Devreplay

Devreplay is static analysis tool based on your own programming pattern.

• Visual Studio Code
• Other Editor Support (Language Server)
• GitHub Application
• Pattern generator

How to use

1. Install on local

$ npm install devreplay
# or
$ yarn global add 

2. Put your own programming pattern(devreplay.json) on the project like bellow

[
  {
    "before": [
      "tmp = $1",
      "$1 = $2",
      "$2 = tmp"
    ],
    "after": [
      "$1, $2 = $2, $1"
    ]
  }
]

3. Run the devreplay

$ devreplay yourfile.py
W:yourfile.py:15:$3 = $1$1 = $2$2 = $3 should be $2, $1 = $1, $2

or get fixed code

$ devreplay --fix yourfile.py

The target source code file will be

- tmp = a
- a = b
- b = a
+ a, b = b, a

• Step up: Make the pattern message and severity. Also after can be more abstract

[
  {
    "before": [
      "$3 = $1",
      "$1 = $2",
      "$2 = $3"
    ],
    "after": [
      "$1, $2 = $2, $1"
    ],
    "author": "Yuki Ueda",
    "message": "Value exchanging can be one line",
    "severity": "Information"
  }
]

• Recommend: Also you can generate pattern file automatically by following command on your git repository

devreplay --init

• severity means how this pattern is important

  • E: Error
  • W: Warning
  • I: Information
  • H: Hint

• Run devreplay again

$ devreplay yourfile.py
I:yourfile.py:15:Value exchanging can be one line by Yuki Ueda

Also, you can use default rules by extends some rules such as

[
  {
    "extends": ["typescript", "react", "vscode"]
  }
]

Make patterns by using Regular Expression

{
  "before": [
    "([a-z]+)-([a-z]+)"
  ],
  "after": [
      "$1 $2"
  ],
  "regex": true
}

That will fix

- print("hello-world")
+ print("hello world")

Default rule languages

• c
• cpp
• dart
• cobol
• java
• javaScript
• typeScript
• python
• ruby
• go
• php

Defaulr rule frameworks

• Angular
• chainer2pytouch
• tensorflow
• rails
• vue

Contribution Link

Thanks

This package is made based on

• tslint
• vscode-python
• vscode-textmate

DevReplay is supported by 2019 Exploratory IT Human Resources Project The MITOU Program, JSPS KAKENHI Grant Numbers JP17H00731, JP15H02683, JP18H03221, and JP18KT0013.

License

MIT © 2019 Yuki Ueda [email protected] (ikuyadeu.github.io)