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
17
from __future__ import absolute_import
5
19
# \subsection{\emph{rio} - simple text metaformat}
7
21
# \emph{r} stands for `restricted', `reproducible', or `rfc822-like'.
9
23
# The stored data consists of a series of \emph{stanzas}, each of which contains
10
24
# \emph{fields} identified by an ascii name, with Unicode or string contents.
11
# The field tag is constrained to alphanumeric characters.
25
# The field tag is constrained to alphanumeric characters.
12
26
# There may be more than one field in a stanza with the same name.
14
28
# The format itself does not deal with character encoding issues, though
15
29
# the result will normally be written in Unicode.
17
31
# The format is intended to be simple enough that there is exactly one character
18
32
# stream representation of an object and vice versa, and that this relation
19
33
# will continue to hold for future versions of bzr.
37
from bzrlib import osutils
23
38
from bzrlib.iterablefile import IterableFile
25
40
# XXX: some redundancy is allowing to write stanzas in isolation as well as
26
# through a writer object.
41
# through a writer object.
28
43
class RioWriter(object):
29
44
def __init__(self, to_file):
183
202
for tag, value in self.items:
185
result.append(tag + ': \n')
204
result.append(tag + u': \n')
187
206
# don't want splitlines behaviour on empty lines
188
val_lines = value.split('\n')
189
result.append(tag + ': ' + val_lines[0] + '\n')
207
val_lines = value.split(u'\n')
208
result.append(tag + u': ' + val_lines[0] + u'\n')
190
209
for line in val_lines[1:]:
191
result.append('\t' + line + '\n')
210
result.append(u'\t' + line + u'\n')
193
result.append(tag + ': ' + value + '\n')
212
result.append(tag + u': ' + value + u'\n')
194
213
return u''.join(result)
196
215
def write(self, to_file):
225
244
for tag, value in self.items:
230
_tag_re = re.compile(r'^[-a-zA-Z0-9_]+$')
231
249
def valid_tag(tag):
232
return bool(_tag_re.match(tag))
250
return _valid_tag(tag)
235
253
def read_stanza(line_iter):
236
254
"""Return new Stanza read from list of lines or a file
238
256
Returns one Stanza that was read, or returns None at end of file. If a
239
257
blank line follows the stanza, it is consumed. It's not an error for
240
258
there to be no blank at end of file. If there is a blank file at the
241
start of the input this is really an empty stanza and that is returned.
259
start of the input this is really an empty stanza and that is returned.
243
261
Only the stanza lines and the trailing blank (if any) are consumed
244
262
from the line_iter.
246
264
The raw lines must be in utf-8 encoding.
248
unicode_iter = (line.decode('utf-8') for line in line_iter)
249
return read_stanza_unicode(unicode_iter)
266
return _read_stanza_utf8(line_iter)
252
269
def read_stanza_unicode(unicode_iter):
266
283
:return: A Stanza object if there are any lines in the file.
273
# TODO: jam 20060922 This code should raise real errors rather than
274
# using 'assert' to process user input, or raising ValueError
275
# rather than a more specific error.
277
for line in unicode_iter:
278
if line is None or line == '':
281
break # end of stanza
282
assert line.endswith('\n')
284
if line[0] == '\t': # continues previous value
286
raise ValueError('invalid continuation line %r' % real_l)
287
accum_value += '\n' + line[1:-1]
288
else: # new tag:value line
290
stanza.add(tag, accum_value)
292
colon_index = line.index(': ')
294
raise ValueError('tag/value separator not found in line %r'
296
tag = str(line[:colon_index])
297
assert valid_tag(tag), \
298
"invalid rio tag %r" % tag
299
accum_value = line[colon_index+2:-1]
301
if tag is not None: # add last tag-value
302
stanza.add(tag, accum_value)
304
else: # didn't see any content
286
return _read_stanza_unicode(unicode_iter)
289
def to_patch_lines(stanza, max_width=72):
290
"""Convert a stanza into RIO-Patch format lines.
292
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
293
It resists common forms of damage such as newline conversion or the removal
294
of trailing whitespace, yet is also reasonably easy to read.
296
:param max_width: The maximum number of characters per physical line.
297
:return: a list of lines
300
raise ValueError(max_width)
301
max_rio_width = max_width - 4
303
for pline in stanza.to_lines():
304
for line in pline.split('\n')[:-1]:
305
line = re.sub('\\\\', '\\\\\\\\', line)
307
partline = line[:max_rio_width]
308
line = line[max_rio_width:]
309
if len(line) > 0 and line[0] != [' ']:
311
break_index = partline.rfind(' ', -20)
313
break_index = partline.rfind('-', -20)
316
break_index = partline.rfind('/', -20)
318
line = partline[break_index:] + line
319
partline = partline[:break_index]
322
partline = re.sub('\r', '\\\\r', partline)
326
elif re.search(' $', partline):
329
lines.append('# ' + partline + '\n')
335
def _patch_stanza_iter(line_iter):
340
return map[match.group(0)]
343
for line in line_iter:
344
if line.startswith('# '):
346
elif line.startswith('#'):
349
raise ValueError("bad line %r" % (line,))
350
if last_line is not None and len(line) > 2:
352
line = re.sub('\r', '', line)
353
line = re.sub('\\\\(.|\n)', mapget, line)
354
if last_line is None:
358
if last_line[-1] == '\n':
361
if last_line is not None:
365
def read_patch_stanza(line_iter):
366
"""Convert an iterable of RIO-Patch format lines into a Stanza.
368
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
369
It resists common forms of damage such as newline conversion or the removal
370
of trailing whitespace, yet is also reasonably easy to read.
374
return read_stanza(_patch_stanza_iter(line_iter))
378
from bzrlib._rio_pyx import (
380
_read_stanza_unicode,
383
except ImportError, e:
384
osutils.failed_to_load_extension(e)
385
from bzrlib._rio_py import (
387
_read_stanza_unicode,