Attempting to serialize a NULL string through msgpack results in
msgpack_sbuffer_write attempting to memcpy from a NULL pointer, which is
undefined behavior.
The existing code would cause utf8len_tab to be declared as non-extern
when main.cpp included globals.h as well as in mbyte.c. This causes the
following warning
Linking C executable ../../bin/nvim
/usr/bin/ld: Warning: size of symbol `utf8len_tab' changed from 256 in CMakeFiles/nvim.dir/main.c.o to 320 in CMakeFiles/nvim.dir/mbyte.c.o
Moving the definition to globals.h and using INIT() ensures the array is
only defined in main.cpp and other places globals.h is included see an
extern declaration.
This fixes failures with TSAN builds like
FATAL: ThreadSanitizer can not mmap the shadow memory (something is mapped at 0x55deea465000 < 0x7cf000000000)
FATAL: Make sure to compile with -fPIE and to link with -pie.
Problem: Command line editing is not tested much.
Solution: Add tests for expanding the file name and 'wildmenu'.
ae3150ec8d
The test_cmdline.vim tests were ported in 23f591dba0 and thus
should have marked 1923 as applied. The test_cmdline.vim invocation
has been moved from test_alot.vim to src/nvim/testdir/Makefile to
better accord with the upstream code.
Problem: When switching to a new buffer and an autocommand applies syntax
highlighting an ml_get error may occur.
Solution: Check "syn_buf" against the buffer in the window. (Alexander von
Buddenbrock, closesvim/vim#676)
b681be175b
Problem: When using :stopinsert in a silent mapping the "INSERT" message
isn't cleared. (Coacher)
Solution: Always clear the message. (Christian Brabandt, closesvim/vim#718)
fd773e9e88
Problem: Overwriting pointer argument.
Solution: Assign to what it points to. (Dominique Pelle)
76ae22fef3
The typos corrected in the original vim patch are no long present
in the neovim code base and the pointer assignment was done correctly
in the porting of patch 1913 (where the changes were introduced).
Problem: When setting 'filetype' there is no check for a valid name.
Solution: Only allow valid characters in 'filetype', 'syntax' and 'keymap'.
d0b5138ba4
* Fix translation of the term 'tag'
Prefer native word instead of transliteration
* Update uk.po automatically
Run make update-po-uk in src/nvim/po
* Update fuzzy translations in uk.po
* Translate missing msgstr
* Fix layout of the front page
* Apply cleanup.vim
Additionally, remove irrelevant and confusing comments
* Copy uk.po into uk.cp1251.po
Problem: Using freed memory after setqflist() and ":caddbuffer". (Dominique)
Solution: Set qf_ptr when adding the first item to the quickfix list.
8b20179c65
In order to not conflict with new error codes that Vim adds, all Neovim
error codes should be above 5000. The three existing sub-5000 error
codes (E926, E951, and E952) are now E50003, E5004, and E5005
respectively.
E953 was removed in 6167ce6df2, so just
remove it from the help.
Problem: Crash when an autocommand changes a quickfix list. (Dominique)
Solution: Check wether an entry is still valid. (Yegappan Lakshmanan,
Hirohito Higashi)
ffec3c5349
Problem: Order of matches for 'omnifunc' is messed up. (Danny Su)
Solution: Do not set compl_curr_match when called from complete_check().
(closesvim/vim#1168)
472e85970e
Command "live preview" is fundamentally a non-recursive concept
("preview of a preview" is not useful). Maintaining this as a
global is less awkward and closer to what we actually want to
express, vs adorning exarg_T, CommandLineState, etc.
command_line_changed:
- Check (current_SID == 0) instead of KeyTyped
- We want to update during mappings (KeyTyped is false then).
- Check vpeekc_any()
- Avoids unnecessary work.
- Avoids triggering live preview during macros.
- Caveat: This makes the redraw "stutter" if user spams (holds a key)
in the replace pattern. But that scenario is not important.
- Update screen if the command is changed to a non-live command.
(`s->live` goes from true => false) => clears the preview
command_line_execute:
- Let CTRL-C cancel live preview
do_sub:
- Enforce a time limit ('redrawtime').
- Unset 'inccommand' if time limit is reached.
Closes#5602Closes#5585
Problem: Vertical movement after CTRL-A ends up in the wrong column.
(Urtica Dioica)
Solution: Set curswant when appropriate. (Hirohito Higashi)
8e08125d3a
- Use a standard scratch buffer instead of a new 'buftype', functions
like curbufIsChanged() already have special handling for scratch bufs.
- Cleanup some stuff from the previous merge.
- Add support for :smagic, :snomagic. Closes#5578
Add tests for undotree().
Helped-by: Björn Linse <bjorn.linse@gmail.com>
When "curhead" points to a valid head, the value of "newhead" is
meaningless (and really should be set to null). The undo state for
a buffer is _logically_ the enum:
enum UndoState {
CurrentHead(head),
NewHead(head),
EmptyTree
}
nvim _represents_ this as: whenever `curbuf->b_u_curhead` is nonnull it
should be used as the current head, and `curbuf->b_u_newhead` is
ignored. If the there is a current head, then this will be redoed on the
next redo, and its parent will be undone on next undo. Only if
`b_u_curhead` is NULL, `b_u_newhead` will be used as the head to undo
(and it is not possible to redo). Also both can be NULL, to indicate an
empty undotree. (To be fair, this only strictly true when calling undo.c
from the outside, in some places _within_ a function in undo.c both
values might be meaningful)
Apparently `undotree()` breaks this non-abstraction, this _cosmetic_
issue can easily be fixed by `ex_substitute` also saving and restoring
`b_u_newhead`, but is doesn't reflect any error really how
`u_undo_and_forget` manipulates the _actual_ state of the undo tree.
