Commit Graph

1553 Commits

Author SHA1 Message Date
James McCoy
de50c003d5 Use vim_strchr(s, c) when c may be NUL (#6656)
As part of the refactoring in #5119, some vim_strchr() were changed to
strchr().  However, vim_strchr() behaves differently than strchr() when
c is NUL, returning NULL instead of a pointer to the NUL.

Revert the strchr() calls where it isn't known whether c is NUL, since
this causes a semantic change the surrounding code doesn't expect.  In
the case of #6650, this led to a heap overrun.

Closes #6650
2017-05-03 10:12:38 +02:00
Justin M. Keyes
52727d98d7 NVIM v0.2.0
FEATURES:
    bc4a2e1576 help, man.vim: "outline" (TOC) feature #5169
    58422f17d8 'guicursor' works in the TUI (and sends info to UIs) #6423
    129f107c0c api: nvim_get_mode() #6247
    0b59f988f4 api/ui: externalize tabline #6583
    bc6d868d00 'listchars': `Whitespace` highlight group #6367
    6afa7d66cd writefile() obeys 'fsync' option #6427
    c60e409471 eval.c refactor (also improves some error messages) #5119
    9d200cd0a3 getcompletion("cmdline") #6376
    2ea7bfc627 terminal: Support extra arguments in 'shell'. #4504
    bf5110266c DirChanged autocmd #5928 #6262
    1743df82f9 'cpoptions': "_" flag to toggle `cw` behaviour #6235
    22337b1c01 CTRL-R omits trailing ^M when pasting to cmdline #6137
    0e44916fff :edit allows unescaped spaces in filename #6119
    abdbfd26bc eval: Add id() function and make printf("%p") useful #6095
    bdfa1479d2 findfile(), :find, gf work in :terminal. #6009
    2f38ed11c9 providers: Disable if `g:loaded_*` exists.
    b5560a69b1 setpos() can set lowercase marks in other buffers #5753
    7c513d646d Throttle :! output, pulse "..." message. #5396
    d2e8c76dc2 v:exiting #5651

    :terminal improvements #6185 #6142
      - cursor keeps position after leaving insert-mode.
      - 4ceec30cd0 Follows output only if cursor is at end of buffer.
      - e7bbd35c81 new option: 'scrollback'
      - fedb8443d5 quasi-support for undo and 'modifiable'
      - b45ddf731b disables 'list' by default
      - disables 'relativenumber' by default

    :help now contains full API documentation at `:help api`.

    man.vim saw numerous improvements.

    Windows support:
      - Windows is no longer "experimental", it is fully supported.
      - Windows package includes a GUI, curl.exe and other utilities.

    "Vim 8" features: partials, lambdas.

SECURITY FIXES:
    CVE-2017-5953 CVE-2017-6349 CVE-2017-6350 #6485

CHANGES:
    NVIM_TUI_ENABLE_CURSOR_SHAPE was removed. Use 'guicursor' instead.
        See https://github.com/neovim/neovim/wiki/Following-HEAD#20170402

    81525dc5c3 'mouse=a' is no longer the default. (This will probably
                 change again after it is improved.) #6022

    0c1f783164 defaults: 'showcmd', 'belloff', 'ruler' #6087
    eb0e94f71b api: {get,set}_option update local options as appropriate #6405
    bdcb2a38b3 "Reading from stdin..." message was removed. #6298

