1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
|
#!/usr/bin/env python3
#
# html2text.py - converts HTML to text
#
# Wireshark - Network traffic analyzer
# By Gerald Combs <gerald@wireshark.org>
# Copyright 1998 Gerald Combs
#
# SPDX-License-Identifier: GPL-2.0-or-later
from __future__ import unicode_literals
__author__ = "Peter Wu <peter@lekensteyn.nl>"
__copyright__ = "Copyright 2015, Peter Wu"
__license__ = "GPL (v2 or later)"
# TODO:
# multiple list indentation levels (modify bullets?)
# maybe allow for ascii output instead of utf-8?
import sys
from textwrap import TextWrapper
try:
from HTMLParser import HTMLParser
from htmlentitydefs import name2codepoint
except ImportError: # Python 3
from html.parser import HTMLParser
from html.entities import name2codepoint
unichr = chr # for html entity handling
class TextHTMLParser(HTMLParser):
"""Converts a HTML document to text."""
def __init__(self):
try:
# Python 3.4
HTMLParser. __init__(self, convert_charrefs=True)
except Exception:
HTMLParser. __init__(self)
# All text, concatenated
self.output_buffer = ''
# The current text block which is being constructed
self.text_block = ''
# Whether the previous element was terminated with whitespace
self.need_space = False
# Whether to prevent word-wrapping the contents (for "pre" tag)
self.skip_wrap = False
# Quoting
self.need_quote = False
self.quote_stack = []
# Suffixes
self.need_suffix = False
self.suffix_stack = []
# track list items
self.list_item_prefix = None
self.ordered_list_index = None
self.stack_list_item_prefix = []
self.stack_ordered_list_index = []
self.list_indent_level = 0
self.list_item_indent = ""
# Indentation (for heading and paragraphs)
self.indent_levels = [0, 0]
# Don't dump CSS, scripts, etc.
self.ignore_tags = ('head', 'style', 'script')
self.ignore_level = 0
# href footnotes.
self.footnotes = []
self.href = None
def _wrap_text(self, text):
"""Wraps text, but additionally indent list items."""
initial_indent = indent = sum(self.indent_levels) * ' '
if self.list_item_prefix:
initial_indent += self.list_item_prefix
indent += ' '
kwargs = {
'width': 72,
'initial_indent': initial_indent,
'subsequent_indent': indent
}
kwargs['break_on_hyphens'] = False
wrapper = TextWrapper(**kwargs)
return '\n'.join(wrapper.wrap(text))
def _commit_block(self, newline='\n\n'):
text = self.text_block
if text:
if not self.skip_wrap:
text = self._wrap_text(text)
self.output_buffer += text + newline
self.text_block = ''
self.need_space = False
def handle_starttag(self, tag, attrs):
# end a block of text on <br>, but also flush list items which are not
# terminated.
if tag == 'br' or tag == 'li':
self._commit_block('\n')
if tag == 'code':
self.need_quote = True
self.quote_stack.append('`')
if tag == 'pre':
self.skip_wrap = True
if tag in ('ol', 'ul'):
self.list_indent_level += 1
self.list_item_indent = " " * (self.list_indent_level - 1)
self.stack_ordered_list_index.append(self.ordered_list_index)
self.stack_list_item_prefix.append(self.list_item_prefix)
# Following list items are numbered.
if tag == 'ol':
self.ordered_list_index = 1
if tag == 'ul':
self.list_item_prefix = self.list_item_indent + ' • '
if tag == 'li' and self.ordered_list_index:
self.list_item_prefix = self.list_item_indent + ' %d. ' % (self.ordered_list_index)
self.ordered_list_index += 1
if tag[0] == 'h' and len(tag) == 2 and \
(tag[1] >= '1' and tag[1] <= '6'):
self.indent_levels = [int(tag[1]) - 1, 0]
if tag == 'p':
self.indent_levels[1] = 1
if tag == 'a':
try:
href = [attr[1] for attr in attrs if attr[0] == 'href'][0]
if '://' in href: # Skip relative URLs and links.
