* Solved Problem with non-alphabetical .gitignore files
When .gitignore file in the user's project directory contained non-alphabetical
characters(Japanese, Korean, Chinese, etc), Nothing works and printed this
weird message in the console('cp949' is the encoding for Korean characters
in this case). It even blocks VSCode's formatting from working. This commit
solves the problem.
Traceback (most recent call last):
File "c:\users\username\anaconda3\envs\project-name\lib\runpy.py", line 193, in _run_module_as_main
"__main__", mod_spec)
File "c:\users\username\anaconda3\envs\project-name\lib\runpy.py", line 85, in _run_code
exec(code, run_globals)
File "C:\Users\username\anaconda3\envs\project-name\Scripts\black.exe\__main__.py", line 7, in <module>
File "c:\users\username\anaconda3\envs\project-name\lib\site-packages\black\__init__.py", line 1056, in patched_main
main()
File "c:\users\username\anaconda3\envs\project-name\lib\site-packages\click\core.py", line 829, in __call__
return self.main(*args, **kwargs)
File "c:\users\username\anaconda3\envs\project-name\lib\site-packages\click\core.py", line 782, in main
rv = self.invoke(ctx)
File "c:\users\username\anaconda3\envs\project-name\lib\site-packages\click\core.py", line 1066, in invoke
return ctx.invoke(self.callback, **ctx.params)
File "c:\users\username\anaconda3\envs\project-name\lib\site-packages\click\core.py", line 610, in invoke
return callback(*args, **kwargs)
File "c:\users\username\anaconda3\envs\project-name\lib\site-packages\click\decorators.py", line 21, in new_func
return f(get_current_context(), *args, **kwargs)
File "c:\users\username\anaconda3\envs\project-name\lib\site-packages\black\__init__.py", line 394, in main
stdin_filename=stdin_filename,
File "c:\users\username\anaconda3\envs\project-name\lib\site-packages\black\__init__.py", line 445, in get_sources
gitignore = get_gitignore(root)
File "c:\users\username\anaconda3\envs\project-name\lib\site-packages\black\files.py", line 122, in get_gitignore
lines = gf.readlines()
UnicodeDecodeError: 'cp949' codec can't decode byte 0xb0 in position 13: illegal multibyte sequence
* Made .gitignore File Reader Detect Its Encoding
* Revert "Made .gitignore File Reader Detect Its Encoding"
This reverts commit 6c3a7ea42b5b1e441cc0026c8205d1cee68c1bba.
* Revert "Solved Problem with non-alphabetical .gitignore files"
This reverts commit b0100b5d91c2f5db544a60f34aafab120f0aa458.
* Made .gitignore Reader Open the File with Auto Encoding Detecting
https://docs.python.org/3.8/library/tokenize.html#tokenize.open
* Revert "Made .gitignore Reader Open the File with Auto Encoding Detecting"
This reverts commit 50dd80422938649ccc8c7f43aac752f9f6481779.
* Made .gitignore Reader Use UTF-8
* Updated CHANGES.md for #2229
* Updated CHANGES.md for #2229
* Update CHANGES.md
* Update CHANGES.md
Co-authored-by: Jelle Zijlstra <jelle.zijlstra@gmail.com>
Co-authored-by: Łukasz Langa <lukasz@langa.pl>
Co-authored-by: Richard Si <63936253+ichard26@users.noreply.github.com>
Commit history before merge:
Black now respects .gitignore files in all levels, not only root/.gitignore file
(apply .gitignore rules like git does).
* Fix: typo
* Fix: respect .gitignore files in all levels.
* Add: CHANGELOG note.
* Fix: TypeError: unsupported operand type(s) for +: 'NoneType' and 'PathSpec'
* Update docs.
* Fix: no parent .gitignore
* Add a comment since the if expression is a bit hard to understand
* Update tests - conver no parent .gitignore case.
* Use main's Pipfile.lock instead
The original changes in Pipfile.lock are whitespace only. The changes
turned the JSON's file indentation from 4 to 2. Effectively this
happened: `json.dumps(json.loads(old_pipfile_lock), indent=2) + "\n"`.
