Commit Graph

204 Commits

Author SHA1 Message Date
Strahinja Val Markovic
196228217f Filename completer now uses threads
This was done by introducing a new ThreadedCompleter class that descends from
Completer. Both JediCompleter and FilenameCompleter descend from
ThreadedCompleter.
2013-04-22 22:30:10 -07:00
Strahinja Val Markovic
8d20637295 User can now pick the root for relative paths 2013-04-22 22:30:10 -07:00
Strahinja Val Markovic
aa9127e3dc Rewriting the code from the pull request
This implements the filename completer and introduces integration with
UltiSnips. The user will now see snippets in the completion menu. After
selecting a snippet, the user should invoke the UltiSnips trigger key
(which should be changed from the default of TAB) to trigger the snippet
expansion.

Fixes #77, Fixes #36
2013-04-22 22:26:35 -07:00
Stanislav Golovanov
bb5839dd74 Add general filename completer & other stuff 2013-04-22 22:24:38 -07:00
Strahinja Val Markovic
14b2220f01 Supporting YcmCorePreload logic
Now the user has the option of writing custom logic before ycm_core.so is
loaded. This can be used to dynamically change the location of where ycm_core.so
is loaded by prepending paths to sys.path.

Very, very few people will need this feature, but I'm one of them so there.
2013-04-22 10:31:16 -07:00
Strahinja Val Markovic
aea154e829 Refactoring ycm_extra_conf handling into module 2013-04-22 10:31:16 -07:00
Stanislav Golovanov
5d37c40fea Add general UltiSnips snippet completer 2013-04-21 01:28:37 +04:00
Stanislav Golovanov
48cda3bb8f Add general completers support 2013-04-21 01:28:37 +04:00
Strahinja Val Markovic
93ed6f7db7 Fix bug with removing flag after "-c"
-c does not take an argument. Why did I think it did?
2013-04-19 10:31:55 -07:00
Thomas Refis
dad08df71b checking if a key is in a dict is done with 'key in dict' not 'hasattr(dict, key)' 2013-04-13 14:29:57 +02:00
Jacek Lach
1b4eb6122c Make caching optional in the omnicompleter 2013-04-11 10:22:16 +01:00
Strahinja Val Markovic
77fbd80c89 Refactoring out some vim.eval calls
s:completion_start_column should not be evaled in Python code. That was a
stopgap measure that ended up lasting far longer than intended.
2013-04-09 19:32:48 -07:00
Strahinja Val Markovic
27d73e0d51 Removing some dead code 2013-04-09 18:43:29 -07:00
Strahinja Val Markovic
085df7e8af Fix traceback print on None for |location|
This happens when the user invokes a GoTo* command on a file with no compilation
flags.
2013-04-08 11:15:27 -07:00
Strahinja Val Markovic
a75c92bb6a Using event.is_set to be compatible with Python2.6
Fixes #227
2013-04-03 19:14:46 -07:00
Strahinja Val Markovic
8e4a2e24f8 Ignoring mappings in our vim 'normal' invocation 2013-03-31 21:01:38 -07:00
Strahinja Val Markovic
1f094e50d0 GoToDefinition/Declaration commands for C-family
These are accessible through the :YcmCompleter command. The docs have more
information.
2013-03-31 20:38:29 -07:00
Strahinja Val Markovic
b4837c81fe Handling exceptions in jedi_completer thread 2013-03-24 18:52:50 -07:00
Strahinja Val Markovic
edbcb3abfe More cleanup for correctness and simplicity
The jedi completer now won't hang on <C-Space>
2013-03-24 18:42:31 -07:00
Strahinja Val Markovic
9845d81fb0 More style fixes 2013-03-24 15:22:31 -07:00
Strahinja Val Markovic
6fb784e61b Making the python thread not hang Vim on exit
As the Pytho docs say, "The entire Python program exits when no alive non-daemon
threads are left." Basically, the daemon threads are safely killed.
2013-03-24 15:04:00 -07:00
Strahinja Val Markovic
1c78971acd More style fixes 2013-03-24 14:57:25 -07:00
Strahinja Val Markovic
8f2715c1a4 Posting an error message when jedi import fails 2013-03-24 14:52:11 -07:00
Strahinja Val Markovic
4d95e32a1e Python code style fixes 2013-03-24 14:41:22 -07:00
Strahinja Val Markovic
0954ff31d9 Bug fixes for the new python semantic completer 2013-03-24 14:38:55 -07:00
Stephen Sugden
eed0434e39 Refactor jedi-completer to use a worker thread 2013-03-23 19:24:26 -07:00
Stephen Sugden
da46be7044 Add jedi-based completion.
This first version only uses the Jedi completion engine after a ".", similar to
how the ClangCompleter works. It is also entirely synchronous and blocks for
quite a while the first time it is called.
2013-03-23 16:21:49 -07:00
Strahinja Val Markovic
74021144c0 Making the omni completer more robust
- There was a bug in not calling int() on the result of the first call to the
omnifunc.
- We need to be more resilient to badly written omnifuncs and check that the
result of the second call is a list or a dict with a list, as the vim docs say
the omnifunc should return on the second call.