FIXES:
    12fc1defd6 ops: fix i<c-r> with multi-byte text #6524
    dd391bfca1 Windows: system() and friends #6497
    13352c00f1 Windows: os_get_hostname() #6413
    16babc6687 tui: Less-noisy mouse seqs #6411
    3a9dd13f9e (vim bug) folding edge-cases  #6207
    f6946c68ae job-control: set CLOEXEC on pty processes. #5986
    d1afd434f3 rplugin: Call s:LoadRemotePlugins() on startup.
    1215084676 backtick-expansion works with `shell=fish` #6224
    e32ec03d67 tui: Improved behavior after resize. #6202
    86c2adc074 edit.c: CTRL-SPC: Insert previously-inserted text. #6090
    c318d8e672 b:changedtick now follows VimL rules #6112
    34e24cb2f7 terminal: Initialize colors in reverse order #6160
    e8899178ec undo: Don't set b_u_curhead in ex_undojoin() #5869
    d25649fa01 undo: :earlier, g-: Set b_u_seq_cur correctly. (#6016)
    043d8ba422 'Visual-mode put from @. register' #5782
    42c922b32c open_buffer(): Do `BufEnter` for directories.
    50d0d89129 inccommand: Preview :sub commands only after delimiter #5932
    1420e10474 CheckHealth improvements #5519
    c8d5e9230e jobstart(): Return -1 if cmd is not executable. #5671
2017-05-01 22:13:23 +02:00
James McCoy
5d73a6e5df Merge pull request #6621 from jamessan/vim-7.4.2231
vim-patch:7.4.2231,7.4.2239,7.4.2244,7.4.2245,7.4.2246,7.4.2263,8.0.0150
2017-04-30 09:46:41 -04:00
Justin M. Keyes
d349f610ac 'guicursor': No color/blink by default
Closes #6577
2017-04-30 13:44:31 +02:00
James McCoy
35d817e68c oldfiles_spec: Add tests for ":filter ... oldfiles" 2017-04-29 23:58:03 -04:00
James McCoy
4488bfcfb3 oldfiles_spec: Set the shada file in the session 2017-04-29 23:48:28 -04:00
Justin M. Keyes
acfd2a2a29 input.c: Process only safe events before blocking.
Introduce multiqueue_process_priority() to process only events at or
above a certain priority.
2017-04-28 19:19:51 +02:00
Justin M. Keyes
3ea1007753 api: nvim_get_mode()
Asynchronous API functions are served immediately, which means pending
input could change the state of Nvim shortly after an async API function
result is returned.

nvim_get_mode() is different:
  - If RPCs are known to be blocked, it responds immediately (without
    flushing the input/event queue)
  - else it is handled just-in-time before waiting for input, after
    pending input was processed. This makes the result more reliable
    (but not perfect).

Internally this is handled as a special case, but _semantically_ nothing
has changed: API users never know when input flushes, so this internal
special-case doesn't violate that. As far as API users are concerned,
nvim_get_mode() is just another asynchronous API function.

In all cases nvim_get_mode() never blocks for more than the time it
takes to flush the input/event queue (~µs).

Note: This doesn't address #6166; nvim_get_mode() will provoke #6166 if
e.g. `d` is operator-pending.

Closes #6159
2017-04-28 19:14:34 +02:00
Justin M. Keyes
7044aa6e82 api/ext_tabline: curtab should be a Tabpage handle. 2017-04-28 15:59:20 +02:00
James McCoy
e20691ccb4 defaults_spec: Test changing :filetype/:syntax in -c, after defaults 2017-04-27 09:20:03 -04:00
Justin M. Keyes
6755262194 test: inccommand_spec: Avoid indeterminism. (#6592) 2017-04-26 14:51:21 +02:00
Justin M. Keyes
6944abad2f api/ext_tabline: List of Dicts. 2017-04-26 03:41:24 +02:00
Justin M. Keyes
c8e1af93de api: nvim_ui_attach(): Flatten ext_* options. 2017-04-26 03:41:24 +02:00
Justin M. Keyes
00843902d3 api/ui: externalize tabline
- Work with a bool[] array parallel to the UIWidget enum.
- Rename some functions.
- Documentation.
2017-04-26 03:41:24 +02:00
Dongdong Zhou
88023d5123 api/ui: externalize tabline 2017-04-26 02:15:31 +02:00
Justin M. Keyes
8f346a322b test/fs: sanity check for literal "~" directory (#6579)
If the CWD contains a directory with the literal name "~" then the tests
will have bogus failures.
2017-04-24 22:45:03 +02:00
Justin M. Keyes
086c354a0a api: Do not translate error messages.
Also re-word some error messages:
- "Key does not exist: %s"
- "Invalid channel: %<PRIu64>"
- "Request array size must be 4 (request) or 3 (notification)"
- "String cannot contain newlines"

