1
0
forked from VimPlug/jedi
Commit Graph

963 Commits

Author SHA1 Message Date
Shane Steinert-Threlkeld
bd5909e7b2 Find active conda environment and set it as default (if there is one) (#1440)
* add detection of conda environments

* changed get_default_environment to look for conda

* updated comment on get_default_environment to mention CONDA_PREFIX

* added myself to authors

* simple fix for mistaken conda paths
2019-12-07 01:04:41 +01:00
Dave Halter
700bd12122 Improve call signature detection by a lot
Fixes #1399
2019-12-04 23:55:26 +01:00
Dave Halter
8aee1e6213 Make sure that decorator signature completion is working, see #1433 2019-12-03 22:20:38 +01:00
Dave Halter
f46f00bc71 Avoid wrong random call signature completion, fixes #1433 2019-12-03 22:12:33 +01:00
Dave Halter
6ffeea7eea Make sure code_lines works on stubs, even if they are builtins 2019-12-01 19:10:08 +01:00
Dave Halter
b895924311 Merge pull request #1451 from pappasam/FIX_SHOW_SYSTEM_FOLDERS
Fix: no longer shows folders recursively to root
2019-12-01 00:12:51 +01:00
Dave Halter
86071dda54 Use a different sys path for import completions and import type inference
Fix tests of the #1451 pull request
2019-12-01 00:12:19 +01:00
Sam Roeca
1ba83414a5 Change search strategy for adding parent paths:
1. skip dirs with __init__.py
2. Stop immediately when above self._path
2019-11-30 10:14:28 -05:00
Sam Roeca
c2fd7b3104 Fix: upward search omits unnecessary paths
In the previous implementation, Jedi's traverse_parents function
traversed parent directories to the system root every time. This would
inadvertently add every folder to the system root every time. Obviously,
this is not the behavior desired for the import system.

This commit collects directories in an upward search until we:

1. Hit any directory without an __init__.py, AND
2. Are above self._path.
2019-11-29 21:12:12 -05:00
Sam Roeca
4bc4f167e9 Revert "Fix: no longer shows folders recursively to root"
This reverts commit 03b4177d3d.
2019-11-29 20:11:23 -05:00
Dave Halter
6e5e706288 Fix file name completions when file name is too long 2019-11-29 14:11:31 +01:00
Dave Halter
0e92be66db Fix an issue around completions in comments before strings 2019-11-29 13:44:12 +01:00
Sam Roeca
03b4177d3d Fix: no longer shows folders recursively to root
In the previous implementation, Jedi would's traverse_parents function
traversed parent directories to the system root every time. This would
inadvertently add every folder to the system root every time. Obviously,
this is not the behavior desired for the import system.

This pull request provides a new argument to the traverse_parents
function, "root", which represents the root parent for the search. This
argument defaults to None, thereby preserving the existing behavior of
the function.

I chose to duplicate some code for performance reasons. Since I'm trying
to avoid too much path manipulation magic, we do:

