cpython/Tools/scripts/highlight.py

266 lines
8.9 KiB
Python
Raw Normal View History

2012-06-30 20:58:06 -03:00
#!/usr/bin/env python3
'''Add syntax highlighting to Python source code'''
2012-07-03 18:42:33 -03:00
2012-07-01 03:19:30 -03:00
__author__ = 'Raymond Hettinger'
2012-06-30 20:58:06 -03:00
import builtins
import functools
import html as html_module
import keyword
import re
import tokenize
2012-07-10 03:52:08 -03:00
#### Analyze Python Source #################################
2012-06-30 20:58:06 -03:00
def is_builtin(s):
'Return True if s is the name of a builtin'
return hasattr(builtins, s)
2012-06-30 20:58:06 -03:00
def combine_range(lines, start, end):
'Join content from a range of lines between start and end'
(srow, scol), (erow, ecol) = start, end
if srow == erow:
return lines[srow-1][scol:ecol], end
rows = [lines[srow-1][scol:]] + lines[srow: erow-1] + [lines[erow-1][:ecol]]
return ''.join(rows), end
def analyze_python(source):
'''Generate and classify chunks of Python for syntax highlighting.
Yields tuples in the form: (category, categorized_text).
'''
2012-07-03 04:15:59 -03:00
lines = source.splitlines(True)
lines.append('')
2012-06-30 20:58:06 -03:00
readline = functools.partial(next, iter(lines), '')
kind = tok_str = ''
tok_type = tokenize.COMMENT
written = (1, 0)
2012-06-30 20:58:06 -03:00
for tok in tokenize.generate_tokens(readline):
prev_tok_type, prev_tok_str = tok_type, tok_str
tok_type, tok_str, (srow, scol), (erow, ecol), logical_lineno = tok
2012-07-03 04:12:27 -03:00
kind = ''
2012-06-30 20:58:06 -03:00
if tok_type == tokenize.COMMENT:
kind = 'comment'
elif tok_type == tokenize.OP and tok_str[:1] not in '{}[](),.:;@':
2012-06-30 20:58:06 -03:00
kind = 'operator'
elif tok_type == tokenize.STRING:
kind = 'string'
if prev_tok_type == tokenize.INDENT or scol==0:
kind = 'docstring'
elif tok_type == tokenize.NAME:
if tok_str in ('def', 'class', 'import', 'from'):
kind = 'definition'
elif prev_tok_str in ('def', 'class'):
kind = 'defname'
elif keyword.iskeyword(tok_str):
kind = 'keyword'
elif is_builtin(tok_str) and prev_tok_str != '.':
kind = 'builtin'
if kind:
text, written = combine_range(lines, written, (srow, scol))
yield '', text
text, written = tok_str, (erow, ecol)
yield kind, text
line_upto_token, written = combine_range(lines, written, (erow, ecol))
yield '', line_upto_token
2012-07-10 03:52:08 -03:00
#### Raw Output ###########################################
def raw_highlight(classified_text):
'Straight text display of text classifications'
result = []
for kind, text in classified_text:
result.append('%15s: %r\n' % (kind or 'plain', text))
2012-07-10 03:52:08 -03:00
return ''.join(result)
#### ANSI Output ###########################################
2012-07-03 18:11:40 -03:00
default_ansi = {
'comment': ('\033[0;31m', '\033[0m'),
'string': ('\033[0;32m', '\033[0m'),
'docstring': ('\033[0;32m', '\033[0m'),
'keyword': ('\033[0;33m', '\033[0m'),
'builtin': ('\033[0;35m', '\033[0m'),
'definition': ('\033[0;33m', '\033[0m'),
'defname': ('\033[0;34m', '\033[0m'),
'operator': ('\033[0;33m', '\033[0m'),
2012-07-03 18:11:40 -03:00
}
def ansi_highlight(classified_text, colors=default_ansi):
'Add syntax highlighting to source code using ANSI escape sequences'
2012-07-03 18:11:40 -03:00
