Problem: No support for stable Python 3 ABI
Solution: Support Python 3 stable ABI
Commits:
1) Support Python 3 stable ABI to allow mixed version interoperatbility
Vim currently supports embedding Python for use with plugins, and the
"dynamic" linking option allows the user to specify a locally installed
version of Python by setting `pythonthreedll`. However, one caveat is
that the Python 3 libs are not binary compatible across minor versions,
and mixing versions can potentially be dangerous (e.g. let's say Vim was
linked against the Python 3.10 SDK, but the user sets `pythonthreedll`
to a 3.11 lib). Usually, nothing bad happens, but in theory this could
lead to crashes, memory corruption, and other unpredictable behaviors.
It's also difficult for the user to tell something is wrong because Vim
has no way of reporting what Python 3 version Vim was linked with.
For Vim installed via a package manager, this usually isn't an issue
because all the dependencies would already be figured out. For prebuilt
Vim binaries like MacVim (my motivation for working on this), AppImage,
and Win32 installer this could potentially be an issue as usually a
single binary is distributed. This is more tricky when a new Python
version is released, as there's a chicken-and-egg issue with deciding
what Python version to build against and hard to keep in sync when a new
Python version just drops and we have a mix of users of different Python
versions, and a user just blindly upgrading to a new Python could lead to
bad interactions with Vim.
Python 3 does have a solution for this problem: stable ABI / limited API
(see https://docs.python.org/3/c-api/stable.html). The C SDK limits the
API to a set of functions that are promised to be stable across
versions. This pull request adds an ifdef config that allows us to turn
it on when building Vim. Vim binaries built with this option should be
safe to freely link with any Python 3 libraies without having the
constraint of having to use the same minor version.
Note: Python 2 has no such concept and this doesn't change how Python 2
integration works (not that there is going to be a new version of Python
2 that would cause compatibility issues in the future anyway).
---
Technical details:
======
The stable ABI can be accessed when we compile with the Python 3 limited
API (by defining `Py_LIMITED_API`). The Python 3 code (in `if_python3.c`
and `if_py_both.h`) would now handle this and switch to limited API
mode. Without it set, Vim will still use the full API as before so this
is an opt-in change.
The main difference is that `PyType_Object` is now an opaque struct that
we can't directly create "static types" out of, and we have to create
type objects as "heap types" instead. This is because the struct is not
stable and changes from version to version (e.g. 3.8 added a
`tp_vectorcall` field to it). I had to change all the types to be
allocated on the heap instead with just a pointer to them.
Other functions are also simply missing in limited API, or they are
introduced too late (e.g. `PyUnicode_AsUTF8AndSize` in 3.10) to it that
we need some other ways to do the same thing, so I had to abstract a few
things into macros, and sometimes re-implement functions like
`PyObject_NEW`.
One caveat is that in limited API, `OutputType` (used for replacing
`sys.stdout`) no longer inherits from `PyStdPrinter_Type` which I don't
think has any real issue other than minor differences in how they
convert to a string and missing a couple functions like `mode()` and
`fileno()`.
Also fixed an existing bug where `tp_basicsize` was set incorrectly for
`BufferObject`, `TabListObject, `WinListObject`.
Technically, there could be a small performance drop, there is a little
more indirection with accessing type objects, and some APIs like
`PyUnicode_AsUTF8AndSize` are missing, but in practice I didn't see any
difference, and any well-written Python plugin should try to avoid
excessing callbacks to the `vim` module in Python anyway.
I only tested limited API mode down to Python 3.7, which seemes to
compile and work fine. I haven't tried earlier Python versions.
2) Fix PyIter_Check on older Python vers / type##Ptr unused warning
For PyIter_Check, older versions exposed them as either macros (used in
full API), or a function (for use in limited API). A previous change
exposed PyIter_Check to the dynamic build because Python just moved it
to function-only in 3.10 anyway. Because of that, just make sure we
always grab the function in dynamic builds in earlier versions since
that's what Python eventually did anyway.
3) Move Py_LIMITED_API define to configure script
Can now use --with-python-stable-abi flag to customize what stable ABI
version to target. Can also use an env var to do so as well.
4) Show +python/dyn-stable in :version, and allow has() feature query
Not sure if the "/dyn-stable" suffix would break things, or whether we
should do it another way. Or just don't show it in version and rely on
has() feature checking.
