Fix common misspellings in text files. It's designed primarily for checking misspelled words in source code (backslash escapes are skipped), but it can be used with other files as well. It does not check for word membership in a complete dictionary, but instead looks for a set of common misspellings. Therefore it should catch errors like "adn", but it will not catch "adnasdfasdf". This also means it shouldn't generate false-positives when you use a niche term it doesn't know about.
Python 3.9 or above.
You can use pip to install codespell with e.g.:
pip install codespellBelow are some simple usage examples to demonstrate how the tool works.
For exhaustive usage information, please check the output of codespell -h.
Run codespell in all files of the current directory:
codespellRun codespell in specific files or directories (specified via their names or glob patterns):
codespell some_file some_dir/ *.extSome noteworthy flags:
codespell -w, --write-changesThe -w flag will actually implement the changes recommended by codespell. Running without the -w flag is the same as doing a dry run. It is recommended to run this with the -i or --interactive flag.
codespell -I FILE, --ignore-words=FILEThe -I flag can be used for a list of certain words to allow that are in the codespell dictionaries. The format of the file is one word per line. Invoke using: codespell -I path/to/file.txt to execute codespell referencing said list of allowed words. See Ignoring Words for more details.
codespell -L word1,word2,word3,word4The -L flag can be used to allow certain words that are comma-separated placed immediately after it. See Ignoring Words for more details.
codespell -x FILE, --exclude-file=FILEIgnore whole lines that match those in FILE. The lines in FILE should match the to-be-excluded lines exactly.
codespell -S, --skip=Comma-separated list of files to skip. It accepts globs as well. Examples:
- to skip .eps & .txt files, invoke
codespell --skip="*.eps,*.txt" - to skip directories, invoke
codespell --skip="./src/3rd-Party,./src/Test"
Useful commands:
codespell -d -q 3 --skip="*.po,*.ts,./src/3rdParty,./src/Test"List all typos found except translation files and some directories. Display them without terminal colors and with a quiet level of 3.
codespell -i 3 -wRun interactive mode level 3 and write changes to file.
We ship a collection of dictionaries that are an improved version of the one available on Wikipedia after applying them in projects like Linux Kernel, EFL, oFono among others. You can provide your own version of the dictionary, but patches for new/different entries are very welcome.
Want to know if a word you're proposing exists in codespell already? It is possible to test a word against the current set dictionaries that exist in codespell_lib/data/dictionary*.txt via:
echo "word" | codespell -
echo "1stword,2ndword" | codespell -You can select the optional dictionaries with the --builtin option.
When ignoring false positives, note that spelling errors are case-insensitive but words to ignore are case-sensitive. For example, the dictionary entry wrod will also match the typo Wrod, but to ignore it you must pass wrod (to match the case of the dictionary entry).
The words to ignore can be passed in two ways:
-I: A file with a word per line to ignore:codespell -I FILE, --ignore-words=FILE
-L: A comma separated list of words to ignore on the command line:codespell -L word1,word2,word3,word4
Some situation might require ignoring a specific word in a specific location. This can be achieved by adding a comment in the source code.
You can either ignore a single word or a list of words. The comment should be in the format of codespell:ignore <words>.
Words should be separated by a comma.
ignore specific word:
def wrod() # codespell:ignore wrod pass
ignore multiple words:
def wrod(wrods) # codespell:ignore pass
Command line options can also be specified in a config file.
When running codespell, it will check in the current directory for an
INI file
named setup.cfg or .codespellrc (or a file specified via --config),
containing an entry named [codespell]. Each command line argument can
be specified in this file (without the preceding dashes), for example:
[codespell]
skip = *.po,*.ts,./src/3rdParty,./src/Test
count =
quiet-level = 3Python's
configparser
module defines the exact format of INI config files. For example,
comments are possible using ; or # as the first character.
Codespell will also check in the current directory for a pyproject.toml
file (or a file specified via --toml), and the [tool.codespell]
entry will be used. For versions of Python prior to 3.11, this requires
the tomli package. For example, here is the TOML equivalent of the
previous config file:
[tool.codespell]
skip = '*.po,*.ts,./src/3rdParty,./src/Test'
count = true
quiet-level = 3The above INI and TOML files are equivalent to running:
codespell --skip "*.po,*.ts,./src/3rdParty,./src/Test" --count --quiet-level 3If several config files are present, they are read in the following order:
pyproject.toml(only if thetomlilibrary is available for Python < 3.11)setup.cfg.codespellrc- any additional file supplied via
--config
If a codespell configuration is supplied in several of these files, the configuration from the most recently read file overwrites previously specified configurations. Any options specified in the command line will override options from the config files.
Values in a config file entry cannot start with a - character, so if
you need to do this, structure your entries like this:
[codespell]
dictionary = mydict,-
ignore-words-list = bar,-fooinstead of these invalid entries:
[codespell]
dictionary = -,mydict
ignore-words-list = -foo,barcodespell also works with pre-commit, using
- repo: https://github.com/codespell-project/codespell
rev: v2.4.1
hooks:
- id: codespellIf one configures codespell using the pyproject.toml file instead use:
- repo: https://github.com/codespell-project/codespell
rev: v2.4.1
hooks:
- id: codespell
additional_dependencies:
- tomliThe format of the dictionaries was influenced by the one they originally came from, i.e. from Wikipedia. The difference is how multiple options are treated and that the last argument is an optional reason why a certain entry could not be applied directly, but should instead be manually inspected. E.g.:
Simple entry: one wrong word / one suggestion:
calulated->calculated
Entry with more than one suggested fix:
fiel->feel, field, file, phial,
Note the last comma! You need to use it, otherwise the last suggestion will be discarded (see below for why). When there is more than one suggestion, an automatic fix is not possible and the best we can do is to give the user the file and line where the error occurred as well as the suggestions.
Entry with one word, but with automatic fix disabled:
clas->class, disabled because of name clash in c++
Note that there isn't a comma at the end of the line. The last argument is treated as the reason why a suggestion cannot be automatically applied.
There can also be multiple suggestions but any automatic fix will again be disabled:
clas->class, clash, disabled because of name clash in c++
As suggested in the Python Packaging User Guide, ensure pip, setuptools, and wheel are up to date before installing from source. Specifically you will need recent versions of setuptools and setuptools_scm:
pip install --upgrade pip setuptools setuptools_scm wheelYou can install required dependencies for development by running the following within a checkout of the codespell source:
pip install -e ".[dev]"To run tests against the codebase run:
make checkIf you have a suggested typo that you'd like to see merged please follow these steps:
Make sure you read the instructions mentioned in the
Dictionary formatsection above to submit correctly formatted entries.Choose the correct dictionary file to add your typo to. See codespell --help for explanations of the different dictionaries.
Sort the dictionaries. This is done by invoking (in the top level directory of
codespell/):make check-dictionaries
If the make script finds that you need to sort a dictionary, please then run:
make sort-dictionaries
Only after this process is complete do we recommend you submit the PR.
Important Notes:
- If the dictionaries are submitted without being pre-sorted the PR will fail via our various CI tools.
- Not all PRs will be merged. This is pending on the discretion of the devs, maintainers, and the community.
To stay current with codespell developments it is possible to build codespell from GitHub via:
pip install --upgrade git+https://github.com/codespell-project/codespell.gitImportant Notes:
Sometimes installing via
pipwill complain about permissions. If this is the case then run with:pip install --user --upgrade git+https://github.com/codespell-project/codespell.git
It has been reported that after installing from
pip, codespell can't be located. Please check the $PATH variable to see if~/.local/binis present. If it isn't then add it to your path.If you decide to install via
pipthen be sure to remove any previously installed versions of codespell (via your platform's preferred app manager).
In the scenario where the user prefers not to follow the development version of codespell yet still opts to benefit from the frequently updated dictionary files, we recommend running a simple set of commands to achieve this:
wget https://raw.githubusercontent.com/codespell-project/codespell/main/codespell_lib/data/dictionary.txt
codespell -D dictionary.txtThe above simply downloads the latest dictionary.txt file and then by utilizing the -D flag allows the user to specify the freshly downloaded dictionary.txt as the custom dictionary instead of the default one.
- You can also do the same thing for the other dictionaries listed here:
- https://github.com/codespell-project/codespell/tree/main/codespell_lib/data
The Python script codespell with its library codespell_lib is available
with the following terms:
(tl;dr: GPL v2)
Copyright (C) 2010-2011 Lucas De Marchi <lucas.de.marchi@gmail.com>
Copyright (C) 2011 ProFUSION embedded systems
This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; version 2 of the License.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program; if not, see <https://www.gnu.org/licenses/old-licenses/gpl-2.0.html>.
dictionary.txt and the other dictionary_*.txt files are derivative works of English Wikipedia and are released under the Creative Commons Attribution-Share-Alike License 3.0.