Adds the API function `nvim_cmd` which allows executing an Ex-command through a Dictionary which can have the same values as the return value of `nvim_parse_cmd()`. This makes it much easier to do things like passing arguments with a space to commands that otherwise may not allow it, or to make commands interpret certain characters literally when they otherwise would not.
It seems range and count can be used together in commands. This PR fixes
the behavior of `nvim_parse_cmd` for those cases by removing the mutual
exclusivity of "range" and "count". It also removes range line number
validation for `nvim_parse_cmd` as it's not its job to validate the
command.
Changes the `range` value in `nvim_parse_cmd` into an array to describe
range information more concisely. Also makes `range` and `count` be
mutually exclusive by making count `-1` when command takes a range
instead of a count. Additionally corrects the behavior of `count` for
built-in commands by making the default count `0`.
Adds range, count and reg to the return values of nvim_parse_cmd. Also makes
line1 and line2 be -1 if the command does not take a range. Also moves
nvim_parse_cmd to vimscript.c because it fits better there.
This commit fixes#9358, where emitting multiple messages with 'echo' or
a single one with 'echom' or 'echoerr' would result in a press-enter
prompt that couldn't be dismissed by pressing enter.
This requires adapting a few tests to spawn a UI before testing whether
press-enter prompts are blocking.
It also fixes#11718, as when combined with #15910 it enables making
sure that neovim never blocks and emits messages on startup.
This includes a partial port of Vim patch 8.2.2569 and some changes to
nvim_eval_statusline() to allow a multibyte fillchar. Literally every
line of C code touched by that patch has been refactored in Nvim, and
that patch contains some irrelevant foldcolumn tests I'm not sure how to
port (as Nvim's foldcolumn behavior has diverged from Vim's).
"works with &opt" is flaky; now it always fails after #15999 for some reason.
This test was skipped in #10773 due to previous flakiness, but after the switch away from appveyor
the check no longer works. Just skip for any CI running the MSVC_32 job.
For anonymous scripts, defer the creation of script items until an attempt to access a script-local
variable is made. This dramatically reduces the number of script items created when using lots of
vim.cmd and nvim_exec especially.
This will mean <SID> usage fails until a script-local variable access is first made.
Based on #13143 (and #11507) with changes:
- Omit script_type_E. Use sn_name == NULL to determine anon items.
- Keep SID_STR. Used by anon :source for .lua files (no item).
- Show SID in get_scriptname output (:verbose set).
- Factor item creation into new_script_item.
- Leave sc_seq = 0 (anon scripts don't re-use the same item when re-sourced).
- Add tests for anon :source.
Co-authored-by: Vikram Pal <vikrampal659@gmail.com>
Co-authored-by: Justin M. Keyes <justinkz@gmail.com>
Adds the following API functions.
- nvim_buf_set_mark(buf, name, line, col)
* Set marks in a buffer.
- nvim_buf_del_mark(buf, name)
* Delete a mark that belongs to buffer.
- nvim_del_mark(name)
* Delete a global mark.
- nvim_get_mark(name)
* Get a global mark.
Tests:
- Adds test to all the new api functions, and adds more for the existing
nvim_buf_get_mark.
* Tests include failure cases.
Documentation:
- Adds documentation for all the new functions, and improves the
existing fucntion docs.
Do not copy a lot of lua strings (dict keys) to just strequal() them
Just compare them directly to a dedicated hash function.
feat(generators): HASHY McHASHFACE
Now remove the addition of "start/*" packages in 'packpath' as
explicit items in 'runtimepath'. This avoids 'runtimepath' from becoming
very long when using a lot of plugins as packages.
To get the effective search path as a list, use |nvim_list_runtime_paths()|
Problem:
Anonymous :source (no args) and nvim_exec() don't support Vimscript line continuations.
Solution:
Factor out the concat logic into concat_continued_line() and a
CONCAT_CONTINUED_LINES macro for simple concatenations where lines are
fetched individually.
Closes#14807
Iterating over PARAM_COUNT is wrong as PARAM_COUNT also counts the last
element of the options array, which has a NULL fullname in order to
signal the end of the array.
- The previous commit lost information in the tests. Instead, add some
more "normalization" substitutions in pcall_err(), so that the general
shape of the stacktrace is included in the asserted text.
- Eliminate contains(), it is redundant with matches()
- We already find ourselves renaming nvim_execute_lua in tests and
scripts, which suggests "exec" is the verb we actually want.
- Add "exec" verb to `:help dev-api`.
- Rename `meth_pcall`.
- Make `pcall_err` raise an error if the function does not fail.
- Add `vim.pesc()` to treat a string as literal where a Lua pattern is
expected.
- Always load files when cleaning up jumplist.
- For Shada: avoids writing duplicate entries, which happens when you read
from a shada file with duplicate entries (merging the jumplist while
writing sometimes produces duplicate entries, bug?) and then write right
away (i.e.: without any `:jumps`, `getjumplist()`, or any jump movement,
that is: nothing that calls `cleanup_jumplist` with `loadfiles == true`).
- For Context: avoids non-idempotent behavior for the same reason (i.e.:
first call to `shada_encode_jumps` does not remove duplicate entries).
- Do not set pcmark when dumping jumplist for Context.
- Retrieving current Context shouldn't add an entry to the jumplist
(which will be removed by a subsequent `cleanup_jumplist` anyway, i.e.:
tail entry matching current position), just act like `getjumplist` for
instance.
luassert uses 3 by default, which is often not enough.
Instead of documenting how to increase it, let's use a more fitting
(sane) default of 100 levels.
- All "chunks" in a paste-stream should form a single undo-block. Side
effect of 7a85792884 was to create an undo-block for each chunk.
- Also: remove old :redraw force logic, irrelevant after 7a85792884.
- Introduce TRY_WRAP() until we have an *architectural* solution.
- TODO: bfredl idea: prepare error-handling at "top level" (nv_event).
- nvim_paste(): Revert luaeval() hack (see parent commit).
- With TRY_WRAP() in nvim_put(), 'nomodifiable' error now correctly
"bubbles up".
- nvim_paste(): Marshal through luaeval() instead of nvim_execute_lua()
because the latter seems to hide some errors.
- Handle 'nomodifiable' in `nvim_put()` explicitly.
- Require explicit `false` from `vim.paste()` in order to "cancel",
otherwise assume true ("continue").
- Show error only once per "paste stream".
- Drain remaining chunks until phase=3.
- Lay groundwork for "cancel".
- Constrain semantics of "cancel" to mean "client must stop"; it is
unrelated to presence of error(s).
Fixes strange behavior where sometimes the buffer contents of a series
of paste chunks (vim._paste) would be out-of-order.
Now the tui_spec.lua screen-tests are much more reliable. But they still
sometimes fail because of off-by-one cursor (caused by "typeahead race"
resulting in wrong mode; fixed later in this patch-series).
