provider#ruby#Detect() to
s:detect(), because the former is run in the sandbox which forbids calling
system().
Problem: When making a vertical split the mode message isn't always
updated, "VISUAL" remains. (Alexei Averchenko)
Solution: Only reset clear_cmdline when filling all columns of the last
screen line. (Tom M. closesvim/vim#2611)
5bab555c2f
Problem: Acessing freed memory after window command in auto command.
(gy741)
Solution: Adjust the pointer in the parent frame. (Christian Brabandt,
closesvim/vim#2467)
6f361c9912
Problem: Virtual replace test fails on MS-Windows.
Solution: Make adding a termcap entry work for a builtin terminal.
Restore terminal keys in a better way.
e780848150
Problem: Crash when setting t_Co to zero when 'termguicolors' is set.
Solution: Use IS_CTERM instead of checking the number of colors.
(closesvim/vim#2710)
f708ac592f
Problem: wrong highlighting with combination of match and 'cursorline'.
Solution: Use "line_attr" when appropriate. (Ozaki Kiichi, closesvim/vim#2111)
But don't highlight more than one character.
0aa398f55a
Problem: Crash if timer closes window while dragging status line.
Solution: Check if the window still exists. (Yasuhiro Matsumoto, closes
vim/vim#1979)
989a70c590
Problem: When a function invoked from a timer calls confirm() and the user
types CTRL-C then Vim hangs.
Solution: Reset typebuf_was_filled. (Ozaki Kiichi, closesvim/vim#1791)
4eb6531b03
Refactor the following affected functions:
- prt_line_number()
- prt_header()
These use prt_set_font(), refactored in 872a76f3dac14fdcb7dc7b34cecd3acb4370d56f for TriState.
This is a regression after PR #7704:
mac: Set $LANG based on the system locale
CFStringGetCStringPtr sometimes returns "lang_region" = NULL,
in this case CFStringGetCString is used instead,
which places output to "buf", but "buf" was not used
by the code.
Problem: The :highlight command causes a redraw even when nothing changed.
Solution: Only set "need_highlight_changed" when an attribute changed.
99433291b1
Problem: With 8 colors the bold attribute is not set properly.
Solution: Move setting HL_TABLE() out of lookup_color. (closesvim/vim#1901)
12d853fae1
Use TriState on lookup_color() to avoid 'NOLINT' comments.
Add const on parameters and variables.
Update declarations to avoid typecasts.
Use `sizeof(*ptr)` for malloc() to reduce effect of type changes.
ie. short to int16_t
Update syn_compare_stub() variable declarations for consistency.
Declare and initialize variables on same line if possible.
Add const to parameters and variables.
Use bool for any parameter,variable using TRUE/FALSE macros.
Replace 'short' type with 'int16_t'.
Problem: There is no way to notice that the quickfix window contents has
changed.
Solution: Increment b:changedtick when updating the quickfix window.
(Yegappan Lakshmanan)
a8788f4d0b
Problem: Minor issues related to quickfix.
Solution: Set the proper return status for all cases in setqflist() and at
test cases for this. Move the "adding" flag outside of
FEAT_WINDOWS. Minor update to the setqflist() help text. (Yegappan
Lakshmanan)
86f100dc09
closes#8362
Vim's code calls `call_shell` directly from `get_system_output_as_rettv`
whereas in Nvim this function has been rewritten to not call `call_shell` but to call
`os_system` via `do_os_system`, losing the support for profiling and verbose.
Changing the code to call `call_shell` from `get_system_output_as_rettv`
seems to be too complicated to be worth it on the current version of the
code. So this commit duplicates the relevant code.
This would need to get `expand`ed to not become empty, and is being
handled by s:check_bin already.
`s:check_bin` will also complain about e.g.
"~/.pyenv/versions/3.6.6/bin/python" not being executable, but that
reflects that the host will fail to start with it.
Fixes#8778
Problem: Signs can be drawn on top of console messages.