References #6150
2017-04-24 00:11:27 +02:00
Sander Bosma
5c9860a0a2 api: Do not truncate errors <1 MB. #6237
Closes #5984
2017-04-23 23:44:29 +02:00
Justin M. Keyes
4524053874 test: api: Do not truncate errors <1 MB. 2017-04-23 23:24:16 +02:00
Justin M. Keyes
a9d37c928e Merge #6513 from ZyX-I/lazier-arg_errmsg-gettext 2017-04-23 01:02:39 +02:00
Justin M. Keyes
10f119ab87 Merge #6539 'More cursor shape modes' 2017-04-21 19:09:50 +02:00
sander2
f50e03f2e3 ex_cmds.c: Fix bug in ex_z (#6557)
vim-patch:8.0.0571
2017-04-21 15:45:51 +02:00
Björn Linse
48f0542ad6 tests: detect invalid helpers.sleep 2017-04-21 14:21:26 +02:00
Björn Linse
7ea5c78687 ui: use an array for mode styles 2017-04-21 12:32:38 +02:00
Björn Linse
2c5751b9b2 ui: add tests for new cursor shape modes 2017-04-21 12:32:38 +02:00
ZyX
d463c9e03a Merge branch 'master' into lazier-arg_errmsg-gettext 2017-04-21 00:33:12 +03:00
Justin M. Keyes
15af803cce Merge #6550 from ZyX-I/pvs-check-comment 2017-04-20 22:00:03 +02:00
Justin M. Keyes
17052946c7 'scrollback': Allow :setlocal -1 on normal buffers
Avoids a spurious :loadview error.
2017-04-20 13:13:39 +02:00
ZyX
c2f3e361c5 *: Add comment to all C files 2017-04-19 19:11:50 +03:00
Justin M. Keyes
45aa465fba test: Cursor after :hi clear|syntax reset
Also enable tests on Windows.
2017-04-17 21:15:51 +02:00
Matthew Malcomson
263849b2dd fold: foldMoveRange(): fix :move bug #6534
Closes #6540

In #6221 there was a mistake in calculating which folds need to be
re-ordered. When there are no folds after those that have been adjusted,
then `move_end` remains 0. This results in reverse_fold_order()
swapping folds that have been adjusted with uninitialised folds in the
remainder of the grow array.

Add a check in foldMoveRange() to account for this case.
2017-04-17 04:45:55 +02:00
Björn Linse
12fc1defd6 ops: fix i<c-r> with multi-byte text (#6524) 2017-04-15 11:19:40 +02:00
ZyX
b54e5c220f unittests: Add a test for TV_CSTRING
Not using enum{} because SIZE_MAX exceeds integer and I do not really like how
enum definition is described in C99:

1. Even though all values must fit into the chosen type (6.7.2.2, p 4) the type
   to choose is still implementation-defined.
2. 6.4.4.3 explicitly states that “an identifier declared as an enumeration
   constant has type `int`”. So it looks like “no matter what type was chosen
   for enumeration, constants will be integers”. Yet the following simple
   program:

        #include <stdint.h>
        #include <stdio.h>
        #include <stddef.h>

        enum { X=SIZE_MAX };

        int main(int argc, char **argv)
        {
          printf("x:%zu m:%zu t:%zu v:%zu",
                 sizeof(X), sizeof(SIZE_MAX), sizeof(size_t), (size_t)X);
        }

    yields one of the following using different compilers:

    - clang/gcc/pathcc: `x:8 m:8 t:8 v:18446744073709551615`
    - pcc/tcc: `x:4 m:8 t:8 v:1844674407370955161`

    If I remove the cast of X to size_t then pcc/tcc both yield `x:4 m:8 t:8
    v:4294967295`, other compilers’ output does not change.

    All compilers were called with `$compiler -std=c99 -xc -` (feeding program
    from echo), except for `tcc` which has missing `-std=c99`. `pcc` seems to
    ignore the argument though: it is perfectly fine with `-std=c1000`.
2017-04-14 23:58:47 +03:00
Justin M. Keyes
58d2ce9bdb test: check_cores(): Escape $TMPDIR path. (#6520)
Lua string:match() considers hyphen to be a special char, we want the
path to be a literal match. Also remove "./", etc.

