man2lore.py

Revision 26264, 7.6 kB (checked in by exarkun, 4 months ago)

Merge minidom-lore-3560-4

Author: exarkun
Reviewer: radix, glyph
Fixes: #3560, #414

Switch Lore from using twisted.web.microdom to using xml.dom.

This explicitly makes the default Lore input document DTD XHTML1-Strict,
but allows either that DTD or XHTML1-Transitional (but not XHTML1-Frameset
or any other DTD).

Line
1	# -- test-case-name: twisted.lore.test.test_man2lore --
2	# Copyright (c) 2001-2009 Twisted Matrix Laboratories.
3	# See LICENSE for details.
4
5	"""
6	man2lore: Converts man page source (i.e. groff) into lore-compatible html.
7
8	This is nasty and hackish (and doesn't support lots of real groff), but is good
9	enough for converting fairly simple man pages.
10	"""
11
12	import re, os
13
14	quoteRE = re.compile(*'"(.?)"'**)
15
16
17
18	def escape(text):
19	text = text.replace('<', '<').replace('>', '>')
20	text = quoteRE.sub('<q>\\1</q>', text)
21	return text
22
23
24
25	def stripQuotes(s):
26	if s[0] == s[-1] == '"':
27	s = s[1:-1]
28	return s
29
30
31
32	class ManConverter(object):
33	"""
34	Convert a man page to the Lore format.
35
36	@ivar tp: State variable for handling text inside a C{TP} token. It can
37	take values from 0 to 3:
38	- 0: when outside of a C{TP} token.
39	- 1: once a C{TP} token has been encountered. If the previous value
40	was 0, a definition list is started. Then, at the first line of
41	text, a definition term is started.
42	- 2: when the first line after the C{TP} token has been handled.
43	The definition term is closed, and a definition is started with
44	the next line of text.
45	- 3: when the first line as definition data has been handled.
46	@type tp: C{int}
47	"""
48	state = 'regular'
49	name = None
50	tp = 0
51	dl = 0
52	para = 0
53
54	def convert(self, inf, outf):
55	self.write = outf.write
56	longline = ''
57	for line in inf.readlines():
58	if line.rstrip() and line.rstrip()[-1] == '\\':
59	longline += line.rstrip()[:-1] + ' '
60	continue
61	if longline:
62	line = longline + line
63	longline = ''
64	self.lineReceived(line)
65	self.closeTags()
66	self.write('</body>\n</html>\n')
67	outf.flush()
68
69
70	def lineReceived(self, line):
71	if line[0] == '.':
72	f = getattr(self, 'macro_' + line[1:3].rstrip().upper(), None)
73	if f:
74	f(line[3:].strip())
75	else:
76	self.text(line)
77
78
79	def continueReceived(self, cont):
80	if not cont:
81	return
82	if cont[0].isupper():
83	f = getattr(self, 'macro_' + cont[:2].rstrip().upper(), None)
84	if f:
85	f(cont[2:].strip())
86	else:
87	self.text(cont)
88
89
90	def closeTags(self):
91	if self.state != 'regular':
92	self.write('</%s>' % self.state)
93	if self.tp == 3:
94	self.write('</dd>\n\n')
95	self.tp = 0
96	if self.dl:
97	self.write('</dl>\n\n')
98	self.dl = 0
99	if self.para:
100	self.write('</p>\n\n')
101	self.para = 0
102
103
104	def paraCheck(self):
105	if not self.tp and not self.para:
106	self.write('<p>')
107	self.para = 1
108
109
110	def macro_TH(self, line):
111	self.write(
112	'<?xml version="1.0"?>\n'
113	'<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"\n'
114	' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">\n')
115	self.write('<html><head>\n')
116	parts = [stripQuotes(x) for x in line.split(' ', 2)] + ['', '']
117	title, manSection = parts[:2]
118	self.write('<title>%s.%s</title>' % (title, manSection))
119	self.write('</head>\n<body>\n\n')
120	self.write('<h1>%s.%s</h1>\n\n' % (title, manSection))
121
122	macro_DT = macro_TH
123
124
125	def macro_SH(self, line):
126	self.closeTags()
127	self.write('<h2>')
128	self.para = 1
129	self.text(stripQuotes(line))
130	self.para = 0
131	self.closeTags()