One of the two fixes probably fixes #198 but since I can't repro the error, I
can't be sure.
2013-03-22 10:26:18 -07:00
Strahinja Val Markovic
d0a51fbf2c More info for conf file load dialog message
The user is now informed of options that can turn off this dialog.
2013-03-19 12:59:17 -07:00
Strahinja Val Markovic
4cb9f17c5f Cache deletion is now async 2013-03-16 12:10:24 -07:00
Strahinja Val Markovic
c391bdcc62 Reducing RAM consumption by deleting unused caches
When the user deletes a buffer, we can delete the clang caches for that file.
Fixes #184.
2013-03-16 10:41:47 -07:00
Strahinja Val Markovic
ba6b40e485 Refactoring the ClangAvailableForBuffer method 2013-03-03 10:48:34 -08:00
Zeh Rizzatti
c6aefaef86 Select correct triggers for multiple filetypes
The base class completer Inner chose the first filetype available and
would use the triggers for it.

The triggers are now chosen considering the first for the current buffer
that is supported by the current completer. If there is no intersection,
it fallsback to considering the first filetype for the buffer.
2013-03-03 01:52:56 -04:00
Zeh Rizzatti
ac1e04fc14 Check for a native completer with multiple filetypes
GetFiletypeCompleter would always return a omnicompleter for the first
filetype in case there was no native completer, and the lookup would
stop.

This changes that behaviour to get all possible completers and tries to
find a native one among them. If no native completer is found, it
returns the omnicompleter for the first filetypes, as it used to.
2013-03-02 22:14:22 -04:00
Zeh Rizzatti
1acd3e84c7 Support for multile filetypes in clang_completer
clang_completer would check if the raw value of '&ft' was one of
supported filetypes for the completer.

Vim allows for multiple filetypes with a '.' separator. A file with
ft=qt.cpp, for example, would not be supported by clang_completer even
though it was a cpp file.

This patch changes that behaviour.
2013-03-02 22:02:29 -04:00
Strahinja Val Markovic
980a1c2ffc Minor post-PR changes 2013-02-28 09:54:22 -08:00
Strahinja Val Markovic
aafcd4264d Merge branch 'globlist' of github.com:kljohann/YouCompleteMe into kljohann-globlist
Conflicts:
	python/completers/cpp/flags.py
2013-02-28 09:51:38 -08:00
Val Markovic
c24d272df6 Merge pull request #155 from kljohann/reload
Implement completer-specific commands
2013-02-28 09:49:11 -08:00
Johann Klähn
1d566bad80 Implement white-/blacklisting 2013-02-28 18:15:09 +01:00
Johann Klähn
3d305f9c74 Implement completer-specific commands
This provides a framework for completer-writers to create
completer-specific commands. I have in mind to use this for the clang
completer to force reloading of a flags module via `:YcmCompleter reload`.
2013-02-28 11:32:07 +01:00
Strahinja Val Markovic
263b405ab4 Very minor post-pull style-related changes. 2013-02-26 18:28:58 -08:00
Strahinja Val Markovic
95b9eeee18 Merge branch 'master' of github.com:kljohann/YouCompleteMe into kljohann-master 2013-02-26 18:21:46 -08:00
Johann Klähn
e9cce29761 Ask before loading .ycm_extra_conf.py files
To prevent the execution of malicious code the new default is
to ask the user before a `.ycm_extra_conf.py` file is loaded.
This can be disabled using the option `g:ycm_confirm_extra_conf`.

