summaryrefslogtreecommitdiff
path: root/checkers/variables.py
Commit message (Collapse)AuthorAgeFilesLines
* Add test cases for good cases, check the targets before the inference.unpacking2cpopa2013-07-251-3/+4
|
* Prefer the symbolic name.cpopa2013-07-241-1/+1
|
* Infer the value before using it, accept astroid.List as well.cpopa2013-07-241-3/+8
|
* Prefer the symbolic message.cpopa2013-07-211-1/+1
|
* Add support for unbalanced tuple unpacking check.cpopa2013-07-211-0/+26
|
* [variables checker] drop check_messages that trigger false positive when ↵Sylvain Th?nault2013-07-081-3/+0
| | | | other messages from this checker are disabled (eg running --errors-only)
* Resolve conflict and take feed-back into account for PR #34Anthony Truchet2013-06-191-1/+6
|\
| * Add @check_messages(...) annotations to help not to spend time on unneeded ↵Anthony Truchet2013-06-181-1/+7
|/ | | | | | | | checks. We only add such annotations is reasonably simple cases. Some visit methods are quite involved and build a checker's state ; such methods are skipped in the first phase.
* Do not warn about redefinitions of dummy variables.Torsten Marek2013-06-181-1/+3
|
* Do not treat all variables starting with _ as dummy variables, only _.Torsten Marek2013-06-181-1/+1
|
* astng has been renamed astroidDavid Douard2013-06-171-30/+30
|
* use .fromlineno instead of .lineno (which is not reliable)Sylvain Th?nault2013-02-251-2/+2
|
* Emit a warning if __all__ contains non-string objects.tmarek2013-01-081-1/+5
| | | | Closes #112728
* Fixed a couple of bugs in the __all__ handling and added a newstabletmarek2013-01-081-8/+16
| | | | | | warning about non-string objects in __all__. Closes #112698
* lint fixesSylvain Th?nault2012-09-191-3/+3
|
* Closes #104572: symbolic warning names in output (by Martin Pool)Martin Pool2012-09-191-0/+16
| | | | triggered whatever the format using a command line option
* drop useless attribute from variables checkerSylvain Th?nault2012-09-191-5/+0
|
* use .iter* version of dict methods when possibleSylvain Th?nault2012-09-191-2/+2
|
* Add test and code for handling __all__ with pylint. Closes #4685.FELD Boris2012-09-191-2/+16
|
* Closes #93591: Correctly emit W0623 on multiple assignment of unpackable ↵tmarek2012-06-071-5/+7
| | | | | | | | | | | | | exceptions eg for code like try: ... except AnyException as (here, there): ... Instead of warning about redefining tuple, recurse into the tuple and check all names.
* Detect assignments to a loop variable in the else branch of a for statement.Torsten Marek2011-11-171-1/+8
| | | | Closes: #9188,#4024
* Fix for #81078: Warn if names in exception handlers clobber overwrite ↵Sylvain Th?nault2011-10-241-3/+12
| | | | existing names
* closes #74747: don't crash when lookup up a special attribute in class scopeSylvain Th?nault2011-09-071-2/+9
| | | | (patch by google)
* closes #74087: handle case where inference of a module return YESSylvain Th?nault2011-09-071-1/+3
| | | | | | | | | | | | | This avoid some cases of "TypeError: '_Yes' object does not support indexing" when conditionally assigning a nonexisting module to intermediate modules in a dotted name. This is only a cosmetic fix to avoid the crash, the actual fix probably needs a lot of work on the inference module to support conditional assignment and inspection of several places of assignment, rather than just the latest one. Probably not worth it, either. patch by Google
* closes #70495: absolute imports fail depending on module path (patch by ↵Sylvain Th?nault2011-07-081-1/+2
| | | | Jacek Konieczny)
* performance: use 'active_msgs' attribute to disable function calls or the likeEmile Anclin2010-12-131-2/+10
|
* pylint itself: remove last pylint error messageEmile Anclin2010-12-131-3/+4
|
* cleanup: remove useless KeyboardInterrupt catch and have a print forEmile Anclin2010-12-091-4/+5
| | | | unhandled exception
* messages: decorate visit and leave methods with check_messagesEmile Anclin2010-12-091-3/+5
| | | | refactor checker/bases.py while adding 'check_messages'
* remove code that's out commented since rev 0Emile Anclin2010-12-061-25/+0
|
* 223 : run python2.x compatible fixersEmile Anclin2010-11-151-1/+1
|
* cleanup; remove a useless "else" indentationEmile Anclin2010-11-041-35/+34
|
* py3k: list comprehension is now a scoped nodeEmile Anclin2010-11-041-0/+18
|
* help / man: more explicit on dummy variable regexEmile Anclin2010-10-251-2/+2
|
* update VariablesChecker to take into account the nested scopes of dictionary ↵Daniel Harding2010-10-091-3/+25
| | | | and set comprehensions
* remove useless lgc.compat importsEmile Anclin2010-10-061-1/+0
|
* copyright update : date up 2010Emile Anclin2010-03-231-1/+1
|
* we may have name node with fromlineno to None for astng built from ↵Sylvain Th?nault2010-03-231-3/+5
| | | | C-compiled modules
* mergeEmile Anclin2010-03-221-2/+1
|\
| * modify astng imports after astng 'rebuild' refactoringEmile Anclin2010-03-221-2/+1
| |
* | test and fix crash on global w/ 'virtual' nodes such as __file__, ↵Sylvain Th?nault2010-03-221-0/+7
|/ | | | __module__, etc.
* applied Colin Moris patch closing #9263: no W0613 for __init__ (method does ↵Sylvain Th?nault2010-03-041-1/+1
| | | | not use all of its arguments)
* fix #20991: class scope definitions ignored in a genexprAdrien Di Mascio2010-02-251-2/+6
| | | | | | | | When checking for variable existence in a method definition, pylint simply skips the class scope in the scope consumer chain. This technique works fine with classic methods but doesn't work when analyzing genexprs defined at class level : for genexprs, we do want the class scope to be considered.
* fix #18862: E0601 false positive with lambda functionsSylvain Th?nault2009-11-251-1/+2
|
* include Dotan Barak spell fixes patchSylvain Th?nault2009-11-231-9/+9
|
* d-t-wSylvain Th?nault2009-08-271-26/+26
|
* do not print Redefining builtin for additional builtinssylvain thenault2009-04-081-8/+7
|
* add test for #5719, fixed in astngsylvain thenault2009-03-241-1/+1
|
* only skip E0203 when catched by an AttributeError or parentsylvain thenault2009-03-241-3/+0
|
* only skip E0602 when catched by a NameError or parentsylvain thenault2009-03-241-2/+2
|