UNMAINTAINED REPOSITORY (issues are closed, no more development)
cleanjs is a tool written in python that checks javascript source files and returns a report about various aspects related to clean code practices (kind of like check style).
The way it works is by parsing the source file then running a series checks on it, outputting info, warning and errors when needed.
The tool is focused on code quality. The following things (and others) are checked:
- file length
- function length
- variable and function name formatting and meaning
- comments
- complexity
- syntax error
- formatting
- ...
The first way to use cleanjs is online: http://cleanjscode.appspot.com/
If you are interested in running it locally on your computer and integrate it into some build system, then read on:
cleanjs is written in python so, first, make sure you have python 2.7 installed and running (seems to work with 2.6 but mostly untested).
Get cleanjs by downloading one of the tags.
Once you've retrieved the code, you can run the tool like this:
python cleanjs_cmdline.py path/to/my_source_file.js path/to/output/file.html
This will review the file path/to/my_source_file.js and create an HTML report file.
You can check the testscript/reports/*.html files to see examples of reports
Optionally, if you are using a text editor like Sublime Text or any other editor able to run build commands on files, you can use the cleanjs_sublime.py entry point instead.
cleanjs can also be integrated into Sublime Text 2 as a plugin.
If you have Sublime Text 2 with Package Control, just launch package control, select "Add Repository", enter https://github.com/captainbrosset/cleanjs and then launch "install package" and select cleanjs from the list.
The default key biding is ctrl+shift+c and for now only outputs messages to a panel and highlights the corresponding lines in the text.
A tooltip/line gutter API is terribly missing in Sublime Text 2 at the moment, vote for it here.
Integrating cleanjs into an existing python program is fairly easy. Open cleanjs_cmdline.py to see how a file is parsed, how reviewers are called and how the output is constructed. You'll see it's easy to create your own reviewers and reporters.
The easiest and most interesting way to contribute is to create more reviewers.
Check the /reviewers/ folder to see how they are made.
Each reviewer must define a Reviewer class and implement a method: review(self, file_data, message_bag)
- file_data is of type
parsers.fileparser.FileData - message_bag is of type
reviewers.messagebag.MessageBag
Check the issues in GitHub to see what is currently planned for development.