Just using main's Pipfile.lock instead of undoing the changes because
1) I don't know how to do that easily and quickly, and 2) there's a
merge conflict.
Co-authored-by: Richard Si <63936253+ichard26@users.noreply.github.com>
* Merge remote-tracking branch 'upstream/main' into i1730 …
conflicts for days ay?
It appears sqlalchemy has recently reformatted their project with
Black 21.5b1.
Most of our dependencies have a lower bound and creating a test
environment with the oldest acceptable dependencies runs the full
Black test suite just fine. The only exception to this is aiohttp-cors.
It's unbounded and the oldest version 0.1.0 until 0.4.0 breaks the
test suite in such an old environment.
Failure with 0.1.0:
```
tests/test_blackd.py:10: in <module>
import blackd
testenv/lib/python3.8/site-packages/blackd/__init__.py:12: in <module>
import aiohttp_cors
testenv/lib/python3.8/site-packages/aiohttp_cors/__init__.py:29: in <module>
from .urldispatcher_router_adapter import UrlDistatcherRouterAdapter
testenv/lib/python3.8/site-packages/aiohttp_cors/urldispatcher_router_adapter.py:27: in <module>
class UrlDistatcherRouterAdapter(RouterAdapter):
testenv/lib/python3.8/site-packages/aiohttp_cors/urldispatcher_router_adapter.py:32: in UrlDistatcherRouterAdapter
def route_methods(self, route: web.Route):
E AttributeError: module 'aiohttp.web' has no attribute 'Route'
```
For 0.2.0:
```
tests/test_blackd.py:10: in <module>
import blackd
testenv/lib/python3.8/site-packages/blackd/__init__.py:12: in <module>
import aiohttp_cors
testenv/lib/python3.8/site-packages/aiohttp_cors/__init__.py:27: in <module>
from .cors_config import CorsConfig
testenv/lib/python3.8/site-packages/aiohttp_cors/cors_config.py:24: in <module>
from .urldispatcher_router_adapter import UrlDistatcherRouterAdapter
testenv/lib/python3.8/site-packages/aiohttp_cors/urldispatcher_router_adapter.py:27: in <module>
class UrlDistatcherRouterAdapter(AbstractRouterAdapter):
testenv/lib/python3.8/site-packages/aiohttp_cors/urldispatcher_router_adapter.py:32: in UrlDistatcherRouterAdapter
def route_methods(self, route: web.Route):
E AttributeError: module 'aiohttp.web' has no attribute 'Route'
```
For 0.3.0:
```
ERROR: Cannot install aiohttp-cors==0.3.0 and aiohttp==3.6.0 because these package versions have conflicting dependencies.
The conflict is caused by:
The user requested aiohttp==3.6.0
aiohttp-cors 0.3.0 depends on aiohttp<=0.20.2 and >=0.18.0
To fix this you could try to:
1. loosen the range of package versions you've specified
2. remove package versions to allow pip attempt to solve the dependency conflict
ERROR: ResolutionImpossible: for help visit https://pip.pypa.io/en/latest/user_guide/#fixing-conflicting-dependencies
```
* Move string-related utility to functions to strings.py, const.py
* Move Leaf/Node-related functionality to nodes.py
* Move comment-related functions to comments.py
* Move caching to cache.py and Mode/TargetVersion/Feature to mode.py
* Move some leftover functions to nodes.py, comments.py, strings.py
* Add missing files to source list for test runs
* Move line-related functionality into lines.py, brackets into brackets.py
* Move transformers to trans.py
* Move file handling, output, parsing, concurrency, debug, and report
* Move two more functions to nodes.py
* Add CHANGES
* Add numeric.py
* Add linegen.py
* More docstrings
* Include new files in tests
Co-authored-by: Jelle Zijlstra <jelle.zijlstra@gmail.com>
Closes#2164.