Solution: don't redraw at a prompt or when scrolled up. (Christian Brabandt,
closesvim/vim#1907)
0792048842
The following (run as a script) used to cause a crash due to :sign using a
special redraw (not updating nvim's specific highlight data structures)
without proper redraw first, as split just flags for redraw later.
set cursorline
sign define piet text=>> texthl=Search
split
sign place 3 line=2 name=piet buffer=1
Problem: Test for MS-Windows $HOME always passes.
Solution: Rename the test function. Make the test pass.
dde6034111
Assume $HOME contains the literal absolute path.
Nested environment variable expansion is unsupported.
Problem: MS-Windows: Problem with $HOME when is was set internally.
Solution: Only use the $HOME default internally. (Yasuhiro Matsumoto, closes
vim/vim#2013)
48340b62e8
Restore vim_getenv() behaviour for $HOME on Windows.
closes#8727
Before:
INFO 180711.233956 11124 main:560: starting main loop
After:
INFO 180711.233956.807 11124 main:560: starting main loop
Note:
- Can't use uv_hrtime() nor uv_now(), they are not "since the epoch".
Also, log.c can't assume a loop exists.
- Log-level name (INFO/ERROR/…) should be in the first column, so that
filtering by log-level is maximally trivial.
- Use 2-digit year. 4-digit year is useless, logs don't survive for
decades without context.
Before:
2018/07/05 17:49:41 INFO 27596 on_process_exit:393: foo
After:
INFO 180705.174941 27596 on_process_exit:393: foo
the first implemented UI protocol clients (python-gui and builitin TUI)
allowed the cleared region to be restricted by setting the scroll region.
This was never used by nvim though, and not documented and implemented by
newer clients, to check we remain compatible with both kind of clients,
ensure the scroll region is in a reset state.
Problem: .po files do not use recommended names.
Solution: Give a warning if the recommended name is not used. Accept the
recommended name for conversion. (Christian Brabandt, Ken Takata)
d1d037e901
Problem: Translating messages is not ideal.
Solution: Add a remark about obsolete messages. Use msgfmt in the check
script. (Christian Brabandt)
aaef1bae3c
Problem: The script to check translations fails if there is more than one
NL in one line.
Solution: Count the number of NL characters. Make count() accept a string.
9966b21a57
Problem: The script to check translations can be improved.
Solution: Restore the view when no errors are found. Check for matching
line break at the end of the message. (Christian Brabandt)
7f93703149
Problem: CTRL-G j in Insert mode is incorrect when 'virtualedit' is set.
Solution: Take coladd into account. (Christian Brabandt, closesvim/vim#2743)
db0eedec16
Problem: :packadd does not load packages from the "start" directory.
(Alejandro Hernandez)
Solution: Make :packadd look in the "start" directory if those packages were
not loaded on startup.
9e1d399e63
Add ext_newgrid and ext_hlstate extensions. These use predefined
highlights and line-segment based updates, for efficiency and
simplicity.. The ext_hlstate extension in addition allows semantic
identification of builtin and syntax highlights.
Reimplement the old char-based updates in the remote UI layer, for
compatibility. For the moment, this is still the default. The bulitin
TUI uses the new line-based protocol.
cmdline uses curwin cursor position when ext_cmdline is active.
Problem: Access memory beyond end of string. (Coverity)
Solution: Keep allocated memory in separate pointer. Avoid outputting the
NUL character.
cdd09aa51a
Problem: ":if 0|syntax {on,off}|endif" skips the default of "syntax on"
because the executor was setting the `did_syntax_onoff` flag even though
"syntax {on,off}" is not actually executed.
closes#8728
Problem: The message for yanking doesn't indicate the register.
Solution: Show the register name in the "N lines yanked" message. (Lemonboy,
closesvim/vim#1803, closesvim/vim#1809)
e45deb7997
If a second and third child exit while we are already in the handler, we
will only see a single additional SIGCHLD. Therefore the handler must
not stop after processing a single child but should check all children.
