convert PTB-style parse tree (essentially, an sexps) to the format for LaTeX's `qtree`/`tikz-qtree` library

treeify.py

#!/usr/bin/env python
# treeify.py: convert PTB parse to LaTeX's `qtree`/`tikz-qtree` format
#
# NB: this only works for documents with a single tree, due to a limitation
# with `nltk.tree`.

import fileinput

from nltk import Tree

# list of LaTeX reserved chars, from:
# http://tex.stackexchange.com/questions/34580/escape-character-in-latex


translations = {"&": r"\&",
                "%": r"\%",
                "$": r"\$",
                "#": r"\#",
                "_": r"\_",
                "{": r"\{",
                "}": r"\}",
                "~": r"\textasciitilde",
                "^": r"\textasciicircum",
                "\\": r"\textbackslash"}
            

if __name__ == "__main__":
    tstring = "".join(fileinput.input()).strip()
    tree = Tree.fromstring(tstring)
    treestring = tree.pprint_latex_qtree()
    newstring = treestring[:5]
    for char in treestring[5:]:
        newstring += translations.get(char, char)
    print(newstring)