Changes behavior of how .gitignore is handled. With this change, the rules in .gitignore are only used as a fallback if no exclusion rule is explicitly passed on the command line or in pyproject.toml. Previously they were used regardless if explicit exclusion rules were specified, preventing any overriding of .gitignore rules.
Those that depend only on .gitignore for their exclusion rules will not be affected. Those that use both .gitignore and exclude will find that exclude will act more like actually specifying exclude and not just another extra-excludes. If the previous behavior was desired, they should move their rules from exclude to extra-excludes.
* Enable ` --experimental-string-processing` on all primer projects
- We want to make this default so need to test it more
- Fixed splat/star bug in extending black args for each project
* Disable sqlalchemy due to crash
Previously the RELAXED_DECORATOR detection would be falsely True on that
example. The problem was that an argument-less parentheses pair didn't
pass the `is_simple_decorator_trailer` check even it should. OTOH a
parentheses pair containing an argument or more passed as expected.
- Allow runs with no code diff output
- This is handy for reducing output to see which file is erroring
Test:
- Edit config for 'channels' to expect no changes and run with `--no-diff` and see no diff output
- It no longer crashes black so we should test on it's code
- Update django reason to name the file causing error
- Seems it has a syntax error on purpose
Such shebangs are only ever used if the file is executed directly, i.e.:
$ /usr/lib/python3.9/site-packages/black_primer/cli.py
But that doesn't work:
$ /usr/lib/python3.9/site-packages/black_primer/cli.py
bash: /usr/lib/python3.9/site-packages/black_primer/cli.py: Permission denied
The lib file even has: "lib is a library, funnily enough"
We're only fixing them so fuzzers don't yell at us when we break "valid"
code. I mean "valid" because some of the examples aren't even accepted by
Python.
Black would previously strip the parenthesis away from statements like this these ones:
assert (spam := 12 + 1)
return (cheese := 1 - 12)
Which happens to be invalid code. Now before making the parenthesis invisible, Black
checks if the assignment expression's parent is an assert stamtment, aborting if True.
Raise, yield, and await are already handled fine.
I added a bunch of test cases from the PEP defining asssignment expressions (PEP 572).
Optional trailing commas put by Black become magic trailing commas on another
pass of the tool. Since they are influencing formatting around optional
parentheses, on rare occasions the tool changes its mind in terms of putting
parentheses or not.
Ideally this would never be the case but sadly the decision to put optional
parentheses or not (which looks at pre-existing "magic" trailing commas) is
happening around the same time as the decision to put an optional trailing
comma. Untangling the two proved to be impractically difficult.
This shameful workaround uses the fact that the formatting instability
introduced by magic trailing commas is deterministic: if the optional trailing
comma becoming a pre-existing "magic" trailing comma changes formatting, the
second pass becomes stable since there is no variable factor anymore on pass 3,
4, and so on.
For most files, this will introduce no performance penalty since `--safe` is
already re-formatting everything twice to ensure formatting stability. We're
using this result and if all's good, the behavior is equivalent. If there is
a difference, we treat the second result as the binding one, and check its
sanity again.
Fixes#1738. Fixes#1812.
Previously, Black removed leading and trailing spaces in multiline docstrings but failed to remove them from one-line docstrings.
* Added support for top-level user configuration
At the user level, a TOML config can be specified in the following locations:
* Windows: ~\.black
* Unix-like: $XDG_CONFIG_HOME/black (~/.config/black fallback)
Instead of changing env vars for the entire black-primer process, they
are now changed only for the black subprocess, using a tmpdir.
Fixes: #1662
Work-around for https://bugs.python.org/issue2142
The test has to slightly mess with its input data, because the utility
functions default to ensuring the test data has a final newline, which
defeats the point of the test.
Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com>
* Add test case to illustrate the issue
* Accept carriage returns as valid separators while enumerating comments
Without this acceptance, escaped multi-line statments that use carriage returns will not be counted into the 'ignored_lines' variable since the emitted line values will end with a CR and not an escape character. That leads to comments associated with the line being incorrectly labeled with the STANDALONE_COMMENT type, affecting comment placement and line space management.