Fixes#8740
channel.c: WIP remove redundant method check and added FUNC_ATTR_NONNULL_ALL macro
channel.c channel_defs.h helpers.c: added Error field to RequestEvent, added no_op handler func
channel.c: use const char* instead of string and cleanup
channel.c; channel_defs.h; helpers.c: removed error from event again; send errors directly to the channel without using handlers and events
channel.c: fixed memory leak and lint errors
api/private/dispatch.c; api/vim.c; msgpack_rpc/channel.c msgpack_rpc/helpers.c added Error* field to msgpack_get_handler_for; further refactored channel.c
channel.c:323 changed order of evaluation in if statement
channel.c: removed superflous whitespace
dispatch.c: review comment
It seems as though in an AppImage there's an extra child process that
dies at some early point, before we have set up a SIGCHLD handler. So
when we later get a SIGCHLD from a child that we do care about,
waitpid(-1, ...) tells us about the extra child - and we don't notice
that the interesting child has exited.
Or something like that!
See also:
* https://patchwork.kernel.org/patch/9949491/ in which perf hit
something similar
* discussion at the AppImage repository:
https://github.com/AppImage/AppImageKit/issues/812#issuecomment-404662110.
Fix is to be explicit about which process we are waitpid()'ing for, so
we never need be distracted by children that we don't know about.
Fixes#8104
Problem: The :global command does not work recursively, which makes it
difficult to execute a command on a line where one pattern matches
and another does not match. (Miles Cranmer)
Solution: Allow for recursion if it is for only one line. (closesvim/vim#1760)
f84b122a99
Show a proper confirmation dialog when trying to unload a terminal buffer while
the confirm option is set or when :confirm is used.
Fixes https://github.com/neovim/neovim/issues/4651
* vim-patch:8.0.0474: the client-server feature is not tested
Problem: The client-server feature is not tested.
Solution: Add a test.
15bf76d40b
* vim-patch:8.0.0475: not enough testing for the client-server feature
Problem: Not enough testing for the client-server feature.
Solution: Add more tests. Add the remote_startserver() function. Fix that
a locally evaluated expression uses function-local variables.
7416f3e73a
* vim-patch:8.0.0492: a failing client-server request can make Vim hang
Problem: A failing client-server request can make Vim hang.
Solution: Add a timeout argument to functions that wait.
81b9d0bd5c
Include src/nvim/testdir/test_clientserver.vim changes from
patches 8.0.0477, 8.0.0479.
* vim-patch:8.0.0633: the client-server test is still a bit flaky
Problem: The client-server test is still a bit flaky.
Solution: Wait a bit for the GUI to start. Check that the version number
can be obtained.
60964f6874
Include src/nvim/testdir/test_clientserver.vim changes
from patches 8.0.0507, 8.0.0511.
* vim-patch:8.0.1251: invalid expressin passed to WaitFor()
Problem: Invalid expressin passed to WaitFor().
Solution: Check if the variable exists.
d97fbf171e
Now that uv.h is directly being included, pre-processing of
test/includes/pre/uv.h fails on Linux with
In file included from «SRCDIR»/neovim/test/includes/pre/uv.h:1:
In file included from /usr/include/uv.h:62:
/usr/include/uv/unix.h:72:11: fatal error: 'uv/pthread-barrier.h' file not found
# include "uv/pthread-barrier.h"
^~~~~~~~~~~~~~~~~~~~~~
1 error generated.
This happens because we're missing -D_GNU_SOURCE (part of ${gen_cdefs}),
which makes the pthread_barrier_* functionality visible.
libuv users are only supposed to directly include uv.h. In v1.21.0, all
the uv-*.h headers were renamed to uv/*.h, which caused the unit tests
to fail with
[123/125] Generating post/uv-errno.h
FAILED: test/includes/post/uv-errno.h
cd «SRCDIR»/src/neovim/build/test/includes && /usr/bin/clang -std=c99 -E -P «SRCDIR»/src/neovim/test/includes/pre/uv-errno.h -I/usr/include -I/usr/include -o «SRCDIR»/neovim/build/test/includes/post/uv-errno.h
«SRCDIR»/src/neovim/test/includes/pre/uv-errno.h:1:10: error: 'uv-errno.h' file not found with <angled> include; use "quotes" instead
#include <uv-errno.h>
^~~~~~~~~~~~
"uv-errno.h"
The intention of the file is to extend libuv's error constants with more
values used by the unit tests. This can just as easily be achieved
without poking into pseudo-private header files.
Problem: Recorded key sequences may become invalid.