# http://en.wikipedia.org/wiki/ANSI_escape_code
result = []
for kind, text in classified_text:
opener, closer = colors.get(kind, ('', ''))
result += [opener, text, closer]
2012-07-03 18:11:40 -03:00
return ''.join(result)
2012-07-10 03:52:08 -03:00
#### HTML Output ###########################################
def html_highlight(classified_text,opener='<pre class="python">\n', closer='</pre>\n'):
'Convert classified text to an HTML fragment'
result = [opener]
for kind, text in classified_text:
if kind:
result.append('<span class="%s">' % kind)
result.append(html_module.escape(text))
2012-06-30 20:58:06 -03:00
if kind:
result.append('</span>')
result.append(closer)
return ''.join(result)
2012-06-30 20:58:06 -03:00
default_css = {
'.comment': '{color: crimson;}',
'.string': '{color: forestgreen;}',
'.docstring': '{color: forestgreen; font-style:italic;}',
2012-06-30 20:58:06 -03:00
'.keyword': '{color: darkorange;}',
'.builtin': '{color: purple;}',
'.definition': '{color: darkorange; font-weight:bold;}',
'.defname': '{color: blue;}',
'.operator': '{color: brown;}',
}
default_html = '''\
2012-07-01 02:19:04 -03:00
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">
<html>
<head>
<meta http-equiv="Content-type" content="text/html;charset=UTF-8">
<title> {title} </title>
2012-07-01 02:19:04 -03:00
<style type="text/css">
{css}
2012-07-01 02:19:04 -03:00
</style>
</head>
2012-06-30 20:58:06 -03:00
<body>
{body}
2012-07-01 02:19:04 -03:00
</body>
</html>
2012-06-30 20:58:06 -03:00
'''
def build_html_page(classified_text, title='python',
css=default_css, html=default_html):
'Create a complete HTML page with colorized source code'
2012-07-01 02:19:04 -03:00
css_str = '\n'.join(['%s %s' % item for item in css.items()])
result = html_highlight(classified_text)
title = html_module.escape(title)
return html.format(title=title, css=css_str, body=result)
2012-06-30 20:58:06 -03:00
2012-07-10 03:52:08 -03:00
#### LaTeX Output ##########################################
2012-07-14 21:58:29 -03:00
default_latex_commands = {
'comment': r'{\color{red}#1}',
'string': r'{\color{ForestGreen}#1}',
'docstring': r'{\emph{\color{ForestGreen}#1}}',
'keyword': r'{\color{orange}#1}',
'builtin': r'{\color{purple}#1}',
'definition': r'{\color{orange}#1}',
'defname': r'{\color{blue}#1}',
'operator': r'{\color{brown}#1}',
2012-07-10 03:52:08 -03:00
}
default_latex_document = r'''
\documentclass{article}
\usepackage{alltt}
2012-07-14 21:58:29 -03:00
\usepackage{upquote}
2012-07-10 03:52:08 -03:00
\usepackage{color}
\usepackage[usenames,dvipsnames]{xcolor}
\usepackage[cm]{fullpage}
2012-07-14 21:58:29 -03:00
%(macros)s
2012-07-10 03:52:08 -03:00
\begin{document}
\center{\LARGE{%(title)s}}
\begin{alltt}
%(body)s
\end{alltt}
\end{document}
'''
def alltt_escape(s):
'Replace backslash and braces with their escaped equivalents'
xlat = {'{': r'\{', '}': r'\}', '\\': r'\textbackslash{}'}
return re.sub(r'[\\{}]', lambda mo: xlat[mo.group()], s)
2012-07-10 03:52:08 -03:00
def latex_highlight(classified_text, title = 'python',
2012-07-14 21:58:29 -03:00
commands = default_latex_commands,
2012-07-10 03:52:08 -03:00
document = default_latex_document):
'Create a complete LaTeX document with colorized source code'
2012-07-14 21:58:29 -03:00
macros = '\n'.join(r'\newcommand{\py%s}[1]{%s}' % c for c in commands.items())
