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
1
# Copyright (C) 2005 by Canonical Ltd
3
# Distributed under the GNU General Public Licence v2
19
5
# \subsection{\emph{rio} - simple text metaformat}
21
7
# \emph{r} stands for `restricted', `reproducible', or `rfc822-like'.
23
9
# The stored data consists of a series of \emph{stanzas}, each of which contains
24
10
# \emph{fields} identified by an ascii name, with Unicode or string contents.
25
# The field tag is constrained to alphanumeric characters.
11
# The field tag is constrained to alphanumeric characters.
26
12
# There may be more than one field in a stanza with the same name.
28
14
# The format itself does not deal with character encoding issues, though
29
15
# the result will normally be written in Unicode.
31
17
# The format is intended to be simple enough that there is exactly one character
32
18
# stream representation of an object and vice versa, and that this relation
33
19
# will continue to hold for future versions of bzr.
38
from .iterablefile import IterableFile
23
from bzrlib.iterablefile import IterableFile
43
25
# XXX: some redundancy is allowing to write stanzas in isolation as well as
44
# through a writer object.
26
# through a writer object.
47
28
class RioWriter(object):
49
29
def __init__(self, to_file):
50
30
self._soft_nl = False
51
31
self._to_file = to_file
53
33
def write_stanza(self, stanza):
55
self._to_file.write(b'\n')
56
36
stanza.write(self._to_file)
57
37
self._soft_nl = True
60
40
class RioReader(object):
61
41
"""Read stanzas from a file as a sequence
63
to_file can be anything that can be enumerated as a sequence of
43
to_file can be anything that can be enumerated as a sequence of
64
44
lines (with newlines.)
67
46
def __init__(self, from_file):
68
47
self._from_file = from_file
136
105
def add(self, tag, value):
137
106
"""Append a name and value to the stanza."""
138
if not valid_tag(tag):
139
raise ValueError("invalid tag %r" % (tag,))
140
if isinstance(value, bytes):
141
value = value.decode('ascii')
142
elif isinstance(value, text_type):
107
assert valid_tag(tag), \
108
("invalid tag %r" % tag)
109
if isinstance(value, str):
110
value = unicode(value)
111
elif isinstance(value, unicode):
113
## elif isinstance(value, (int, long)):
114
## value = str(value) # XXX: python2.4 without L-suffix
145
116
raise TypeError("invalid type for rio value: %r of type %s"
146
117
% (value, type(value)))
147
118
self.items.append((tag, value))
150
def from_pairs(cls, pairs):
155
120
def __contains__(self, find_tag):
156
121
"""True if there is any field in this stanza with the given tag."""
157
122
for tag, value in self.items:
181
146
def to_lines(self):
182
147
"""Generate sequence of lines for external version of this file.
184
149
The lines are always utf-8 encoded strings.
186
151
if not self.items:
187
152
# max() complains if sequence is empty
190
for text_tag, text_value in self.items:
191
tag = text_tag.encode('ascii')
192
value = text_value.encode('utf-8')
194
result.append(tag + b': \n')
155
for tag, value in self.items:
156
assert isinstance(tag, str), type(tag)
157
assert isinstance(value, unicode)
159
result.append(tag + ': \n')
196
161
# don't want splitlines behaviour on empty lines
197
val_lines = value.split(b'\n')
198
result.append(tag + b': ' + val_lines[0] + b'\n')
162
val_lines = value.split('\n')
163
result.append(tag + ': ' + val_lines[0].encode('utf-8') + '\n')
199
164
for line in val_lines[1:]:
200
result.append(b'\t' + line + b'\n')
165
result.append('\t' + line.encode('utf-8') + '\n')
202
result.append(tag + b': ' + value + b'\n')
167
result.append(tag + ': ' + value.encode('utf-8') + '\n')
205
170
def to_string(self):
206
171
"""Return stanza as a single string"""
207
return b''.join(self.to_lines())
209
def to_unicode(self):
210
"""Return stanza as a single Unicode string.
212
This is most useful when adding a Stanza to a parent Stanza
218
for tag, value in self.items:
220
result.append(tag + u': \n')
222
# don't want splitlines behaviour on empty lines
223
val_lines = value.split(u'\n')
224
result.append(tag + u': ' + val_lines[0] + u'\n')
225
for line in val_lines[1:]:
226
result.append(u'\t' + line + u'\n')
228
result.append(tag + u': ' + value + u'\n')
229
return u''.join(result)
172
return ''.join(self.to_lines())
231
174
def write(self, to_file):
232
175
"""Write stanza to a file"""
260
203
for tag, value in self.items:
208
_tag_re = re.compile(r'^[-a-zA-Z0-9_]+$')
265
209
def valid_tag(tag):
266
return _valid_tag(tag)
210
return bool(_tag_re.match(tag))
269
213
def read_stanza(line_iter):
270
214
"""Return new Stanza read from list of lines or a file
272
216
Returns one Stanza that was read, or returns None at end of file. If a
273
217
blank line follows the stanza, it is consumed. It's not an error for
274
218
there to be no blank at end of file. If there is a blank file at the
275
start of the input this is really an empty stanza and that is returned.
219
start of the input this is really an empty stanza and that is returned.
277
221
Only the stanza lines and the trailing blank (if any) are consumed
278
222
from the line_iter.
280
224
The raw lines must be in utf-8 encoding.
282
return _read_stanza_utf8(line_iter)
285
def read_stanza_unicode(unicode_iter):
286
"""Read a Stanza from a list of lines or a file.
288
The lines should already be in unicode form. This returns a single
289
stanza that was read. If there is a blank line at the end of the Stanza,
290
it is consumed. It is not an error for there to be no blank line at
291
the end of the iterable. If there is a blank line at the beginning,
292
this is treated as an empty Stanza and None is returned.
294
Only the stanza lines and the trailing blank (if any) are consumed
295
from the unicode_iter
297
:param unicode_iter: A iterable, yeilding Unicode strings. See read_stanza
298
if you have a utf-8 encoded string.
299
:return: A Stanza object if there are any lines in the file.
302
return _read_stanza_unicode(unicode_iter)
305
def to_patch_lines(stanza, max_width=72):
306
"""Convert a stanza into RIO-Patch format lines.
308
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
309
It resists common forms of damage such as newline conversion or the removal
310
of trailing whitespace, yet is also reasonably easy to read.
312
:param max_width: The maximum number of characters per physical line.
313
:return: a list of lines
316
raise ValueError(max_width)
317
max_rio_width = max_width - 4
319
for pline in stanza.to_lines():
320
for line in pline.split(b'\n')[:-1]:
321
line = re.sub(b'\\\\', b'\\\\\\\\', line)
323
partline = line[:max_rio_width]
324
line = line[max_rio_width:]
325
if len(line) > 0 and line[:1] != [b' ']:
327
break_index = partline.rfind(b' ', -20)
329
break_index = partline.rfind(b'-', -20)
332
break_index = partline.rfind(b'/', -20)
334
line = partline[break_index:] + line
335
partline = partline[:break_index]
338
partline = re.sub(b'\r', b'\\\\r', partline)
342
elif re.search(b' $', partline):
345
lines.append(b'# ' + partline + b'\n')
347
lines.append(b'# \n')
351
def _patch_stanza_iter(line_iter):
352
map = {b'\\\\': b'\\',
357
return map[match.group(0)]
360
230
for line in line_iter:
361
if line.startswith(b'# '):
363
elif line.startswith(b'#'):
366
raise ValueError("bad line %r" % (line,))
367
if last_line is not None and len(line) > 2:
369
line = re.sub(b'\r', b'', line)
370
line = re.sub(b'\\\\(.|\n)', mapget, line)
371
if last_line is None:
375
if last_line[-1:] == b'\n':
378
if last_line is not None:
382
def read_patch_stanza(line_iter):
383
"""Convert an iterable of RIO-Patch format lines into a Stanza.
385
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
386
It resists common forms of damage such as newline conversion or the removal
387
of trailing whitespace, yet is also reasonably easy to read.
391
return read_stanza(_patch_stanza_iter(line_iter))
395
from ._rio_pyx import (
397
_read_stanza_unicode,
400
except ImportError as e:
401
osutils.failed_to_load_extension(e)
402
from ._rio_py import (
404
_read_stanza_unicode,
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