Dave Halter
0392524dfc
Merge pull request #789 from blueyed/parser.utils.clear_cache-clear-self.__index
...
parser.utils.clear_cache: clear self.__index
2016-10-23 18:03:43 +02:00
Dave Halter
0475bb5fd0
First function execution that is working.
2016-10-23 03:02:57 +02:00
Dave Halter
75b67af000
Starting to improve function calls.
2016-10-22 21:02:15 +02:00
Dave Halter
2e6603cc2e
A lot of small improvements.
2016-10-22 17:40:42 +02:00
Dave Halter
4ccfbb4962
Use super().__getattribute__ instead of custom AttributeErrors in __getattr__.
2016-10-21 02:12:54 +02:00
Dave Halter
cad9ae8ab1
Start implemeting contexts with specialized TreeNameDefinitions to avoid gambling with the parser.
2016-10-20 19:36:44 +02:00
Dave Halter
3654de97b0
Better filter for arrays.
2016-10-16 14:57:08 +02:00
Dave Halter
0a0cb2a722
Fix generators.
2016-10-16 04:17:11 +02:00
Dave Halter
4ca3556c3b
Fix the fundamentally wrong cache.
2016-10-16 04:04:31 +02:00
Dave Halter
129c669bc0
Fix private variables in filters.
2016-10-15 19:12:46 +02:00
Dave Halter
5c0b2d7aae
Fixed a lot of class tests.
2016-10-14 18:09:29 +02:00
Daniel Hahler
a5480c054d
parser.utils.clear_cache: clear self.__index
...
This fixes a potential FileNotFoundError when clearing the cache
manually, using the method from
https://github.com/davidhalter/jedi-vim/pull/625 .
Traceback:
```
File "…/jedi/evaluate/imports.py", line 342, in _do_import
module = _load_module(self._evaluator, module_path, source, sys_path, parent_module)
File "…/jedi/evaluate/imports.py", line 457, in _load_module
cached = load_parser(path)
File "…/jedi/parser/utils.py", line 72, in load_parser
return ParserPickling.load_parser(path, p_time)
File "…/jedi/parser/utils.py", line 126, in load_parser
with open(self._get_hashed_path(path), 'rb') as f:
FileNotFoundError: [Errno 2] No such file or directory: '…/.cache/jedi/cpython-35/759d60e96c76f41ffd882d9b8d844899.pkl'
```
2016-10-14 14:25:18 +02:00
Daniel Hahler
f7f966805f
sys_path: prepend/prefer egg-link files
...
With `pip install -e` the generated .egg-link file gets preferred over
any normally installed distribution, and `pip uninstall` will first
remove the egg-link before the normal package.
2016-10-14 14:18:09 +02:00
Dave Halter
1752598353
A small class name lookup improvement.
2016-10-12 15:54:54 +02:00
Dave Halter
fce0eff18a
Get rid of all names_dicts in the completion api.
2016-10-12 03:40:24 +02:00
Dave Halter
482103e796
Replace names_dicts with filters in trailer completion.
2016-10-12 02:34:50 +02:00
Dave Halter
862e4a6176
Add filters for the sub module dicts and module attributes dicts.
2016-10-12 02:19:52 +02:00
Dave Halter
5f46b48433
Add a filter for global names.
2016-10-11 16:01:26 +02:00
Dave Halter
37ba971787
Trying to start implementing instance filters.
2016-10-06 16:35:53 +02:00
Dave Halter
2f1e9d634f
FunctionExecution improvement.
2016-10-02 19:54:03 +02:00
Dave Halter
c2873792eb
Filters for compiled objects and also FunctionExecution.
2016-10-02 15:36:24 +02:00
Dave Halter
249049b10c
Start using filters for name resolution.
2016-09-30 13:29:20 +02:00
Dave Halter
a96eec8058
Fix an issue with mixed objects.
2016-09-27 01:28:42 +02:00
Dave Halter
9b85d5517f
Fix more issues in the diff parser.
2016-09-27 00:29:11 +02:00
Dave Halter
09a5f27068
Add a test for wrong whitespace.
2016-09-26 23:56:37 +02:00
Dave Halter
c728148ece
Fix an issue with dividing suites and remove a lot of print statements.
2016-09-23 17:31:29 +02:00
Dave Halter
e371b670f5
Remove a comprehension hack in the parser that would have made the diff parsers world hell.
2016-09-22 18:26:09 +02:00
Dave Halter
c161e33119
Fix one more issue with the fast parser.
2016-09-21 20:36:54 +02:00
Dave Halter
6eb3b15e9b
Make a test a bit better testable (more flexible).
2016-09-21 18:13:18 +02:00
Dave Halter
37e3e79faa
Check in the diff tests that the graph is valid.
2016-09-19 05:41:59 +02:00
Dave Halter
ccc325616a
Temporarily fix an issue with list comprehensions.
2016-09-19 05:28:35 +02:00
Dave Halter
8aeeaec9c3
Remove some print statements
2016-09-18 20:43:52 +02:00
Dave Halter
b594a7d861
Merge pull request #776 from Alexey-T/patch-1
...
For Py3.5 embeddable, which misses pydoc_data module
2016-09-18 18:58:34 +02:00
Dave Halter
959f7b5e00
We don't need to reset the last failed start_pos anymore, because this is something that was necessary with the fucked up old parser.
2016-09-18 00:52:22 +02:00
Dave Halter
885cf62a12
Remove the position_modifier from the parser.
2016-09-18 00:50:31 +02:00
Uvview
e08209f35e
For Py3.5 embeddable, which misses pydoc_data module
2016-09-17 03:47:56 +04:00
Dave Halter
ed71d05ed7
Small test changes.
2016-09-15 09:37:35 +02:00
Dave Halter
74058fbf28
Finally passing all diff parser tests.
2016-09-15 01:26:28 +02:00
Dave Halter
8132055428
Fix an issue with parser endings and therefore adapt a few tests.
2016-09-14 17:23:49 +02:00
Dave Halter
47028c947a
Better debugging and solving a test with for stmts.
2016-09-13 20:34:02 +02:00
Dave Halter
f1a45ee4e6
Some error leaf handling.
2016-09-13 09:37:59 +02:00
Dave Halter
70e3719fb9
Small bug fixes.
2016-09-12 02:26:45 +02:00
Dave Halter
2eeafe23f8
Use differ for all diff tests.
2016-09-12 02:26:29 +02:00
Dave Halter
994e6615b1
Ifs in two directions.
2016-09-11 22:42:47 +02:00
Dave Halter
dfdda4a2f1
Copying an if (and other flows) is now working.
2016-09-11 21:51:44 +02:00
Dave Halter
c764976ef2
Merge branch 'remove_names_dicts' into diff
2016-09-11 13:24:11 +02:00
Dave Halter
7667cba17e
Remove old indent/dedent usages. Now they are not needed anymore.
2016-09-11 13:20:24 +02:00
Dave Halter
1226962922
Remove dedents from the parser tree. No need for them.
2016-09-11 13:03:29 +02:00
Dave Halter
cc5a2cd219
Small changes.
2016-09-09 17:38:07 +02:00
Dave Halter
5923765369
get_parsed_node should return the right thing.
2016-09-08 18:14:13 +02:00
