Disclaimer: This is a side project of mine and has not been rigorously tested.
pymatlabparser is a parser for Matlab/Octave language, a popular language in engineering and engineering education. This parser can handle many of the unique syntax rules of Matlab, such as:
- Apostrophe's dual role for quotation ('strings') and matrix transpose (matrix')
- Input: {'ABC' ABC'}
code_block ──── statement ──── cell_array ──── matrx_elements ─┬── 'ABC'
└── "'" oper ──── ABC
- Dot's dual role for accessing a struct's field (a.b) and matrix operations (a.^b)
- Input: a.b + a.^2
code_block ──── statement ──── "+" oper ─┬── dot ─┬── a
│ └── b
└── ".^" oper ─┬── a
└── 2
- Anonymous functions
- Input: @(x, y, z) x.^2 + y.^2 + z.^2
code_block ──── statement ──── anonym_func ─┬── args ─┬── x
│ ├── y
│ └── z
└── expr ──── "+" oper ─┬── "+" oper ─┬── ".^" oper ─┬── x
│ │ └── 2
│ └── ".^" oper ─┬── y
│ └── 2
└── ".^" oper ─┬── z
└── 2
- Matlab's for-loops, while blocks, if blocks, etc.
- Input: for i = 1 : 10; disp(i); if i == 5; break; end; end
code_block ──── for_loop ─┬── for ──── assign ─┬── i
│ └── range ─┬── 1
│ └── 10
└── body ─┬── statement ──── func_call/array_idxing ─┬── expr ──── disp
│ └── args ──── i
└── if_block ─┬── if ──── "==" oper ─┬── i
│ └── 5
└── body ──── statement ──── func_call/array_idxing ─┬── expr ──── disp
└── args ──── i
Among other features.
For possible future projects on alternative Matlab/Octave interpreters, IDE plugins, education, etc.
Currently this project is experimental; Unexpected problems can occur from time to time, and the author strives to improve it continuously.
This project uses the lexer-parser framework, implemented using sly Python package. This is the only dependency you'd need to install, which is available using pip.
To generate the tree views like the ones shown above, simply execute the matlab_parser.py file. To use it in code, create a MatlabParser object and call its parse() function (just like how it works in sly or ply package). This will generate an abstract syntax tree (AST) node.
Every AST node is a tuple that has two elements: the first element is a string name that describes this node (e.g. for_loop, * operation, function call, etc.), and the second element is a tuple of children node(s) which follow the same format of AST node.
To keep the code simple, there is no dedicated class for AST node; only these tuples are used.