This commit introduces a helper class `FlagsModules` that keeps track of
and caches the currently loaded modules. To introduce further criteria
for a module look at `FlagsModules.ShouldLoad`.

Also `:YcmDebugInfo` now lists the file that was used to determine
the current set of flags.

`Flags.ModuleForFile` could be used in a user-facing command that
opens the `.ycm_extra_conf.py` corresponding to the current file.
A second command could then force a reloding of this module via
`Flags.ReloadModule`.
2013-02-26 11:16:26 +01:00
Johann Klähn
4b3e0a1895 Add code to ask user for confirmation in vimsupport 2013-02-25 10:50:44 +01:00
Johann Klähn
70a2a722fe Add GetBoolValue helper in vimsupport 2013-02-25 10:50:39 +01:00
Strahinja Val Markovic
c819c9f31e Typo fix 2013-02-23 13:08:56 -08:00
Strahinja Val Markovic
98ff82ece5 Removing falsy items from omnifunc returned items
Depending on the omnifunc the user has set, it could return empty strings etc in
the list of items.

Fixes #146, fixes #147
2013-02-23 09:55:13 -08:00
Strahinja Val Markovic
c1700c7d19 Handle omnifunc returning dict with 'words' 2013-02-23 09:35:07 -08:00
Strahinja Val Markovic
101d949a88 Splitting echo text on \n and echoing lines 2013-02-21 10:50:38 -08:00
Strahinja Val Markovic
e60928d721 Updating error message on missing ycm_extra_conf 2013-02-20 13:16:18 -08:00
bear
66289d7522 let g:ycm_global_ycm_extra_conf can use "~" 2013-02-18 15:10:13 +08:00
Val Markovic
9e3aa21a77 Merge pull request #119 from olajep/let-local-conf-override-global
Let local YCM config file override global config file
2013-02-17 13:07:38 -08:00
Ola Jeppsson
3d1a86c382 Let local YCM config file override global config file
Currently, when VIM opens a source file, YCM always defaults to
'g:global_ycm_extra_conf_file' if it exists.