5) Documentation first draft. Still need to implement v:python3_version
6) Fix PyIter_Check build breaks when compiling against Python 3.8
7) Add CI coverage stable ABI on Linux/Windows / make configurable on Windows
This adds configurable options for Windows make files (both MinGW and
MSVC). CI will also now exercise both traditional full API and stable
ABI for Linux and Windows in the matrix for coverage.
Also added a "dynamic" option to Linux matrix as a drive-by change to
make other scripting languages like Ruby / Perl testable under both
static and dynamic builds.
8) Fix inaccuracy in Windows docs
Python's own docs are confusing but you don't actually want to use
`python3.dll` for the dynamic linkage.
9) Add generated autoconf file
10) Add v:python3_version support
This variable indicates the version of Python3 that Vim was built
against (PY_VERSION_HEX), and will be useful to check whether the Python
library you are loading in dynamically actually fits it. When built with
stable ABI, it will be the limited ABI version instead
(`Py_LIMITED_API`), which indicates the minimum version of Python 3 the
user should have, rather than the exact match. When stable ABI is used,
we won't be exposing PY_VERSION_HEX in this var because it just doesn't
seem necessary to do so (the whole point of stable ABI is the promise
that it will work across versions), and I don't want to confuse the user
with too many variables.
Also, cleaned up some documentation, and added help tags.
11) Fix Python 3.7 compat issues
Fix a couple issues when using limited API < 3.8
- Crash on exit: In Python 3.7, if a heap-allocated type is destroyed
before all instances are, it would cause a crash later. This happens
when we destroyed `OptionsType` before calling `Py_Finalize` when
using the limited API. To make it worse, later versions changed the
semantics and now each instance has a strong reference to its own type
and the recommendation has changed to have each instance de-ref its
own type and have its type in GC traversal. To avoid dealing with
these cross-version variations, we just don't free the heap type. They
are static types in non-limited-API anyway and are designed to last
through the entirety of the app, and we also don't restart the Python
runtime and therefore do not need it to have absolutely 0 leaks.
See:
- https://docs.python.org/3/whatsnew/3.8.html#changes-in-the-c-api
- https://docs.python.org/3/whatsnew/3.9.html#changes-in-the-c-api
- PyIter_Check: This function is not provided in limited APIs older than
3.8. Previously I was trying to mock it out using manual
PyType_GetSlot() but it was brittle and also does not actually work
properly for static types (it will generate a Python error). Just
return false. It does mean using limited API < 3.8 is not recommended
as you lose the functionality to handle iterators, but from playing
with plugins I couldn't find it to be an issue.
- Fix loading of PyIter_Check so it will be done when limited API < 3.8.
Otherwise loading a 3.7 Python lib will fail even if limited API was
specified to use it.
12) Make sure to only load `PyUnicode_AsUTF8AndSize` in needed in limited API
We don't use this function unless limited API >= 3.10, but we were
loading it regardless. Usually it's ok in Unix-like systems where Python
just has a single lib that we load from, but in Windows where there is a
separate python3.dll this would not work as the symbol would not have
been exposed in this more limited DLL file. This makes it much clearer
under what condition is this function needed.
closes: #12032
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Yee Cheng Chin <ychin.git@gmail.com>
Problem: no support for custom cmdline completion
Solution: Add new vimscript functions
Add the following two functions:
- getcmdcompltype() returns custom and customlist functions
- getcompletion() supports both custom and customlist
closes: #12228
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Shougo Matsushita <Shougo.Matsu@gmail.com>
Problem: '.-' no allowed in highlight group names
Solution: Allow dot and hyphen characters in highlight group names
Allow dots and hyphens in group names. There does not seem
to be any reason for these to be disallowed.
closes: #12807
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Gregory Anders <greg@gpanders.com>
Co-authored-by: Sean Dewar <seandewar@users.noreply.github.com>
Problem: Not able to get the virtual text property
Solution: Make prop_list() return virtual text and alignment
closes: #12860
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Yegappan Lakshmanan <yegappan@yahoo.com>
Problem: g<kEnd> behaves different from g<end>
Solution: Make g<kEnd> behave like g<End>
closes: #12861
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: zeertzjq <zeertzjq@outlook.com>
Problem: vim9 class problem with new() constructor
Solution: Don't allow a return type for the new() class constructor.
closes: #12863closes: #12040
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Yegappan Lakshmanan <yegappan@yahoo.com>
Problem: can't move to last non-blank char
Solution: Make g<end> behave like that
Make it possible to move to last non-blank char on a line
We can distinguish between g0 and g^ to move to the very first character
and the first non-blank char.
