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.
35
from bzrlib import osutils
23
36
from bzrlib.iterablefile import IterableFile
25
38
# XXX: some redundancy is allowing to write stanzas in isolation as well as
26
# through a writer object.
39
# through a writer object.
28
41
class RioWriter(object):
29
42
def __init__(self, to_file):
171
188
"""Return stanza as a single string"""
172
189
return ''.join(self.to_lines())
191
def to_unicode(self):
192
"""Return stanza as a single Unicode string.
194
This is most useful when adding a Stanza to a parent Stanza
200
for tag, value in self.items:
202
result.append(tag + u': \n')
204
# don't want splitlines behaviour on empty lines
205
val_lines = value.split(u'\n')
206
result.append(tag + u': ' + val_lines[0] + u'\n')
207
for line in val_lines[1:]:
208
result.append(u'\t' + line + u'\n')
210
result.append(tag + u': ' + value + u'\n')
211
return u''.join(result)
174
213
def write(self, to_file):
175
214
"""Write stanza to a file"""
176
215
to_file.writelines(self.to_lines())
203
242
for tag, value in self.items:
208
_tag_re = re.compile(r'^[-a-zA-Z0-9_]+$')
209
247
def valid_tag(tag):
210
return bool(_tag_re.match(tag))
248
return _valid_tag(tag)
213
251
def read_stanza(line_iter):
214
252
"""Return new Stanza read from list of lines or a file
216
254
Returns one Stanza that was read, or returns None at end of file. If a
217
255
blank line follows the stanza, it is consumed. It's not an error for
218
256
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.
257
start of the input this is really an empty stanza and that is returned.
221
259
Only the stanza lines and the trailing blank (if any) are consumed
222
260
from the line_iter.
224
262
The raw lines must be in utf-8 encoding.
264
return _read_stanza_utf8(line_iter)
267
def read_stanza_unicode(unicode_iter):
268
"""Read a Stanza from a list of lines or a file.
270
The lines should already be in unicode form. This returns a single
271
stanza that was read. If there is a blank line at the end of the Stanza,
272
it is consumed. It is not an error for there to be no blank line at
273
the end of the iterable. If there is a blank line at the beginning,
274
this is treated as an empty Stanza and None is returned.
276
Only the stanza lines and the trailing blank (if any) are consumed
277
from the unicode_iter
279
:param unicode_iter: A iterable, yeilding Unicode strings. See read_stanza
280
if you have a utf-8 encoded string.
281
:return: A Stanza object if there are any lines in the file.
284
return _read_stanza_unicode(unicode_iter)
287
def to_patch_lines(stanza, max_width=72):
288
"""Convert a stanza into RIO-Patch format lines.
290
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
291
It resists common forms of damage such as newline conversion or the removal
292
of trailing whitespace, yet is also reasonably easy to read.
294
:param max_width: The maximum number of characters per physical line.
295
:return: a list of lines
298
raise ValueError(max_width)
299
max_rio_width = max_width - 4
301
for pline in stanza.to_lines():
302
for line in pline.split('\n')[:-1]:
303
line = re.sub('\\\\', '\\\\\\\\', line)
305
partline = line[:max_rio_width]
306
line = line[max_rio_width:]
307
if len(line) > 0 and line[0] != [' ']:
309
break_index = partline.rfind(' ', -20)
311
break_index = partline.rfind('-', -20)
314
break_index = partline.rfind('/', -20)
316
line = partline[break_index:] + line
317
partline = partline[:break_index]
320
partline = re.sub('\r', '\\\\r', partline)
324
elif re.search(' $', partline):
327
lines.append('# ' + partline + '\n')
333
def _patch_stanza_iter(line_iter):
338
return map[match.group(0)]
230
341
for line in line_iter:
231
if line == None or line == '':
234
break # end of stanza
235
line = line.decode('utf-8')
236
assert line[-1] == '\n'
238
if line[0] == '\t': # continues previous value
240
raise ValueError('invalid continuation line %r' % real_l)
241
accum_value += '\n' + line[1:-1]
242
else: # new tag:value line
244
stanza.add(tag, accum_value)
246
colon_index = line.index(': ')
248
raise ValueError('tag/value separator not found in line %r' % real_l)
249
tag = str(line[:colon_index])
250
assert valid_tag(tag), \
251
"invalid rio tag %r" % tag
252
accum_value = line[colon_index+2:-1]
253
if tag is not None: # add last tag-value
254
stanza.add(tag, accum_value)
256
else: # didn't see any content
342
if line.startswith('# '):
344
elif line.startswith('#'):
347
raise ValueError("bad line %r" % (line,))
348
if last_line is not None and len(line) > 2:
350
line = re.sub('\r', '', line)
351
line = re.sub('\\\\(.|\n)', mapget, line)
352
if last_line is None:
356
if last_line[-1] == '\n':
359
if last_line is not None:
363
def read_patch_stanza(line_iter):
364
"""Convert an iterable of RIO-Patch format lines into a Stanza.
366
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
367
It resists common forms of damage such as newline conversion or the removal
368
of trailing whitespace, yet is also reasonably easy to read.
372
return read_stanza(_patch_stanza_iter(line_iter))
376
from bzrlib._rio_pyx import (
378
_read_stanza_unicode,
381
except ImportError, e:
382
osutils.failed_to_load_extension(e)
383
from bzrlib._rio_py import (
385
_read_stanza_unicode,