* Remove comment linking to ephemeral build log
To avoid hitting a mypy bug causes pre-commit to always fail on CPython
3.9. Even though it's still an outdated version, the bug effectively
blocks development on CPython 3.9 so that's why this commit exists
instead of waiting for cooperlees to finish his bump to 0.790 PR.
Also this fixes primer to ensure it always raises CalledProcessError
with an int error code. I stole the patch from cooperlees's mypy bump
PR.
It's funny how mypy 0.790 is already asked for in our
Pipfile.lock file, but oh well mypy is probably more commonly run
through pre-commit than standalone I guess.
Oh and if you're curious why the bug doesn't up on CPython 3.8 or lower:
there was some subscription AST changes in CPython 3.9.
* Provide a stdin-filename to allow stdin to respect exclude/force-exclude rules
This will allow automatic tools to enforce the project's
exclude/force-exclude rules even if they pass the file through stdin to
update its buffer.
This is a similar solution to --stdin-display-name in flake8.
* Update src/black/__init__.py
Co-authored-by: Richard Si <63936253+ichard26@users.noreply.github.com>
* --stdin-filename should only respect --exclude-filename
* Update README with the new --stdin-filename option
* Write some tests for the new stdin-filename functionality
* Apply suggestions from code review
Co-authored-by: Hugo van Kemenade <hugovk@users.noreply.github.com>
* Force stdin output when we asked for stdin even if the file exists
* Add an entry in the changelog regarding --stdin-filename
* Reduce disk reads if possible
Co-authored-by: Richard Si <63936253+ichard26@users.noreply.github.com>
* Check for is_stdin and p.is_file before checking for p.is_dir()
Co-authored-by: Richard Si <63936253+ichard26@users.noreply.github.com>
Co-authored-by: Hugo van Kemenade <hugovk@users.noreply.github.com>
The `fix_docstring` function expanded all tabs, which caused a
difference in the AST representation when those tabs were inline and not
leading. This changes the function to only expand leading tabs so inline
tabs are preserved.
Fixes#1601.
- If no aiohttp* deps exist nicely print a helpful message and exit
- There seems to be no nice way to optionally install the entry point, so lets make the entry point nicer
Test:
```
cooper-mbp1:black cooper$ /tmp/tb/bin/pip install .
cooper-mbp1:black cooper$ /tmp/tb/bin/blackd
aiohttp dependency is not installed: No module named 'aiohttp'. Please re-install black with the '[d]' extra install to obtain aiohttp_cors: `pip install black[d]`
cooper-mbp1:black cooper$ /tmp/tb/bin/pip install .[d]
...
Successfully installed aiohttp-3.6.3 aiohttp-cors-0.7.0 black
cooper-mbp1:black cooper$ /tmp/tb/bin/blackd
blackd version 20.8b2.dev31+gdd2f86a.d20201013 listening on localhost port 45484
```
Fixes#1688
* Repair colorama wrapping on non-Windows platforms
The wrap_stream_for_windows() function calls
colorama.initialise.wrap_stream() function to apply colorama's magic to
wrapper to the output stream. Except this wrapper is only applied on
Windows platforms that need it, otherwise the original stream is
returned as-is.
The colorama wrapped stream lacks a detach() method, so a no-op lambda
was being assigned to the wrapped stream.
The problem is that the no-op lambda was being assigned unconditionally
whether or not colorama actually returns a wrapped stream, thus
replacing the original TextIOWrapper's detach() method. Replacing the
detach() method with a no-op lambda is the root cause of the problem
observed in #1664.
The solution is to only assign the no-op detach() method if the stream
lacks its own detach() method.
Repairs #1664
Black used to erroneously remove all empty lines between non-function
code and decorators when formatting typing stubs. Now a single empty
line is enforced.
I chose for putting empty lines around decorated classes that have empty
bodies since removing empty lines around such classes would cause a
formatting issue that seems to be impossible to fix.
For example:
```
class A: ...
@some_decorator
class B: ...
class C: ...
class D: ...
@some_other_decorator
def foo(): -> None: ...