This commit changes YCM's behaviour so that it first tries to find the config
file in the source file's folder (or any of its parents folder), before
falling back to 'g:global_ycm_extra_conf_file'.
2013-02-17 19:06:48 +01:00
Strahinja Val Markovic
e743076e14 Option to collect identifiers n comments/strings
Fixes #98.
2013-02-16 14:00:46 -08:00
Strahinja Val Markovic
7833cc1cf1 More doc updates for Completer API 2013-02-16 13:33:19 -08:00
Strahinja Val Markovic
1cd6f4d05e Some extra docs for the Completer API 2013-02-16 13:26:56 -08:00
Stephen Sugden
7006033d3e typo fix 2013-02-13 21:18:51 -08:00
Strahinja Val Markovic
4000da089d Only triggering omni_completer when omnifunc set 2013-02-13 10:27:40 -08:00
Strahinja Val Markovic
ad8345aa35 Now checking ycm_core version for compatibility 2013-02-12 20:54:27 -08:00
Strahinja Val Markovic
21dac46ecc Docs for the Completer API 2013-02-12 20:01:22 -08:00
Strahinja Val Markovic
452f7d1fec Better triggering of semantic completion
Now there's a nice user-configurable setting for when YCM should trigger
semantic completion. This is very useful for the new omni_completer that uses
data coming from Vim's omnicomplete system.
2013-02-11 21:46:06 -08:00
Strahinja Val Markovic
786e6182ff Initial version of omnifunc-based omni_completer
Still a work in progress (needs better triggering and bug fixes)
2013-02-10 20:03:56 -08:00
Strahinja Val Markovic
6e27176ebd Shortening some function names
They could be shorter and still readable
2013-02-10 20:03:56 -08:00
Strahinja Val Markovic
b0c0a12df8 Removing unused file 2013-02-10 20:03:56 -08:00
Strahinja Val Markovic
6be3cd37ed Making sure filename exists before using it 2013-02-06 18:46:57 -08:00
Strahinja Val Markovic
6f10d7aad7 YcmDiags now doesn't hang when no flags available
Fixes #46
2013-02-05 21:22:50 -08:00
Strahinja Val Markovic
a1c89f5108 Fix rare traceback on <leader>d. Fixes #43 2013-02-05 20:12:43 -08:00
Strahinja Val Markovic
ace95db4f0 Emit warning about missing extra_conf file once
Otherwise we spam the user, which is annoying. Fixes #41.
2013-02-05 19:03:38 -08:00
Strahinja Val Markovic
f0e98fa21f Reordering some imports 2013-02-04 23:12:22 -08:00
Strahinja Val Markovic
cc14981774 Renaming utils.py to ycm_utils.py 2013-02-04 15:33:18 -08:00
Strahinja Val Markovic
f671fe72b2 Adding our paths to sys.path start, not end
Otherwise things can conflict and we could end up loading the wrong module.
2013-01-31 17:32:39 -08:00
Strahinja Val Markovic
c67658bbce Supporting multiple filetypes set for current file
Vim allows setting the filetype string to something like "cpp.c", which means
that the file is both cpp and c (nonsense, but allowed). We need to support such
filetype strings.
2013-01-31 16:19:56 -08:00
Strahinja Val Markovic
8452914046 Adding a command to force recompilation and diags 2013-01-30 14:46:58 -08:00
Strahinja Val Markovic
79f834f97e Now possible to specify a global ycm_extra_conf 2013-01-30 13:23:57 -08:00
Strahinja Val Markovic
424736e395 Making sure that 'flags' is a list 2013-01-28 12:56:45 -08:00
Strahinja Val Markovic
d3874ebb65 Code style fix 2013-01-28 10:05:13 -08:00
Strahinja Val Markovic
7e929b7831 clang_completer debug info now shows file flags 2013-01-28 10:00:15 -08:00
Strahinja Val Markovic
a6a1e7c949 Clear error message on ycm_core ImportError 2013-01-26 17:29:18 -08:00
Strahinja Val Markovic
e8b60fd537 Adding the YcmDebugInfo command 2013-01-26 11:45:14 -08:00
Strahinja Val Markovic
6579a1d05d Trivial refactoring 2013-01-24 17:14:10 -08:00
Strahinja Val Markovic
19637cadda Renaming ycm_clang_options to ycm_extra_conf 2013-01-24 09:59:21 -08:00
Strahinja Val Markovic
90fe31f98e Async querying of CompilationDatabase supported 2013-01-23 17:23:51 -08:00
Strahinja Val Markovic
970b2fb56c Kiling a memory leak... I hate C interfaces 2013-01-22 17:29:05 -08:00
Strahinja Val Markovic
6ba2dd8caa Printing an error when no ycm_clang_options.py file 2013-01-22 17:13:14 -08:00
Strahinja Val Markovic
482de517ba Better support for CompilationDatabase
We now use the compilation working directory for a file that is specified in the
CompilationDatabase. We don't actually change the working directory of the
process, even temporarily (that would be annoying to users); we munge the flags
coming from the database so that all the relative paths in them are resolved to
absolute ones.
2013-01-22 16:30:49 -08:00
Strahinja Val Markovic
00db8fd8b1 compilation_database.json now fully supported 2013-01-19 18:53:30 -08:00
Strahinja Val Markovic
f9a516704b Now explicitly adding the clang includes to flags 2013-01-17 17:08:28 -08:00
Strahinja Val Markovic
0922864eda Moving clang includes to clang_includes folder 2013-01-17 17:08:11 -08:00
Strahinja Val Markovic
5ac3d40691 No error when no clang support and cpp file opened
Previously, when the user opened a cpp/c/objc/objcpp file and clang support was
not compiled in, there would be an error message printed in vim.
2013-01-13 20:56:10 -08:00
Strahinja Val Markovic
fa21622655 Updating local clang includes to 3.2
We have to have these beneath the folder where ycm_core.so is placed so that we
get fast completions. If these files are not here, then clang fails to
precompile a file preamble and completions are slow.
2013-01-13 12:49:49 -08:00
Strahinja Val Markovic
c9adb9e86e Comment update in vimsupport.py 2012-09-24 19:20:33 -07:00
Strahinja Val Markovic
04991d874d Moving a comment closer to the related code 2012-09-05 21:49:51 -07:00
Strahinja Val Markovic
0c17c49a66 Per-filetype turning off of filetype completion
So things like the ClangCompleter can be turned off fully and the user can rely
on identifier completion only.
2012-08-15 21:29:43 -07:00
Strahinja Val Markovic
dc2f52edbf Adding a max_diagnostics_to_display user option 2012-08-15 20:58:21 -07:00
Strahinja Val Markovic
daef17feb4 Support for showing extra info for diagstics
This was intended to show the full clang output for a given diagnostic,
including notes. But it appears that libclang does not provide this
functionality...
2012-08-15 19:39:03 -07:00
Strahinja Val Markovic
f500ecaffb Fix for hang on insert enter after "foo.bar"
If the user had code like "foo.bar" and then entered insert mode after the 'r'
in "bar", YCM would cause vim to hang.