References #6483
2017-04-14 20:34:54 +02:00
Justin M. Keyes
7c4e5dfd27 win: os_shell_is_cmdexe() + tests 2017-04-12 02:28:43 +02:00
Rui Abreu Ferreira
d31d177a0c win: default shellxescape, shellxquote to empty
Calling cmd.exe in Windows follows a very different pattern from Vim.
The primary difference is that Vim does a nested call to cmd.exe, e.g.
the following call in Vim

    system('echo a 2>&1')

spawns the following processes

    "C:\Program Files (x86)\Vim\vim80\vimrun" -s C:\Windows\system32\cmd.exe /c (echo a 2^>^&1
        ^>C:\Users\dummy\AppData\Local\Temp\VIoC169.tmp 2^>^&1)
    C:\Windows\system32\cmd.exe /c C:\Windows\system32\cmd.exe /c (echo a 2^>^&1
        ^>C:\Users\dummy\AppData\Local\Temp\VIo3C6C.tmp 2^>^&1)
    C:\Windows\system32\cmd.exe  /c (echo a 2>&1
        >C:\Users\dummy\AppData\Local\Temp\VIo3C6C.tmp 2>&1)

The escaping with ^ is needed because cmd.exe calls itself and needs to
preserve the special metacharacters for the last call. However in nvim
no nested call is made, system('') spawns a single cmd.exe process.
Setting shellxescape to "" disables escaping with ^.

The previous default for shellxquote=( wrapped any command in
parenthesis, in Vim this is more meaningful due to the use of tempfiles
to store the output and redirection (also see &shellquote). There is
a slight benefit in having the default be empty because some expressions
that run in console will not run within parens e.g. due to unbalanced
double quotes

    system('echo "a b')
2017-04-12 02:10:34 +02:00
Rui Abreu Ferreira
f3cc843755 win: libuv_process_spawn(): special-case cmd.exe
Disable CommandLineToArgvW-standard quoting for cmd.exe.

libuv assumes spawned processes follow the convention expected by
CommandLineToArgvW(). But cmd.exe is non-conformant, so for cmd.exe:
- With system([]), the caller has full control (and responsibility) to
  quote arguments correctly.
- With system(''), shell* options are used.

libuv quoting is disabled if argv[0] is:
- cmd.exe
- cmd
- $COMSPEC resolving to a path with filename cmd.exe

Closes #6329
References #6387
2017-04-12 02:10:34 +02:00
Rui Abreu Ferreira
799443c994 win/test: Enable more system() tests 2017-04-12 02:10:33 +02:00
Justin M. Keyes
d6e5f94ae9 win: defaults: 'shellredir', 'shellxquote', 'shellxescape' 2017-04-12 01:35:49 +02:00
Björn Linse
69775f603f ci: install Turkish locale and make locale tests more reliable 2017-04-11 10:24:19 +02:00
Justin M. Keyes
119f0ca854 test: helpers.execute() => helpers.feed_command() 2017-04-11 02:37:39 +02:00
Justin M. Keyes
dab3f86d09 win/test: Enable recover_spec.lua 2017-04-11 02:37:39 +02:00
Justin M. Keyes
2d29638744 test: :file {name} 2017-04-11 00:39:12 +02:00
Justin M. Keyes
6cbf290d56 test/rmdir(): fallback to Nvim delete()
Lua has too many pitfalls here:
- os.execute() requires shell-escaping
- os.execute() has breaking changes between Lua 5.1 and 5.2
- No native way in Lua to handle "readonly" etc. on Windows
2017-04-11 00:39:12 +02:00
Justin M. Keyes
0ce9618918 test/rmdir(): Remove readonly attr on Windows. 2017-04-10 22:59:30 +02:00
Justin M. Keyes
fec53f0bdf Merge #6479 from bfredl/tolower
remove vim_tolower/etc functions with broken locale-dependent behavior
2017-04-10 15:23:44 +02:00
Justin M. Keyes
dd7f41e5a0 Merge #6488 from ZyX-I/coverity-fixes 2017-04-10 14:04:19 +02:00
Björn Linse
a3a06d0248 test: add tests for gu/gU behavior in Turkish locale 2017-04-10 12:02:25 +02:00
ZyX
fa7ace446e coverity/56795: Fix NULL dereference in :syn keyword non-printable
Bug was introduced 3 years earlier, in 13848aa: NULL keyword_copy was 
incorrectly treated as an indicator of OOM.
2017-04-09 23:38:05 +03:00
ZyX
35584594f5 coverity/13689: Check file header with memcmp
Not that it is actually useful (would fail in any case), but should fix coverity 
report.
2017-04-09 22:30:48 +03:00