Additional modifications:
- More `const` qualifiers in tested functions.
- `tv_list_find_str()` second argument is more in-line with other
`tv_list_find*()` functions.
Lesser form of include-what-you-use: at least guarantees that header
file did not forget to include something through some other included
file.
Activate run_single_includes_tests on CI.
Fix some IWYU violations.
References #5321
1. Use `char *` for strings.
2. Add `const` qualifiers.
3. Add attributes and documentation.
4. Handle skipping *inside*.
5. Handle non-heredoc argument also inside: deferring this to the caller is
pointless because all callers need the same thing. Though new ex_lua caller
may live without allocations in this case, allocating nevertheless produces
cleaner code.
6. Note that all callers call script_get with `eap` and `eap->arg`. Thus second
argument is useless in practice: it is one and the same always and can be
reached through the first argument.
During free_all_mem, somehow ex_tabonly() may free aucmd_win. But it
isn't fully destroyed (maybe autocmd_busy?). When win_free_all() tries
to free aucmd_win directly, it double-frees the sub-fields.
Tried unnsuccessfully to work around this by invoking `:tabonly!` with
autocmds disabled:
diff --git a/src/nvim/memory.c b/src/nvim/memory.c
index 58c01fbe7a12..91c845e94d22 100644
--- a/src/nvim/memory.c
+++ b/src/nvim/memory.c
@@ -565,9 +565,9 @@ void free_all_mem(void)
/* Close all tabs and windows. Reset 'equalalways' to avoid redraws. */
p_ea = false;
if (first_tabpage->tp_next != NULL)
- do_cmdline_cmd("tabonly!");
+ do_cmdline_cmd("noautocmd tabonly!");
if (firstwin != lastwin)
- do_cmdline_cmd("only!");
+ do_cmdline_cmd("noautocmd only!");
/* Free all spell info. */
spell_free_all();
Problem: Crash when editing a new buffer and BufUnload autocommand wipes
out the new buffer. (Norio Takagi)
Solution: Don't allow wiping out this buffer. (partly by Hirohito Higashi)
Move old style test13 into test_autocmd. Avoid ml_get error when
editing a file.
e0ab94e712
Problem: Illegal memory access with ":1@". (Dominique Pelle)
Solution: Correct cursor column after setting the line number. Also avoid
calling end_visual_mode() when not in Visual mode.
4930a76a03
Tokenize p_sh if used as default in ex_terminal(). Previously p_sh was
used as the first arg in a list when calling termopen(), this would try
to call an untokenized version of shell, meaning if you had an argument
in 'shell':
set shell=/bin/bash\ --login
the command would fail.
Helped-by: oni-link <knil.ino@gmail.com>
Closes#3999
Fix a problem when filtering manually folded lines
When foldMarkAdjustRecurse() is called to adjust folds that start inside
the range of lines that are being moved and end outside that range, it
calculates `amount_after` for its recursive call incorrectly.
The calculation assumes that folds inside the changed range are being
deleted, but this is not always the case.
This means nested folds that start after the changed range of lines are
shifted an incorrect amount.
We fix this by calculating the `amount_after` differently if the folds
inside the changed range are not being deleted.
Problem: An address relative to the current line, ":.,+3y", does not work
properly on a closed fold. (Efraim Yawitz)
Solution: Correct for including the closed fold. (Christian Brabandt)
ded2782783
Problem: More buf_valid() calls can be optimized.
Solution: Use bufref_valid() instead.
NOTE: Some changes related to channels and the Python and Netbeans interfaces
were obviously left out.
7c0a2f367f
Closes#731
References #851
Note: This does not remove some intentional legacy usages of strncpy.
- memcpy isn't equivalent because it doesn't check the string
length of `src`, and doesn't zero-out the remainder of `dst`.
- xstrlcpy isn't equivalent because it doesn't zero-out the
remainder of `dst`. Some Vim logic depends on that (e.g.
ex_append which calls vim_strnsave).
Helped-by: Douglas Schneider <ds3@ualberta.ca>
Helped-by: oni-link <knil.ino@gmail.com>
Helped-by: James McCoy <jamessan@jamessan.com>
Abuse NOTDONE to give some nuance to the return value of readfile(), so
that open_buffer() can distinguish between "failed, lol" and "failed
because the path is a directory".
Before this change, Vim *already* creates a new buffer when a directory
is edited. So there is no reason it should not raise BufEnter, that was
an implementation detail of ye olde readfile().
Most of the changes in this commit merely preserve the old semantics.
The "implicit" change that we actually are interested in, is this line
in `open_buffer()`, where `retval` being non-FAIL allows EVENT_BUFENTER
to be applied:
apply_autocmds_retval(EVENT_BUFENTER, NULL, NULL, FALSE, curbuf, &retval);
References https://github.com/vim/vim/issues/1353
Problem: The short form of ":noswapfile" is ":noswap" instead of ":now".
(Kent Sibilev)
Solution: Only require three characters. Add a test for the short forms.
3bcfca3ab4
Problem: When there is an CmdUndefined autocmd then the error for a missing
command is E464 instead of E492. (Manuel Ortega)
Solution: Don't let the pointer be NULL.
eac784eced
Fix session load with winminheight/winminwidth >1.
Problem: Vim temporarily sets winheight/winwidth to 1 while loading
session. If user has his minimum window size set higher, this
causes an error.
Solution: Temporarily set winminheight/winminwidth, and restore the
original values (in the right order) when finishing session read.
The order of the compound 'set' command is important, if it is
wrong there will be an error.