```
It is easy to enforce no empty lines between class A, B, and C.
Just return 0, 0 for a line that is a decorator and precedes an stub
class. Fortunately before this commit, empty lines after that class
would be removed already.
Now let's look at the empty line between class D and function foo. In
this case, there should be an empty line there since it's class code next
to function code. The problem is that when deciding to add X empty lines
before a decorator, you can't tell whether it's before a class or a
function. If the decorator is before a function, then an empty line
is needed, while no empty lines are needed when the decorator is
before a class.
So even though I personally prefer no empty lines around decorated
classes, I had to go the other way surrounding decorated classes with
empty lines.
Co-authored-by: Jelle Zijlstra <jelle.zijlstra@gmail.com>
- tox recently adopted Black
a7903508fa
- attrs already adopted Black but they updated to 20.08b1 + did a format pass and removed some trailing commas
f680c5b83e
* Fix incorrect space before colon in if/while stmts
Previously Black would format this code
```
if (foo := True):
print(foo)
```
as
```
if (foo := True) :
print(foo)
```
adding an incorrect space after the RPAR. Buggy code in the
normalize_invisible_parens function caused the colon to be wrapped in
invisible parentheses. The LPAR of that pair was then prefixed with a
single space at the request of the whitespace function.
This commit fixes the accidental skipping of a pre-condition check
which must return True before parenthesis normalization of a specific
child Leaf or Node can happen. The pre-condition check being skipped
was why the colon was wrapped in invisible parentheses.
* Add an entry in CHANGES.md
The quotes of multiline docstrings are now only normalized when string
normalization is off, instead of the string normalization setting being
ignored and the quotes being *always* normalized.
I had to make a new test case and data file since the current pair for
docstrings only worked when there is no formatting difference between the
formatting results with string normalization on and off. I needed to add
tests for when there *are* differences between the two. So I split
test_docstring's test code when string normalization is disabled into a
new test case along with a new data file.
On a second pass of Black on the same file, inserted trailing commas are now
pre-existing. Doesn't make sense to differentiate between the passes then.
Reformatted projects I have acceess to:
- aioexabgp
- bandersnatch
- flake8-bugbear
```
-- primer results 📊 --
13 / 16 succeeded (81.25%) ✅
0 / 16 FAILED (0.0%) 💩
- 3 projects disabled by config
- 0 projects skipped due to Python version
- 0 skipped due to long checkout
```
* Also re-enable pytest
```
-- primer results 📊 --
14 / 16 succeeded (87.5%) ✅
0 / 16 FAILED (0.0%) 💩
- 2 projects disabled by config
- 0 projects skipped due to Python version
- 0 skipped due to long checkout
real 2m26.207s
user 17m55.404s
sys 0m43.061s
```
This addresses a few crashers, namely:
* producing non-equivalent code due to mangling escaped newlines,
* invalid hugging quote characters in the docstring body to the docstring outer
triple quotes (causing a quadruple quote which is a syntax error),
* lack of handling for docstrings that start on the same line as the `def`, and
* invalid stripping of outer triple quotes when the docstring contained
a string prefix.
As a bonus, tests now also run when string normalization is disabled.
This required some hackery. Long story short, we need to reuse the ability to
omit rightmost bracket pairs (which glues them together and splits on something
else instead), for use with pre-existing trailing commas.
This form of user-controlled formatting is brittle so we have to be careful not
to cause a scenario where Black first formats code without trailing commas in
one way, and then looks at the same file with pre-existing trailing commas
(that it itself put on the previous run) and decides to format the code again.
One particular ugly edge case here is handling of optional parentheses. In
particular, the long-standing `line_length=1` hack got in the way of
pre-existing trailing commas and had to be removed. Instead, a more
intelligent but costly solution was put in place: a "second opinion" if the
formatting that omits optional parentheses ended up causing lines to be too
long. Again, for efficiency purposes, Black reuses Leaf objects from blib2to3
and modifies them in place, which was invalid for having two separate
formattings. Line cloning was used to mitigate this.
Fixes#1619
