1
# Copyright (C) 2005 by Canonical Ltd
3
# Distributed under the GNU General Public Licence v2
1
# Copyright (C) 2005 Canonical Ltd
3
# This program is free software; you can redistribute it and/or modify
4
# it under the terms of the GNU General Public License as published by
5
# the Free Software Foundation; either version 2 of the License, or
6
# (at your option) any later version.
8
# This program is distributed in the hope that it will be useful,
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
# GNU General Public License for more details.
13
# You should have received a copy of the GNU General Public License
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
5
17
# \subsection{\emph{rio} - simple text metaformat}
7
19
# \emph{r} stands for `restricted', `reproducible', or `rfc822-like'.
9
21
# The stored data consists of a series of \emph{stanzas}, each of which contains
10
22
# \emph{fields} identified by an ascii name, with Unicode or string contents.
11
# The field tag is constrained to alphanumeric characters.
23
# The field tag is constrained to alphanumeric characters.
12
24
# There may be more than one field in a stanza with the same name.
14
26
# The format itself does not deal with character encoding issues, though
15
27
# the result will normally be written in Unicode.
17
29
# The format is intended to be simple enough that there is exactly one character
18
30
# stream representation of an object and vice versa, and that this relation
19
31
# will continue to hold for future versions of bzr.
33
45
def write_stanza(self, stanza):
47
self._to_file.write('\n')
36
48
stanza.write(self._to_file)
37
49
self._soft_nl = True
40
52
class RioReader(object):
41
53
"""Read stanzas from a file as a sequence
43
to_file can be anything that can be enumerated as a sequence of
55
to_file can be anything that can be enumerated as a sequence of
44
56
lines (with newlines.)
46
58
def __init__(self, from_file):
171
181
"""Return stanza as a single string"""
172
182
return ''.join(self.to_lines())
184
def to_unicode(self):
185
"""Return stanza as a single Unicode string.
187
This is most useful when adding a Stanza to a parent Stanza
193
for tag, value in self.items:
195
result.append(tag + ': \n')
197
# don't want splitlines behaviour on empty lines
198
val_lines = value.split('\n')
199
result.append(tag + ': ' + val_lines[0] + '\n')
200
for line in val_lines[1:]:
201
result.append('\t' + line + '\n')
203
result.append(tag + ': ' + value + '\n')
204
return u''.join(result)
174
206
def write(self, to_file):
175
207
"""Write stanza to a file"""
176
208
to_file.writelines(self.to_lines())
213
244
def read_stanza(line_iter):
214
245
"""Return new Stanza read from list of lines or a file
216
247
Returns one Stanza that was read, or returns None at end of file. If a
217
248
blank line follows the stanza, it is consumed. It's not an error for
218
249
there to be no blank at end of file. If there is a blank file at the
219
start of the input this is really an empty stanza and that is returned.
250
start of the input this is really an empty stanza and that is returned.
221
252
Only the stanza lines and the trailing blank (if any) are consumed
222
253
from the line_iter.
224
255
The raw lines must be in utf-8 encoding.
257
unicode_iter = (line.decode('utf-8') for line in line_iter)
258
return read_stanza_unicode(unicode_iter)
261
def read_stanza_unicode(unicode_iter):
262
"""Read a Stanza from a list of lines or a file.
264
The lines should already be in unicode form. This returns a single
265
stanza that was read. If there is a blank line at the end of the Stanza,
266
it is consumed. It is not an error for there to be no blank line at
267
the end of the iterable. If there is a blank line at the beginning,
268
this is treated as an empty Stanza and None is returned.
270
Only the stanza lines and the trailing blank (if any) are consumed
271
from the unicode_iter
273
:param unicode_iter: A iterable, yeilding Unicode strings. See read_stanza
274
if you have a utf-8 encoded string.
275
:return: A Stanza object if there are any lines in the file.
227
278
stanza = Stanza()
229
280
accum_value = None
230
for line in line_iter:
231
if line == None or line == '':
282
# TODO: jam 20060922 This code should raise real errors rather than
283
# using 'assert' to process user input, or raising ValueError
284
# rather than a more specific error.
286
for line in unicode_iter:
287
if line is None or line == '':
232
288
break # end of file
234
290
break # end of stanza
235
line = line.decode('utf-8')
236
assert line[-1] == '\n'
238
292
if line[0] == '\t': # continues previous value
246
300
colon_index = line.index(': ')
247
301
except ValueError:
248
raise ValueError('tag/value separator not found in line %r' % real_l)
302
raise ValueError('tag/value separator not found in line %r'
249
304
tag = str(line[:colon_index])
250
assert valid_tag(tag), \
251
"invalid rio tag %r" % tag
305
if not valid_tag(tag):
306
raise ValueError("invalid rio tag %r" % (tag,))
252
307
accum_value = line[colon_index+2:-1]
253
309
if tag is not None: # add last tag-value
254
310
stanza.add(tag, accum_value)
256
312
else: # didn't see any content
316
def to_patch_lines(stanza, max_width=72):
317
"""Convert a stanza into RIO-Patch format lines.
319
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
320
It resists common forms of damage such as newline conversion or the removal
321
of trailing whitespace, yet is also reasonably easy to read.
323
:param max_width: The maximum number of characters per physical line.
324
:return: a list of lines
327
raise ValueError(max_width)
328
max_rio_width = max_width - 4
330
for pline in stanza.to_lines():
331
for line in pline.split('\n')[:-1]:
332
line = re.sub('\\\\', '\\\\\\\\', line)
334
partline = line[:max_rio_width]
335
line = line[max_rio_width:]
336
if len(line) > 0 and line[0] != [' ']:
338
break_index = partline.rfind(' ', -20)
340
break_index = partline.rfind('-', -20)
343
break_index = partline.rfind('/', -20)
345
line = partline[break_index:] + line
346
partline = partline[:break_index]
349
partline = re.sub('\r', '\\\\r', partline)
353
elif re.search(' $', partline):
356
lines.append('# ' + partline + '\n')
362
def _patch_stanza_iter(line_iter):
367
return map[match.group(0)]
370
for line in line_iter:
371
if line.startswith('# '):
373
elif line.startswith('#'):
376
raise ValueError("bad line %r" % (line,))
377
if last_line is not None and len(line) > 2:
379
line = re.sub('\r', '', line)
380
line = re.sub('\\\\(.|\n)', mapget, line)
381
if last_line is None:
385
if last_line[-1] == '\n':
388
if last_line is not None:
392
def read_patch_stanza(line_iter):
393
"""Convert an iterable of RIO-Patch format lines into a Stanza.
395
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
396
It resists common forms of damage such as newline conversion or the removal
397
of trailing whitespace, yet is also reasonably easy to read.
401
return read_stanza(_patch_stanza_iter(line_iter))