Solution: Add back KE_SNIFF removed in 7.4.1433. Use fixed numbers for the
key_extra enum.
8858498516
Longer summary lines for commits may be wrapped when downloading the
commit from GitHub. Unwrapping the lines ensures they can be compared
with the raw Vim commits.
The difference before and after this patch using:
./scripts/vim-patch.sh -r 8684
Before:
✘ Wrong commit message.
Expected:
vim-patch:8.0.1464: completing directory after :find does not add slash
Problem: Completing directory after :find does not add slash.
Solution: Adjust the flags for globpath(). (Genki Sky)
vim/vim@8a37b03
Actual:
vim-patch:8.0.1464: completing directory after :find does not
add slash
Problem: Completing directory after :find does not add slash.
Solution: Adjust the flags for globpath(). (Genki Sky)
After:
✔ Found expected commit message.
[ci skip]
Problem: MS-Windows: when 'clipboard' is "unnamed" yyp does not work in a
:global command.
Solution: When setting the clipboard was postponed, do not clear the
register.
3fcfa35f82
closes#8466closes#8664
Regression by 0d7daaad98.
- Fix length comparison.
- Fix loop(s) which iterated over all fields of array `pcc` even if it
was not filled up (try unicode 0x9f as statusline character).
Note about the tests:
- To input unicode with more than two hex digits you can use <C-v>U...:
a + U+fe20: a︠
a + U+fe20 + U+fe21: a︠︡
urxvt supports the xterm sequence for focus reporting in master / CVS
since 2016 [1].
This has not been released yet, but is expected to be in 9.23.
In 3d0ee17 a special case for rxvt was added, which requires a custom
urxvt extension, see #7578.
Since urxvt 9.23 is not released still, and extensions for handling this
manually appear to be in use, this sends both escape sequences for rxvt.
Fixes https://github.com/neovim/neovim/issues/8695.
1: 75264fa544 (diff-2c8f5590ce4fa7495edcf7572c89c77b)
It was not working for me in different terminals.
This patch makes it work in the same way like reverse.
Test:
:hi jediUsage cterm=standout | hi jediUsage
Problem: Still too many old style tests.
Solution: Convert a few more tests to new style. (Yegappan Lakshmanan,
closesvim/vim#2272)
430dc5d360
Skip Test_tag_file_encoding.
083_tag_search_with_file_encoding_spec.lua handles the same test case.
Problem: When typing CTRL-L in a window that's not the first one, another
redraw will happen later. (Christian Brabandt)
Solution: Reset must_redraw after calling screenclear().
9f5f7bf4d5
Use LuaJIT FFI to create char pointer.
Validate output with utf_ptr2char(), vim_iswordc() and vim_iswordp().
Use const for LuaJIT string-to-char conversion.
Problem: Characters below 256 that are not one byte are not always
recognized as word characters.
Solution: Make vim_iswordc() and vim_iswordp() work the same way. Add a test
for this. (Ozaki Kiichi)
4019cf90b8
Problem: Duplication of code for adding a list or dict return value.
Solution: Add rettv_dict_set() and rettv_list_set(). (Yegappan Lakshmanan)
45cf6e910c
closes#7383closes#7715
This implements the compromise described in #7383:
* low-priority CursorLine if foreground is not set
* high-priority ("same as Vim" priority) CursorLine if foreground is set
ref d1874ab282
ref 56eda2aa17
Problem: Cannot resize the window in a FileType autocommand. (Ingo Karkat)
Solution: Add the CMDWIN flag to :resize. (test by Ingo Karkat,
closesvim/vim#1804)
9c4fefffb6
Problem: Setting 'filetype' internally may cause the current buffer and
window to change unexpectedly.
Solution: Set curbuf_lock. (closesvim/vim#1734)
1814183b86
Problem: Freeing wrong memory when manipulating buffers in autocommands.
(James McCoy)
Solution: Also set the w_s pointer if w_buffer was NULL.
f1d13478e3
Problem: Problems with autocommands when opening help.
Solution: Avoid using invalid "varp" value. Allow using :wincmd if buffer
is locked. (closesvim/vim#1806, closesvim/vim#1804)
163095f088
Problem: writefile() continues after detecting an error.
Solution: Bail out as soon as an error is detected. (suggestions by Nikolai
Pavlov, closesvim/vim#1476)
8cf91286ca
Problem: Saving the redo buffer only works one time, resulting in the "."
command not working well for a function call inside another
function call. (Ingo Karkat)
Solution: Save the redo buffer at every user function call. (closesvim/vim#1619)
d4863aa99e
* vim-patch:8.0.0538: no test for falling back to default term value
Problem: No test for falling back to default term value.
Solution: Add a test.
85045a73db
* vim-patch:8.0.0539: startup test fails on Mac
Problem: Startup test fails on Mac.
Solution: Use another term name, "unknown" is known. Avoid a 2 second delay.
08f88b139d
* oldtest: nvim does not support ':set term'
The `external_blobs` target was not copying dependency files into
the correct subdirectories. This caused a bug where nvim-qt wouldn't
start because `qwindows.dll` was not located in the `platforms`
subdirectory.
closes https://github.com/equalsraf/neovim-qt/issues/422
Problem: When 'wrapscan' is off "gn" does not select the whole pattern when
it's the last one in the text. (KeyboardFire)
Solution: Check if the search fails. (Christian Brabandt, closesvim/vim#1683)
add8dce38d
Compare `$VIRTUAL_ENV` to `python_bin`.
This is necessary when `g:python_host_prog` is set to an absolute path,
and looking up `pyname` in `$PATH` yields another result.
vim-patch:8.0.1836: buffer-local window options may not be recent
Problem: Buffer-local window options may not be recent if the buffer is
still open in another window.
Solution: Copy the options from the window instead of the outdated window
options. (Bjorn Linse, closesvim/vim#2336)
25782a7ff4
Problem: The ":stag" command does not respect 'switchbuf'.
Solution: Check 'switchbuf' for tag commands that may open a new window.
(Ingo Karkat, closesvim/vim#1681) Define macros for the return values
of getfile().
8ad80dea08
Problem: After aborting an Ex command g< does not work. (Marcin
Szamotulski)
Solution: Postpone clearing scrollback messages to until the command line
has been entered. Also fix that the screen isn't redrawn if after
g< the command line is cancelled.
f2405ed232
vim-patch:8.0.0632: the quotestar test is still a bit flaky
Problem: The quotestar test is still a bit flaky.
Solution: Kill any existing server to make the retry work. Wait for the
register to be filled.
1c13c0fe3e
vim-patch:8.0.1536: quotestar test is flaky when using the GUI
Problem: Quotestar test is flaky when using the GUI.
Solution: Add check that the star register arrived at the server. Increase
timeouts.
791010e648
Problem: When setting the cterm background with ":hi Normal" the value of
'background' may be set wrongly.
Solution: Check that the color is less than 16. Don't set 'background' when
it was set explicitly. (Lemonboy, closesvim/vim#1710)
1615b36b91
Restore reset_option_was_set(), removed in 419da839e0
ref #8595
ref #8597
vim-patch:8.0.0469: compiler warnings on MS-Windows
Problem: Compiler warnings on MS-Windows.
Solution: Add type casts. (Christian Brabandt)
0c0d4eca4d
vim-patch:8.0.0581: moving folded text is sometimes not correct
Problem: Moving folded text is sometimes not correct.
Solution: Bail out when "move_end" is zero. (Matthew Malcomson)
94be619e30
vim-patch:8.0.0583: fold test hangs on MS-Windows
Problem: Fold test hangs on MS-Windows.
Solution: Avoid overflow in compare.
b11c826ddc
Problem: In Insert mode, CTRL-N at start of the buffer does not work
correctly. (zuloloxi)
Solution: Wrap around the start of the buffer. (Christian Brabandt)
24a9e348aa
Problem: Swap file exists briefly when opening the command window.
Solution: Set the noswapfile command modifier before splitting the window.
(James McCoy, closesvim/vim#1620)
3bab93998d
Problem: Illegal memory access with :z and large count.
Solution: Check for number overflow, using long instead of int. (Dominique
Pelle, closesvim/vim#1612)
fa0ad0bb0b
This ensures that special characters in the variables are not
interpreted as escapes/format characters in the printf string, as was
seen with upstream patch 8.0.0615.
$ ./scripts/vim-patch.sh -p 8.0.0615
Updating Vim sources: /$HOME/src/neovim/.vim-src
✔ Updated Vim sources.
✔ Found Vim revision 'bf15b8d78b22661db8b19d662b62bb9a061cdd37'.
Creating patch...
Pre-processing patch...
✔ Saved patch to '/$HOME/src/neovim/vim-8.0.0615.patch'.
Fetching 'origin/master'.
✔ From https://github.com/neovim/neovim
* branch master -> FETCH_HEAD
Creating new branch 'vim-8.0.0615' based on 'origin/master'.
✔ Switched to a new branch 'vim-8.0.0615'
Branch 'vim-8.0.0615' set up to track remote branch 'master' from 'origin'.
Creating empty commit with correct commit message.
./scripts/vim-patch.sh: line 40: printf: `w': invalid format character
✔ [vim-8.0.0615 db7fbb016] vim-patch:8.0.0615: using ./scripts/vim-patch.sh: line 44: printf: `w': invalid format character
✘ [vim-8.0.0615 db7fbb016] vim-patch:8.0.0615: using
[ci-skip]
Problem: The mode test may hang in Test_mode(). (Michael Soyka)
Solution: Set 'complete' to only search the current buffer (as suggested by
Michael)
ffea8c99d9
DWIM: avoid empty buffer 1 when stdin was empty. If other files were
specified at startup, we assume that stdin is only accidentally
not-a-TTY: user did not intend to send text from it.
ref #8560
ref #8561
If stdin is not a TTY we read it into buffer 1, as text. But if the
stdin pipe is empty, Nvim was most likely invoked for some other reason.
DWIM: select buffer 2 (if it exists). Example:
echo file1 | xargs nvim
closes#8560closes#8561
ref https://github.com/equalsraf/neovim-qt/issues/417
There's a mix of CXX and C related variables being set/referenced in our
CMake files. Since we only use C, use an explicit language list of "C"
instead of the implicit "C CXX" and replace all uses of CXX variables
with their C counterparts
The Debian hurd-i386 [build] failed (partly) due to -D_GNU_SOURCE not be
defined:
[215/286] /usr/bin/cc -DINCLUDE_GENERATED_DECLARATIONS -Iconfig -I../src -Isrc/nvim/auto -Iinclude -I/usr/include/luajit-2.1 -g -O2 -fdebug-prefix-map=/<<PKGBUILDDIR>>=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -DDISABLE_LOG -Wdate-time -D_FORTIFY_SOURCE=2 -Wconversion -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=1 -DNVIM_MSGPACK_HAS_FLOAT32 -DNVIM_UNIBI_HAS_VAR_FROM -O2 -g -DMIN_LOG_LEVEL=3 -Og -g -Wall -Wextra -pedantic -Wno-unused-parameter -Wstrict-prototypes -std=gnu99 -Wimplicit-fallthrough -Wvla -fstack-protector-strong -fdiagnostics-color=auto -Wno-array-bounds -MD -MT src/nvim/CMakeFiles/nvim.dir/os/pty_process_unix.c.o -MF src/nvim/CMakeFiles/nvim.dir/os/pty_process_unix.c.o.d -o src/nvim/CMakeFiles/nvim.dir/os/pty_process_unix.c.o -c ../src/nvim/os/pty_process_unix.c
../src/nvim/os/pty_process_unix.c: In function 'pty_process_tty_name':
../src/nvim/os/pty_process_unix.c:121:10: warning: implicit declaration of function 'ptsname'; did you mean 'ttyname'? [-Wimplicit-function-declaration]
return ptsname(ptyproc->tty_fd);
Hurd is obviously not Linux, but it is using a GNU compiler and glibc so
it needs -D_GNU_SOURCE for the ptsname() definition to be visible.
[build]: https://buildd.debian.org/status/fetch.php?pkg=neovim&arch=hurd-i386&ver=0.3.0-2&stamp=1528981349&raw=0
Enabling CMake's USE_FOLDERS option and adding the FOLDER property to
targets allows some IDEs to list the targets in an organized
hierarchy of folders.