* a search to a valid specified root, OR
* a simple upward search until hitting the system root when there is no
valid root specified.
2019-11-28 23:04:08 -05:00
Dave Halter
4d332c32c0 Use create_name instead of duplicated logic 2019-09-05 00:04:24 +02:00
Dave Halter
47d6ae3da1 SimpleParamName -> AnonymousParamName 2019-09-04 01:20:44 +02:00
Dave Halter
a0cadd9375 Use Context.create_name instead of weird playing with params everywhere 2019-09-02 09:38:54 +02:00
Dave Halter
edb17b8e7c Refactor params and what execution contexts need 2019-09-01 14:14:42 +02:00
Dave Halter
4572503c9f Fix usages in context of the new parso parameter include_setitem=True 2019-08-28 22:56:16 +02:00
Dave Halter
c94bce315a Merge branch 'refactoring' 2019-08-24 14:38:45 +02:00
Dave Halter
ec6fa0c97c Differentiate between a public name and an internal string_name 2019-08-24 13:35:15 +02:00
Dave Halter
88cf198552 Avoid function executions if they are not necessary
This also means that annotations are prefered to docstring types
2019-08-24 12:23:33 +02:00
Dave Halter
6d361e03ac Avoid import recursions in other ways 2019-08-24 02:06:57 +02:00
Dave Halter
60a73f6bac Move get_global_filters to the context module 2019-08-23 21:19:17 +02:00
Dave Halter
7573e2033a Fix comprehension parent issues, fixes #1215 2019-08-23 14:34:16 +02:00
Dave Halter
86f4f7be45 Remove a private access 2019-08-23 11:55:49 +02:00
Dave Halter
9f3a2f93c4 Remove get_statement_of_position. It's not used anymore 2019-08-23 00:13:18 +02:00
Dave Halter
193ba47f50 Simplify get_user_context 2019-08-23 00:10:56 +02:00
Dave Halter
05fe29a156 Get rid of the node_is_value parameter 2019-08-23 00:02:39 +02:00
Dave Halter
faf6752ff8 Move create_context to a context 2019-08-22 22:47:26 +02:00
Dave Halter
eee6810576 Small cleanup of code 2019-08-22 17:20:07 +02:00
Dave Halter
02c96b37db Some more value -> context renames 2019-08-21 09:31:23 +02:00
Dave Halter
8a2b7f18cd Get all tests working except usage tests 2019-08-21 00:50:34 +02:00
Dave Halter
85f8f2a764 Fix os path resolving issues 2019-08-21 00:22:34 +02:00
Dave Halter
14fc5ed289 Fix more issues with where contexts are used and where values are used 2019-08-20 09:59:10 +02:00
Dave Halter
39b294e085 Fix some interpreter issues 2019-08-20 09:09:19 +02:00
Dave Halter
0c419a5094 Fix class tests 2019-08-17 23:52:52 +02:00
Dave Halter
2629ff55f3 Fix some array tests 2019-08-17 15:42:13 +02:00
Dave Halter
c6d2aa6da2 Some small improvements 2019-08-16 16:44:03 +02:00
Dave Halter
165639c1dd Start implementing the bulk of the context/value separation 2019-08-16 16:12:12 +02:00
Dave Halter
03920502c4 infer_state -> inference_state 2019-08-16 11:44:30 +02:00
Dave Halter
fffb39227e InferState -> InferenceState 2019-08-16 11:43:21 +02:00
Dave Halter
600272366f parent_value -> parent_context 2019-08-15 09:36:46 +02:00
Dave Halter
9986d8c9aa Context -> Value 2019-08-15 01:26:11 +02:00
Dave Halter
49f996867d NO_CONTEXTS -> NO_VALUES 2019-08-15 01:24:28 +02:00
Dave Halter
ad4f546aca context -> value 2019-08-15 01:23:06 +02:00
Dave Halter
9e23f4d67b Move base_context -> base_value 2019-08-15 00:41:02 +02:00
Dave Halter
a5dff65142 Evaluator -> InferState 2019-08-15 00:37:51 +02:00
Dave Halter
8157d119a7 eval_ -> infer_ 2019-08-15 00:20:01 +02:00
Dave Halter
3b4f292464 Move the evaluate package to inference 2019-08-15 00:14:26 +02:00
Matthias Bussonnier
f47211c129 Use scandir on py3.5+ for less disk access on filename completion
On Python 3.5+, we can make use of scandir that not only list the
content of the directory as an iterator but caches some infomations (for
example, `is_dir()`; this avoid extra stats call to the underlying
filesytem and can be – according to pep 471 –  2x to 20 time faster
especially on NFS filesystem where stats call is expensive.

From a quick this is the only place where scandir would make sens, as
most other places only require the name.

Fixes 1381
2019-08-12 17:56:29 -07:00