______________________________________________________________________

Branch	Build	Unittest	Release	License
main

Code-Text parser is a custom tree-sitter's grammar parser for extract raw source code into class and function level. We support 10 common programming languages:

• Python
• Java
• JavaScript
• PHP
• Ruby
• Rust
• C
• C++
• C#
• Go

Installation

codetext package require python 3.7 or above and tree-sitter. Setup environment and install dependencies manually from source:

git https://github.com/FSoft-AI4Code/CodeText-parser.git; cd CodeText-parser
pip install -r requirement.txt
pip install -e .

Or install via pypi package:

pip install codetext

Getting started

codetext CLI Usage

codetext [options] [PATH or FILE] ...

For example extract any python file in src/ folder:

codetext src/ --language Python

If you want to store extracted class and function, use flag --json and give a path to destination file:

codetext src/ --language Python --output_file ./python_report.json --json

Options

positional arguments:
  paths                 list of the filename/paths.

optional arguments:
  -h, --help            show this help message and exit
  --version             show program's version number and exit
  -l LANGUAGE, --language LANGUAGE
                        Target the programming languages you want to analyze.
  -o OUTPUT_FILE, --output_file OUTPUT_FILE
                        Output file (e.g report.json).
  --json                Generate json output as a transform of the default
                        output
  --verbose             Print progress bar

Example

File circle_linkedlist.py analyzed:
==================================================
Number of class    : 1
Number of function : 2
--------------------------------------------------

Class summary:
+-----+---------+-------------+
|   # | Class   | Arguments   |
+=====+=========+=============+
|   0 | Node    |             |
+-----+---------+-------------+

Class analyse: Node
+-----+---------------+-------------+--------+---------------+
| #   | Method name   | Paramters   | Type   | Return type   |
+=====+===============+=============+========+===============+
| 0   | __init__      | self        |        |               |
|     |               | data        |        |               |
+-----+---------------+-------------+--------+---------------+

Function analyse:
+-----+-----------------+-------------+--------+---------------+
| #   | Function name   | Paramters   | Type   | Return type   |
+=====+=================+=============+========+===============+
| 0   | push            | head_ref    |        | Node          |
|     |                 | data        | Any    | Node          |
| 1   | countNodes      | head        | Node   |               |
+-----+-----------------+-------------+--------+---------------+

Using codetext as Python module

Build your language

codetext need tree-sitter language file (i.e .so file) to work properly. You can manually compile language (see more) or automatically build use our pre-defined function (the <language>.so will saved in a folder name /tree-sitter/):

from codetext.utils import build_language

language = 'rust'
build_language(language)

# INFO:utils:Not found tree-sitter-rust, attempt clone from github
# Cloning into 'tree-sitter-rust'...
# remote: Enumerating objects: 2835, done. ...
# INFO:utils:Attempt to build Tree-sitter Language for rust and store in .../tree-sitter/rust.so

Using Language Parser

Each programming language we supported are correspond to a custome language_parser. (e.g Python is PythonParser()). language_parser take input as raw source code and use breadth-first search to traveser through all syntax node. The class, method or stand-alone function will then be collected:

from codetext.utils import parse_code

raw_code = """
    /**
    * Sum of 2 number
    * @param a int number
    * @param b int number
    */
    double sum2num(int a, int b) {
        return a + b;
    } 
"""

# Auto parse code into tree-sitter.Tree
root = parse_code(raw_code, 'cpp')
root_node = root.root_node

Get all function nodes inside a specific node:

from codetext.utils.parser import CppParser

function_list = CppParser.get_function_list(root_node)
print(function_list)

# [<Node type=function_definition, start_point=(6, 0), end_point=(8, 1)>]

Get function metadata (e.g. function's name, parameters, (optional) return type)

function = function_list[0]

metadata = CppParser.get_function_metadata(function, raw_code)

# {'identifier': 'sum2num', 'parameters': {'a': 'int', 'b': 'int'}, 'type': 'double'}

Get docstring (documentation) of a function

docstring = CppParser.get_docstring(function, code_sample)

# ['Sum of 2 number \n@param a int number \n@param b int number']

We also provide 2 command for extract class object

class_list = CppParser.get_class_list(root_node)
# and
metadata = CppParser.get_metadata_list(root_node)

Limitations

codetext heavly depends on tree-sitter syntax:

• Since we use tree-sitter grammar to extract desire node like function, class, function's name (identifier) or class's argument list, etc. codetext is easily vulnerable by tree-sitter update patch or syntax change in future.

• While we try our best to capture all possiblity, there are still plenty out there. We open for community to contribute into this project.