Before this change, -E/-Es without `-u NONE` reads stdin as Ex commands.
It should always read stdin as text (into buffer 1), like this:
echo foo | nvim -Es +'%p'
foo
echo foo | nvim -Es -u NORC +'%p'
foo
This changes Ex mode (Q, -e) to work like Vim's "improved Ex mode"
(gQ, -E). That brings some small behavior differences, but should not
impact most Ex scripts (unless, for example, they depend on mappings
being disabled--but that can be solved for -e by skipping user config).
Before this change:
* the screen test hangs.
After this change:
* Q acts like gQ.
* -e/-es differs from -E/-Es only in its treatment of stdin.
This moves towards potentially removing getexmodeline().
(HINT: That does NOT mean "removing Ex mode", it means removing the
Vi-compatible Ex mode, which differs from Vim's "improved Ex mode" only
in some minor details (e.g. mappings are disabled).)
ref #1089 :-)~
Problem: For some people the hint about quitting is not sufficient.
Solution: Put <Enter> separately. Also use ":qa!" to get out even when
there are changes.
28a8193e31
Problem: shellescape() always escapes a newline, which does not work with
some shells. (Harm te Hennepe)
Solution: Only escape a newline when the "special" argument is non-zero.
(Christian Brabandt, closesvim/vim#1590)
206155280d
Problem: When gF fails to edit the file the cursor still moves to the found
line number.
Solution: Check the return value of do_ecmd(). (Michael Hwang)
2a79ed293c
Problem: Test_edit causes older xfce4-terminal to close. (Dominique Pelle)
Solution: Reduce number of columns to 2000. Try to restore the window
position.
ba6ec18297
Problem: Buffer overflow when 'columns' is very big. (Nikolai Pavlov)
Solution: Correctly compute where to truncate. Fix translation.
(closesvim/vim#1600)
658a3a2caf
Fix PVS warnings:
fileio.c:7293 Medium V547 Expression is always true.
fileio.c:7351 Medium V547 Expression 'event == 100' is always false.
event_name2nr returns event_T, so PVS infers that nothing outside of
that range could possibly be returned.
According to POSIX[0], only octal escapes are supported by the printf
command. GNU coreutils' printf and some shells' builtin printf versions
which support hex escapes, but dash and non-GNU printf do not.
[0]: http://pubs.opengroup.org/onlinepubs/9699919799/utilities/printf.html
Store text in ScreenLines as UTF-8, so it can be sent as-is to the UI
layer. `utfc_char2bytes(off,buf)` is removed, as `ScreenLines[off]` now
already contains this representation.
To recover the codepoints that the screen arrays previously contained, use
utfc_ptr2char (or utf_ptr2char to ignore composing chars).
NB: This commit does NOT change how screen.c processes incoming UTF-8 data
from buffers, cmdline, messages etc. Any algorithm that operates on UCS-4
(like arabic shaping, treatment of non-printable chars)
is left unchanged for now.
Problem: Crash when complete() is called after complete_add() in
'completefunc'. (Lifepillar)
Solution: Bail out if compl_pattern is NULL. (closesvim/vim#1668)
Also avoid using freed memory.
4475b62396
Environment variables are used to detect when the project is being built
from within Clion or Visual Studio, so that the build process can be
simplified by automatically building the bundled dependencies for them.
2018-06-10 09:57:12 -05:00
685 changed files with 66609 additions and 19817 deletions
@@ -27,6 +27,9 @@ You can see the name of the current swap file being used with the command:
:sw[apname] *:sw* *:swapname*
Or you can use the |swapname()| function, which also allows for seeing the
swap file name of other buffers.
The name of the swap file is normally the same as the file you are editing,
with the extension ".swp".
- On Unix, a '.' is prepended to swap file names in the same directory as the
@@ -170,4 +173,4 @@ Once you are sure the recovery is ok delete the swap file. Otherwise, you
will continue to get warning messages that the ".swp" file already exists.
vim:tw=78:ts=8:ft=help:norl:
vim:tw=78:ts=8:noet:ft=help:norl:
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