2012-07-10 03:52:08 -03:00
result = []
for kind, text in classified_text:
if kind:
2012-07-14 21:58:29 -03:00
result.append(r'\py%s{' % kind)
result.append(alltt_escape(text))
2012-07-10 03:52:08 -03:00
if kind:
result.append('}')
2012-07-14 21:58:29 -03:00
return default_latex_document % dict(title=title, macros=macros, body=''.join(result))
2012-07-10 03:52:08 -03:00
2012-06-30 20:58:06 -03:00
if __name__ == '__main__':
import argparse
import os.path
import sys
import textwrap
import webbrowser
2012-06-30 20:58:06 -03:00
parser = argparse.ArgumentParser(
description = 'Add syntax highlighting to Python source code',
formatter_class=argparse.RawDescriptionHelpFormatter,
epilog = textwrap.dedent('''
examples:
# Show syntax highlighted code in the terminal window
$ ./highlight.py myfile.py
# Colorize myfile.py and display in a browser
$ ./highlight.py -b myfile.py
# Create an HTML section to embed in an existing webpage
./highlight.py -s myfile.py
# Create a complete HTML file
$ ./highlight.py -c myfile.py > myfile.html
2012-07-10 03:52:08 -03:00
# Create a PDF using LaTeX
$ ./highlight.py -l myfile.py | pdflatex
'''))
2012-07-03 04:12:27 -03:00
parser.add_argument('sourcefile', metavar = 'SOURCEFILE',
help = 'file containing Python sourcecode')
2012-06-30 20:58:06 -03:00
parser.add_argument('-b', '--browser', action = 'store_true',
help = 'launch a browser to show results')
2012-07-03 21:55:23 -03:00
parser.add_argument('-c', '--complete', action = 'store_true',
help = 'build a complete html webpage')
2012-07-10 03:52:08 -03:00
parser.add_argument('-l', '--latex', action = 'store_true',
help = 'build a LaTeX document')
parser.add_argument('-r', '--raw', action = 'store_true',
help = 'raw parse of categorized text')
2012-07-03 04:12:27 -03:00
parser.add_argument('-s', '--section', action = 'store_true',
help = 'show an HTML section rather than a complete webpage')
2012-06-30 20:58:06 -03:00
args = parser.parse_args()
2012-07-03 18:11:40 -03:00
2012-07-03 21:55:23 -03:00
if args.section and (args.browser or args.complete):
2012-07-03 04:12:27 -03:00
parser.error('The -s/--section option is incompatible with '
2012-07-03 21:55:23 -03:00
'the -b/--browser or -c/--complete options')
2012-06-30 20:58:06 -03:00
2012-07-03 04:12:27 -03:00
sourcefile = args.sourcefile
2012-06-30 20:58:06 -03:00
with open(sourcefile) as f:
2012-07-03 18:42:33 -03:00
source = f.read()
classified_text = analyze_python(source)
2012-07-03 18:11:40 -03:00
2012-07-10 03:52:08 -03:00
if args.raw:
encoded = raw_highlight(classified_text)
elif args.complete or args.browser:
encoded = build_html_page(classified_text, title=sourcefile)
2012-07-03 18:11:40 -03:00
elif args.section:
encoded = html_highlight(classified_text)
2012-07-10 03:52:08 -03:00
elif args.latex:
encoded = latex_highlight(classified_text, title=sourcefile)
2012-07-03 18:11:40 -03:00
else:
encoded = ansi_highlight(classified_text)
2012-07-03 18:11:40 -03:00
2012-06-30 20:58:06 -03:00
if args.browser:
htmlfile = os.path.splitext(os.path.basename(sourcefile))[0] + '.html'
with open(htmlfile, 'w') as f:
2012-07-03 18:11:40 -03:00
f.write(encoded)
2012-06-30 20:58:06 -03:00
webbrowser.open('file://' + os.path.abspath(htmlfile))
else:
2012-07-03 18:11:40 -03:00
sys.stdout.write(encoded)