And while we can move to the last screen char, we cannot go to the last
non-blank screen char.
Since I think g$ is the more widely used and known movement command (and
g<end> is synonymous to it) change the behaviour of g<end> to move to
last non-screen char instead and don't have this be the same command as
the g$ command anymore.
If you want to keep the old behaviour, you can use:
```
nnoremap g<end> g$
```
Add a test to verify the behaviour.
closes: #12593
Signed-off-by: Christian Brabandt <cb@256bit.org>
Problem: missing winid argument for virtcol()
Solution: Add a {winid} argument to virtcol()
Other functions col(), charcol() and virtcol2col() support a {winid}
argument, so it makes sense for virtcol() to also support than.
Also add test for virtcol2col() with 'showbreak' and {winid}.
closes: #12633
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: zeertzjq <zeertzjq@outlook.com>
Problem: minor problems with the teapot()
Solution: remove the null check, update documentation
Update the builtin-function-list entry. (It implicitly returns 0, but such
functions usually use "none")
Also, tv_get_string_strict() can not return NULL, so remove that check
closes: #12647
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Sean Dewar <seandewar@users.noreply.github.com>
Problem: virtcol2col returns last byte of a multi-byte char
Solution: Make it return the first byte for a multi-byte char
closes: #12786closes: #12799
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Yegappan Lakshmanan <yegappan@yahoo.com>
Problem: sidescrolloff and scrolloff options work slightly
different than other global-local options
Solution: Make it behave consistent for all global-local options
It was noticed, that sidescrolloff and scrolloff options behave
differently in comparison to other global-local window options like
'listchars'
So make those two behave like other global-local options. Also add some
extra documentation for a few special local-window options.
Add a few tests to make sure all global-local window options behave
similar
closes: #12956closes: #12643
Signed-off-by: Christian Brabandt <cb@256bit.org>
Problem: Cannot use positional arguments for printf()
Solution: Support positional arguments in string formatting
closes: #12140
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Christ van Willegen <cvwillegen@gmail.com>
* Dedicate upcoming Vim 9.1 to Bram
Also replace in a few more places Brams email address and mention new
maintainers.
* Remove Bram from any Maintainer role
* runtime: Align Header
* it's mailing list not mailinglist
Problem: cannot store custom data in quickfix list
Solution: add `user_data` field for the quickfix list
closes: #11818
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: Tom Praschan <13141438+tom-anders@users.noreply.github.com>
Problem: undotree() only works for the current buffer
Solution: Add an optional "buffer number" parameter to undotree(). If
omitted, use the current buffer for backwards compatibility.
closes: #4001closes: #12292
Signed-off-by: Christian Brabandt <cb@256bit.org>
Co-authored-by: zeertzjq <zeertzjq@outlook.com>
Co-authored-by: Devin J. Pohly <djpohly@gmail.com>
Using xxd(1) to filter and edit binary files causes the input files
to have dual nature, so to speak, which effectively makes restoring
the cursor position broken. Fix that by ignoring the "xxd" file type
in the code that restores the cursor position.
Interactive rebasing in git causes files to be edited in vim, which,
similarly to commit messages, are rarely the same as the last one
edited. Thus, also add "gitrebase" to the list of file types for
which the cursor position isn't restored.
While there, refactor the code a bit to possibly save a few CPU cycles
and to keep the line lengths in check, and use the long form of the
commands and variables, to make the code slightly more consistent and
more understandable to newcomers.
Update the relevant comments in the code and the associated parts of
the documentation, to keep them in sync with the updated code.
Remove some redundant trailing whitespace as well, as spotted.
Save and restore the view position before and after saving the buffer,
respectively, to keep the current view of the xxd(1)'s hex dump
unchanged after doing ":w", which previously caused the window to
scroll back to the very beginning of the buffer. I believe it's
needless to say how annoying and counterproductive that was.
Get rid of the "Press ENTER or type command to continue" message, which
was previously displayed after opening larger binary files. The use
of "silent" and "redraw" commands is tailored specifically to avoid
screen flickering, e.g. when doing ":w", which is caused by the buffer
being filtered by an external command.
