| 1 |
# -*- coding: utf-8 -*- |
| 2 |
"""Pygments lexer for text containing ANSI color codes.""" |
| 3 |
from __future__ import ( |
| 4 |
absolute_import, |
| 5 |
unicode_literals, |
| 6 |
) |
| 7 |
|
| 8 |
import itertools |
| 9 |
import re |
| 10 |
|
| 11 |
import pygments.lexer |
| 12 |
import pygments.token |
| 13 |
|
| 14 |
Color = pygments.token.Token.Color |
| 15 |
|
| 16 |
_ansi_code_to_color = { |
| 17 |
0: 'Black', |
| 18 |
1: 'Red', |
| 19 |
2: 'Green', |
| 20 |
3: 'Yellow', |
| 21 |
4: 'Blue', |
| 22 |
5: 'Magenta', |
| 23 |
6: 'Cyan', |
| 24 |
7: 'White', |
| 25 |
} |
| 26 |
|
| 27 |
|
| 28 |
def _token_from_lexer_state(bold, fg_color, bg_color): |
| 29 |
"""Construct a token given the current lexer state. |
| 30 |
|
| 31 |
We can only emit one token even though we have a multiple-tuple state. |
| 32 |
To do work around this, we construct tokens like "BoldRed". |
| 33 |
""" |
| 34 |
token_name = '' |
| 35 |
|
| 36 |
if bold: |
| 37 |
token_name += 'Bold' |
| 38 |
|
| 39 |
if fg_color: |
| 40 |
token_name += fg_color |
| 41 |
|
| 42 |
if bg_color: |
| 43 |
token_name += 'BG' + bg_color |
| 44 |
|
| 45 |
if token_name == '': |
| 46 |
return pygments.token.Text |
| 47 |
else: |
| 48 |
return getattr(Color, token_name) |
| 49 |
|
| 50 |
|
| 51 |
def color_tokens(fg_colors, bg_colors): |
| 52 |
"""Return color tokens for a given set of colors. |
| 53 |
|
| 54 |
Pygments doesn't have a generic "color" token; instead everything is |
| 55 |
contextual (e.g. "comment" or "variable"). That doesn't make sense for us, |
| 56 |
where the colors actually *are* what we care about. |
| 57 |
|
| 58 |
This function will register combinations of tokens (things like "Red" or |
| 59 |
"BoldRedBGGreen") based on the colors passed in. |
| 60 |
|
| 61 |
You can also define the tokens yourself, but note that the token names are |
| 62 |
*not* currently guaranteed to be stable between releases as I'm not really |
| 63 |
happy with this approach. |
| 64 |
|
| 65 |
Usage: |
| 66 |
|
| 67 |
fg_colors = bg_colors = { |
| 68 |
'Black': '#000000', |
| 69 |
'Red': '#EF2929', |
| 70 |
'Green': '#8AE234', |
| 71 |
'Yellow': '#FCE94F', |
| 72 |
'Blue': '#3465A4', |
| 73 |
'Magenta': '#c509c5', |
| 74 |
'Cyan': '#34E2E2', |
| 75 |
'White': '#ffffff', |
| 76 |
} |
| 77 |
class MyStyle(pygments.styles.SomeStyle): |
| 78 |
styles = dict(pygments.styles.SomeStyle.styles) |
| 79 |
styles.update(color_tokens(fg_colors, bg_colors)) |
| 80 |
""" |
| 81 |
styles = {} |
| 82 |
|
| 83 |
for bold, fg_color, bg_color in itertools.product( |
| 84 |
(False, True), |
| 85 |
{None} | set(fg_colors), |
| 86 |
{None} | set(bg_colors), |
| 87 |
): |
| 88 |
token = _token_from_lexer_state(bold, fg_color, bg_color) |
| 89 |
if token is not pygments.token.Text: |
| 90 |
value = [] |
| 91 |
if bold: |
| 92 |
value.append('bold') |
| 93 |
if fg_color: |
| 94 |
value.append(fg_colors[fg_color]) |
| 95 |
if bg_color: |
| 96 |
value.append('bg:' + bg_colors[bg_color]) |
| 97 |
styles[token] = ' '.join(value) |
| 98 |
|
| 99 |
return styles |
| 100 |
|
| 101 |
|
| 102 |
class AnsiColorLexer(pygments.lexer.RegexLexer): |
| 103 |
