The asttokens module annotates Python abstract syntax trees (ASTs) with the positions of tokens and text in the source code that generated them.

It makes it possible for tools that work with logical AST nodes to find the particular text that resulted in those nodes, for example for automated refactoring or highlighting.

asttokens is available on PyPI: https://pypi.python.org/pypi/asttokens/:

pip install asttokens

The code is on GitHub: https://github.com/gristlabs/asttokens.

The API Reference is here: http://asttokens.readthedocs.io/en/latest/api-index.html.

ASTTokens works with both Python2 and Python3.

ASTTokens can annotate both trees built by ast, AND those built by astroid.

Here's an example:

import asttokens, ast
source = "Robot('blue').walk(steps=10*n)"
atok = asttokens.ASTTokens(source, parse=True)

Once the tree has been marked, nodes get .first_token, .last_token attributes, and the ASTTokens object offers helpful methods:

attr_node = next(n for n in ast.walk(atok.tree) if isinstance(n, ast.Attribute))
print(atok.get_text(attr_node))
start, end = attr_node.last_token.startpos, attr_node.last_token.endpos
print(atok.text[:start] + 'RUN' + atok.text[end:])

Which produces this output:

Robot('blue').walk
Robot('blue').RUN(steps=10*n)

The ASTTokens object also offers methods to walk and search the list of tokens that make up the code (or a particular AST node), which is more useful and powerful than dealing with the text directly.

To contribute:

1. Fork this repository, and clone your fork.

2. Install the package with test dependencies (ideally in a virtualenv) with:

   pip install -e '.[test]'

3. Run tests in your current interpreter with the command pytest or python -m pytest.
4. Run tests across all supported interpreters with the tox command. You will need to have the interpreters installed separately. We recommend pyenv for that. Use tox -p auto to run the tests in parallel.
5. By default certain tests which take a very long time to run are skipped, but they are run in CI. These are marked using the pytest marker slow and can be run on their own with pytest -m slow or as part of the full suite with pytest -m ''.