Increase the number of octets per line, produced by xxd(1), from the
default value of 16 to 32. This puts bigger chunks of the hex dump
on the screen and makes the whole thing much more usable.
While there, reformat the code to make it more readable, and use the
long form of the commands and variables to make the code slightly more
consistent and more understandable to newcomers.
This is a collection of various PRs from github that all require a minor
patch number:
1) https://github.com/vim/vim/pull/12612
Do not conflate dictionary key with end of block
2) https://github.com/vim/vim/pull/12729:
When saving and restoring 'undolevels', the constructs `&undolevels` and
`:set undolevels` are problematic.
The construct `&undolevels` reads an unpredictable value; it will be the
local option value (if one has been set), or the global option value
(otherwise), making it unsuitable for saving a value for later
restoration.
Similarly, if a local option value has been set for 'undolevels',
temporarily modifying the option via `:set undolevels` changes the local
value as well as the global value, requiring extra work to restore both
values.
Saving and restoring the option value in one step via the construct
`:let &undolevels = &undolevels` appears to make no changes to the
'undolevels' option, but if a local option has been set to a different
value than the global option, it has the unintended effect of changing
the global 'undolevels' value to the local value.
Update the documentation to explain these issues and recommend explicit
use of global and local option values when saving and restoring. Update
some unit tests to use `g:undolevels`.
3) https://github.com/vim/vim/pull/12702:
Problem: Pip requirements files are not recognized.
Solution: Add a pattern to match pip requirements files.
4) https://github.com/vim/vim/pull/12688:
Add indent file and tests for ABB Rapid
5) https://github.com/vim/vim/pull/12668:
Use Lua 5.1 numeric escapes in tests and add to CI
Only Lua 5.2+ and LuaJIT understand hexadecimal escapes in strings. Lua
5.1 only supports decimal escapes:
> A character in a string can also be specified by its numerical value
> using the escape sequence \ddd, where ddd is a sequence of up to three
> decimal digits. (Note that if a numerical escape is to be followed by a
> digit, it must be expressed using exactly three digits.) Strings in Lua
> can contain any 8-bit value, including embedded zeros, which can be
> specified as '\0'.
To make sure this works with Lua 5.4 and Lua 5.1 change the Vim CI to
run with Lua 5.1 as well as Lua 5.4
6) https://github.com/vim/vim/pull/12631:
Add hurl filetype detection
7) https://github.com/vim/vim/pull/12573:
Problem: Files for haskell persistent library are not recognized
Solution: Add pattern persistentmodels for haskell persistent library
closes: #12612closes: #12729closes: #12702closes: #12688closes: #12668closes: #12631closes: #12573
Co-authored-by: lacygoill <lacygoill@lacygoill.me>
Co-authored-by: Michael Henry <drmikehenry@drmikehenry.com>
Co-authored-by: ObserverOfTime <chronobserver@disroot.org>
Co-authored-by: KnoP-01 <knosowski@graeffrobotics.de>
Co-authored-by: James McCoy <jamessan@jamessan.com>
Co-authored-by: Jacob Pfeifer <jacob@pfeifer.dev>
Co-authored-by: Borys Lykah <lykahb@fastmail.com>
Problem: charidx() and utf16idx() result is not consistent with byteidx().
Solution: When the index is equal to the length of the text return the
lenght of the text instead of -1. (Yegappan Lakshmanan,
closes#12503)
Problem: Some commands for opening a file don't use 'switchbuf'.
Solution: Use 'switchbuf' for more commands. (Yegappan Lakshmanan,
closes#12383, closes#12381)
Problem: No error when calling remote_startserver() with an empty string.
Solution: Give an error for an empty string. (Hirohito Higashi,
closes#12327)
Problem: no functions for converting from/to UTF-16 index.
Solution: Add UTF-16 flag to existing funtions and add strutf16len() and
utf16idx(). (Yegappan Lakshmanan, closes#12216)
Problem: Decrypting with libsodium may fail if the library changes.
Solution: Add parameters used to the encrypted file header. (Christian
Brabandt, closes#12279)
Problem: Highlight for popupmenu kind and extra cannot be set.
Solution: Add PmenuKind, PmenuKindSel, PmenuExtra and PmenuExtraSel
highlight groups and use them. (Gianmaria Bajo, closes#12114)