name = 'ANSI Color' |
| 104 |
aliases = ('ansi-color', 'ansi', 'ansi-terminal') |
| 105 |
flags = re.DOTALL | re.MULTILINE |
| 106 |
|
| 107 |
def __init__(self, *args, **kwargs): |
| 108 |
super(AnsiColorLexer, self).__init__(*args, **kwargs) |
| 109 |
self.reset_state() |
| 110 |
|
| 111 |
def reset_state(self): |
| 112 |
self.bold = False |
| 113 |
self.fg_color = None |
| 114 |
self.bg_color = None |
| 115 |
|
| 116 |
@property |
| 117 |
def current_token(self): |
| 118 |
return _token_from_lexer_state( |
| 119 |
self.bold, self.fg_color, self.bg_color, |
| 120 |
) |
| 121 |
|
| 122 |
def process(self, match): |
| 123 |
"""Produce the next token and bit of text. |
| 124 |
|
| 125 |
Interprets the ANSI code (which may be a color code or some other |
| 126 |
code), changing the lexer state and producing a new token. If it's not |
| 127 |
a color code, we just strip it out and move on. |
| 128 |
|
| 129 |
Some useful reference for ANSI codes: |
| 130 |
* http://ascii-table.com/ansi-escape-sequences.php |
| 131 |
""" |
| 132 |
# "after_escape" contains everything after the start of the escape |
| 133 |
# sequence, up to the next escape sequence. We still need to separate |
| 134 |
# the content from the end of the escape sequence. |
| 135 |
after_escape = match.group(1) |
| 136 |
|
| 137 |
# TODO: this doesn't handle the case where the values are non-numeric. |
| 138 |
# This is rare but can happen for keyboard remapping, e.g. |
| 139 |
# '\x1b[0;59;"A"p' |
| 140 |
parsed = re.match( |
| 141 |
r'([0-9;=]*?)?([a-zA-Z])(.*)$', |
| 142 |
after_escape, |
| 143 |
re.DOTALL | re.MULTILINE, |
| 144 |
) |
| 145 |
if parsed is None: |
| 146 |
# This shouldn't ever happen if we're given valid text + ANSI, but |
| 147 |
# people can provide us with utter junk, and we should tolerate it. |
| 148 |
text = after_escape |
| 149 |
else: |
| 150 |
value, code, text = parsed.groups() |
| 151 |
|
| 152 |
if code == 'm': # "m" is "Set Graphics Mode" |
| 153 |
# Special case \x1b[m is a reset code |
| 154 |
if value == '': |
| 155 |
self.reset_state() |
| 156 |
else: |
| 157 |
values = value.split(';') |
| 158 |
for value in values: |
| 159 |
try: |
| 160 |
value = int(value) |
| 161 |
except ValueError: |
| 162 |
# Shouldn't ever happen, but could with invalid |
| 163 |
# ANSI. |
| 164 |
continue |
| 165 |
else: |
| 166 |
fg_color = _ansi_code_to_color.get(value - 30) |
| 167 |
bg_color = _ansi_code_to_color.get(value - 40) |
| 168 |
if fg_color: |
| 169 |
self.fg_color = fg_color |
| 170 |
elif bg_color: |
| 171 |
self.bg_color = bg_color |
| 172 |
elif value == 1: |
| 173 |
self.bold = True |
| 174 |
elif value == 22: |
| 175 |
self.bold = False |
| 176 |
elif value == 39: |
| 177 |
self.fg_color = None |
| 178 |
elif value == 49: |
| 179 |
self.bg_color = None |
| 180 |
elif value == 0: |
| 181 |
self.reset_state() |
| 182 |
|
| 183 |
yield match.start(), self.current_token, text |
| 184 |
|
| 185 |
tokens = { |
| 186 |
# states have to be native strings |
| 187 |
str('root'): [ |
| 188 |
(r'\x1b\[([^\x1b]*)', process), |
| 189 |
(r'[^\x1b]+', pygments.token.Text), |
| 190 |
], |
| 191 |
} |