The problem happened because a sorting task was created that would try to sort
on the latest clang result but none would be created because a clang task was
not created in this occasion. clang_data_ready_ would remain false and would
never be set to true, thus causing an infinite loop in SortingThreadMain since
the thread would forever wait on the mutex.

This was rectified with better handling of the clang results cache. Now the
cache is a full class and it also stores the line & column number of the
location for which the results were computed. Better logic is in place for the
cache invalidation.
2012-08-12 19:01:39 -07:00
Strahinja Val Markovic
f99e89f812 Fix for "lclose not allowed here" error on open
The problem was caused by a race condition of all things. ClangCompleter would
set possibly_completions_ready when starting the first parse pass for the file
and then would try to extract diagnostics for the file before the diagnostics
were done. Technically this was not a problem because only an empty diagnostics
vector would be returned, but this triggered Syntastic because hey, we have some
diagnostics to show (even though we don't).

And then Syntastic would try to close the location list window during startup
when this operation is not available. Technically it's Syntastic's fault, but a
more principled way to check for done diagnostics is to return and use a future
for file parsing operations and this solution also works around the Syntastic
issue.
2012-08-12 14:36:36 -07:00
Strahinja Val Markovic
fef702eef0 Reversed the logic for the no-results-found var 2012-08-12 14:07:38 -07:00
Strahinja Val Markovic
d793919969 Removing old debugging statement 2012-08-11 21:19:06 -07:00
Strahinja Val Markovic
f88c9feb4f Refactored the clang completer; many bugs fixed
This change should fix the random hangs and segfaults when using the clang
completer. Also, assertion errors printed to the console on vim exit should go
away too, same thing with segfaults on vim exit. These "on exit" errors were
caused by not cleanly shutting down the background threads; both the identifier
completer and the clang one now join the threads on destruction. This results in
a clean shutdown.

The new clang completer architecture now uses only one clang thread (again)
instead of a completion and parsing thread. Since the parsing task needs to wait
on the completion task if it was started first (and vice-versa) there's no point
to using two threads. The desired "simplicity" of using two threads for these
two tasks actually created needless complexity (and bugs). Sigh. Such is life.

A TranslationUnit abstraction was also created and this in turn also reduces the
complexity of the clang completer.

The clang completer now also has some (very) basic tests.
2012-08-11 19:37:08 -07:00