self.href = href
except IndexError:
self.href = None
if tag == 'span':
try:
el_class = [attr[1] for attr in attrs if attr[0] == 'class'][0]
if 'menuseq' in el_class:
self.need_quote = True
self.quote_stack.append('"')
except IndexError:
pass
if tag == 'div':
try:
el_class = [attr[1] for attr in attrs if attr[0] == 'class'][0]
if 'title' in el_class.split(' '):
self.need_suffix = True
self.suffix_stack.append(':')
except IndexError:
pass
if tag == 'sup':
self.need_quote = True
self.quote_stack.append('^')
if tag in self.ignore_tags:
self.ignore_level += 1
def handle_data(self, data):
quote = ''
if self.need_quote:
quote = self.quote_stack[-1]
suffix = ''
if self.need_suffix:
suffix = self.suffix_stack.pop()
if self.ignore_level > 0:
return
elif self.skip_wrap:
block = data
else:
if self.href and data == self.href:
# This is a self link. Don't create a footnote.
self.href = None
# For normal text, fold multiple whitespace and strip
# leading and trailing spaces for the whole block (but
# keep spaces in the middle).
block = quote
if data.strip() and data[:1].isspace():
# Keep spaces in the middle
self.need_space = True
if self.need_space and data.strip() and self.text_block:
block = ' ' + quote
block += ' '.join(data.split()) + suffix
self.need_space = data[-1:].isspace()
self.text_block += block
self.need_quote = False
self.need_suffix = False
def handle_endtag(self, tag):
block_elements = 'p li ul pre ol h1 h2 h3 h4 h5 h6 tr'
#block_elements += ' dl dd dt'
if tag in block_elements.split():
self._commit_block()
if tag in ('code', 'span'):
# XXX This span isn't guaranteed to match its opening.
self.text_block += self.quote_stack.pop()
if tag in ('ol', 'ul'):
self.list_indent_level -= 1
self.list_item_indent = " " * (self.list_indent_level - 1)
self.ordered_list_index = self.stack_ordered_list_index.pop()
self.list_item_prefix = self.stack_list_item_prefix.pop()
if tag == 'pre':
self.skip_wrap = False
if tag == 'a' and self.href:
self.footnotes.append(self.href)
self.text_block += '[{0}]'.format(len(self.footnotes))
if tag == 'sup':
self.quote_stack.pop()
if tag in self.ignore_tags:
self.ignore_level -= 1
def handle_charref(self, name):
self.handle_data(unichr(int(name)))
def handle_entityref(self, name):
self.handle_data(unichr(name2codepoint[name]))
def close(self):
HTMLParser.close(self)
self._commit_block()
if len(self.footnotes) > 0:
self.list_item_prefix = None
self.indent_levels = [1, 0]
self.text_block = 'References'
self._commit_block()
self.indent_levels = [1, 1]
footnote_num = 1
for href in self.footnotes:
self.text_block += '{0:>2}. {1}\n'.format(footnote_num, href)
footnote_num += 1
self._commit_block('\n')
byte_output = self.output_buffer.encode('utf-8')
if hasattr(sys.stdout, 'buffer'):
sys.stdout.buffer.write(byte_output)
else:
sys.stdout.write(byte_output)
def main():
htmlparser = TextHTMLParser()
if len(sys.argv) > 1 and sys.argv[1] != '-':
filename = sys.argv[1]
f = open(filename, 'rb')
else:
filename = None
f = sys.stdin
try:
if hasattr(f, 'buffer'):
# Access raw (byte) buffer in Python 3 instead of decoded one
f = f.buffer
# Read stdin as a Unicode string
htmlparser.feed(f.read().decode('utf-8'))
finally:
if filename is not None:
f.close()
htmlparser.close()
if __name__ == '__main__':
sys.exit(main())
|
