Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/quasilyte/astnorm

AST normalization experiment
https://github.com/quasilyte/astnorm

ast canonicalize go golang library normalization normalize source-code types

Last synced: 12 days ago
JSON representation

AST normalization experiment

Awesome Lists containing this project

README

        

[![Go Report Card](https://goreportcard.com/badge/github.com/Quasilyte/astnorm)](https://goreportcard.com/report/github.com/Quasilyte/astnorm)
[![GoDoc](https://godoc.org/github.com/Quasilyte/astnorm?status.svg)](https://godoc.org/github.com/Quasilyte/astnorm)
[![Build Status](https://travis-ci.org/Quasilyte/astnorm.svg?branch=master)](https://travis-ci.org/Quasilyte/astnorm)

![logo](/logo.jpg)

# astnorm

Go AST normalization experiment.

> THIS IS NOT A PROPER LIBRARY (yet?).

> DO NOT USE.

> It will probably be completely re-written before it becomes usable.

## Normalized code examples

1. Swap values.


Before
After




```go
tmp := xs[i]
xs[i] = ys[i]
ys[i] = tmp
```



```go
xs[i], ys[i] = ys[i], xs[i]
```

2. Remove elements that are equal to `toRemove+1`.


Before
After




```go
const toRemove = 10
var filtered []int
filtered = xs[0:0]
for i := int(0); i < len(xs); i++ {
x := xs[i]
if toRemove+1 != x {
filtered = append(filtered, x)
}
}
return (filtered)
```



```go
filtered := []int(nil)
filtered = xs[:0]
for _, x := range xs {
if x != 11 {
filtered = append(filtered, x)
}
}
return filtered
```

## Usage examples

* [cmd/go-normalize](/cmd/go-normalize): normalize given Go file
* [cmd/grepfunc](/cmd/grepfunc): turn Go code into a pattern for `gogrep` and run it

Potential workflow for code searching:

### 1. Code search

* Normalize the entire Go stdlib
* Then normalize your function
* Run `grepfunc` against normalized stdlib
* If function you implemented has implementation under the stdlib, you'll probably find it

Basically, instead of stdlib you can use any kind of Go corpus.

Another code search related tasks that can be simplified by `astnorm` are code similarity
evaluation and code duplication detection of any kind.

### 2. Static analysis

Suppose we have `badcode.go` file:

```go
package badpkg

func NotEqual(x1, x2 int) bool {
return (x1) != x1
}
```

There is an obvious mistake there, `x1` used twice, but because of extra parenthesis, linters may not detect this issue:

```bash
$ staticcheck badcode.go
# No output
```

Let's normalize the input first and then run `staticcheck`:

```bash
go-normalize badcode.go > normalized_badcode.go
staticcheck normalized_badcode.go
normalized_badcode.go:4:9: identical expressions on the left and right side of the '!=' operator (SA4000)
```

And we get the warning we deserve!
No changes into `staticcheck` or any other linter are required.

See also: [demo script](/example/demo.bash).