132	self.write('</h2>\n\n')
133
134
135	def macro_B(self, line):
136	words = line.split()
137	words[0] = '\\fB' + words[0] + '\\fR '
138	self.text(' '.join(words))
139
140
141	def macro_NM(self, line):
142	if not self.name:
143	self.name = line
144	self.text(self.name + ' ')
145
146
147	def macro_NS(self, line):
148	parts = line.split(' Ns ')
149	i = 0
150	for l in parts:
151	i = not i
152	if i:
153	self.text(l)
154	else:
155	self.continueReceived(l)
156
157
158	def macro_OO(self, line):
159	self.text('[')
160	self.continueReceived(line)
161
162
163	def macro_OC(self, line):
164	self.text(']')
165	self.continueReceived(line)
166
167
168	def macro_OP(self, line):
169	self.text('[')
170	self.continueReceived(line)
171	self.text(']')
172
173
174	def macro_FL(self, line):
175	parts = line.split()
176	self.text('\\fB-%s\\fR' % parts[0])
177	self.continueReceived(' '.join(parts[1:]))
178
179
180	def macro_AR(self, line):
181	parts = line.split()
182	self.text('\\fI %s\\fR' % parts[0])
183	self.continueReceived(' '.join(parts[1:]))
184
185
186	def macro_PP(self, line):
187	self.closeTags()
188
189
190	def macro_IC(self, line):
191	cmd = line.split(' ', 1)[0]
192	args = line[line.index(cmd) + len(cmd):]
193	args = args.split(' ')
194	text = cmd
195	while args:
196	arg = args.pop(0)
197	if arg.lower() == "ar":
198	text += " \\fU%s\\fR" % (args.pop(0),)
199	elif arg.lower() == "op":
200	ign = args.pop(0)
201	text += " [\\fU%s\\fR]" % (args.pop(0),)
202
203	self.text(text)
204
205
206	def macro_TP(self, line):
207	"""
208	Handle C{TP} token: start a definition list if it's first token, or
209	close previous definition data.
210	"""
211	if self.tp == 3:
212	self.write('</dd>\n\n')
213	self.tp = 1
214	else:
215	self.tp = 1
216	self.write('<dl>')
217	self.dl = 1
218
219
220	def macro_BL(self, line):
221	self.write('<dl>')
222	self.tp = 1
223
224
225	def macro_EL(self, line):
226	if self.tp == 3:
227	self.write('</dd>')
228	self.tp = 1
229	self.write('</dl>\n\n')
230	self.tp = 0
231
232
233	def macro_IT(self, line):
234	if self.tp == 3:
235	self.write('</dd>')
236	self.tp = 1
237	self.continueReceived(line)
238
239
240	def text(self, line):
241	"""
242	Handle a line of text without detected token.
243	"""
244	if self.tp == 1:
245	self.write('<dt>')
246	if self.tp == 2:
247	self.write('<dd>')
248	self.paraCheck()
249
250	bits = line.split('\\')
251	self.write(escape(bits[0]))
252	for bit in bits[1:]:
253	if bit[:2] == 'fI':
254	self.write('<em>' + escape(bit[2:]))
255	self.state = 'em'
256	elif bit[:2] == 'fB':
257	self.write('<strong>' + escape(bit[2:]))
258	self.state = 'strong'
259	elif bit[:2] == 'fR':
260	self.write('</%s>' % self.state)
261	self.write(escape(bit[2:]))
262	self.state = 'regular'
263	elif bit[:2] == 'fU':
264	# fU doesn't really exist, but it helps us to manage underlined
265	# text.
266	self.write('<u>' + escape(bit[2:]))
267	self.state = 'u'
268	elif bit[:3] == '(co':
269	self.write('©' + escape(bit[3:]))
270	else:
271	self.write(escape(bit))
272
273	if self.tp == 1:
274	self.write('</dt>')
275	self.tp = 2
276	elif self.tp == 2:
277	self.tp = 3
278
279
280
281	class ProcessingFunctionFactory:
282
283	def generate_lore(self, d, filenameGenerator=None):
284	ext = d.get('ext', '.html')
285	return lambda file,_: ManConverter().convert(open(file),
286	open(os.path.splitext(file)[0]+ext, 'w'))
287
288
289
290	factory = ProcessingFunctionFactory()
291
292
293	if __name__ == '__main__':
294	import sys
295	mc = ManConverter().convert(open(sys.argv[1]), sys.stdout)

Note: See TracBrowser for help on using the browser.

root / trunk / twisted / lore / man2lore.py

Download in other formats: