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
1
# Copyright (C) 2005 by Canonical Ltd
3
# Distributed under the GNU General Public Licence v2
17
5
# \subsection{\emph{rio} - simple text metaformat}
19
7
# \emph{r} stands for `restricted', `reproducible', or `rfc822-like'.
21
9
# The stored data consists of a series of \emph{stanzas}, each of which contains
22
10
# \emph{fields} identified by an ascii name, with Unicode or string contents.
23
# The field tag is constrained to alphanumeric characters.
11
# The field tag is constrained to alphanumeric characters.
24
12
# There may be more than one field in a stanza with the same name.
26
14
# The format itself does not deal with character encoding issues, though
27
15
# the result will normally be written in Unicode.
29
17
# The format is intended to be simple enough that there is exactly one character
30
18
# stream representation of an object and vice versa, and that this relation
31
19
# will continue to hold for future versions of bzr.
21
# In comments, $\min(1,10)$
36
from .iterablefile import IterableFile
38
27
# XXX: some redundancy is allowing to write stanzas in isolation as well as
39
# through a writer object.
28
# through a writer object.
42
30
class RioWriter(object):
44
31
def __init__(self, to_file):
45
32
self._soft_nl = False
46
33
self._to_file = to_file
48
35
def write_stanza(self, stanza):
50
self._to_file.write(b'\n')
51
38
stanza.write(self._to_file)
52
39
self._soft_nl = True
55
42
class RioReader(object):
56
43
"""Read stanzas from a file as a sequence
58
to_file can be anything that can be enumerated as a sequence of
45
to_file can be anything that can be enumerated as a sequence of
59
46
lines (with newlines.)
62
48
def __init__(self, from_file):
63
49
self._from_file = from_file
131
91
def add(self, tag, value):
132
92
"""Append a name and value to the stanza."""
133
if not valid_tag(tag):
134
raise ValueError("invalid tag %r" % (tag,))
135
if isinstance(value, bytes):
136
value = value.decode('ascii')
137
elif isinstance(value, str):
93
assert valid_tag(tag), \
94
("invalid tag %r" % tag)
95
if isinstance(value, (str, unicode)):
97
## elif isinstance(value, (int, long)):
98
## value = str(value) # XXX: python2.4 without L-suffix
140
raise TypeError("invalid type for rio value: %r of type %s"
141
% (value, type(value)))
100
raise ValueError("invalid value %r" % value)
142
101
self.items.append((tag, value))
145
def from_pairs(cls, pairs):
150
103
def __contains__(self, find_tag):
151
104
"""True if there is any field in this stanza with the given tag."""
152
105
for tag, value in self.items:
174
127
return iter(self.items)
176
129
def to_lines(self):
177
"""Generate sequence of lines for external version of this file.
179
The lines are always utf-8 encoded strings.
130
"""Generate sequence of lines for external version of this file."""
181
131
if not self.items:
182
132
# max() complains if sequence is empty
185
for text_tag, text_value in self.items:
186
tag = text_tag.encode('ascii')
187
value = text_value.encode('utf-8', 'surrogateescape')
189
result.append(tag + b': \n')
135
for tag, value in self.items:
136
assert isinstance(value, (str, unicode))
138
result.append(tag + ': \n')
191
140
# don't want splitlines behaviour on empty lines
192
val_lines = value.split(b'\n')
193
result.append(tag + b': ' + val_lines[0] + b'\n')
141
val_lines = value.split('\n')
142
result.append(tag + ': ' + val_lines[0] + '\n')
194
143
for line in val_lines[1:]:
195
result.append(b'\t' + line + b'\n')
144
result.append('\t' + line + '\n')
197
result.append(tag + b': ' + value + b'\n')
146
result.append(tag + ': ' + value + '\n')
200
149
def to_string(self):
201
150
"""Return stanza as a single string"""
202
return b''.join(self.to_lines())
204
def to_unicode(self):
205
"""Return stanza as a single Unicode string.
207
This is most useful when adding a Stanza to a parent Stanza
213
for tag, value in self.items:
215
result.append(tag + u': \n')
217
# don't want splitlines behaviour on empty lines
218
val_lines = value.split(u'\n')
219
result.append(tag + u': ' + val_lines[0] + u'\n')
220
for line in val_lines[1:]:
221
result.append(u'\t' + line + u'\n')
223
result.append(tag + u': ' + value + u'\n')
224
return u''.join(result)
151
return ''.join(self.to_lines())
226
153
def write(self, to_file):
227
154
"""Write stanza to a file"""
252
"""Return a dict containing the unique values of the stanza.
255
for tag, value in self.items:
178
_tag_re = re.compile(r'^[-a-zA-Z0-9_]+$')
260
179
def valid_tag(tag):
261
return _valid_tag(tag)
180
return bool(_tag_re.match(tag))
264
183
def read_stanza(line_iter):
265
184
"""Return new Stanza read from list of lines or a file
267
186
Returns one Stanza that was read, or returns None at end of file. If a
268
187
blank line follows the stanza, it is consumed. It's not an error for
269
188
there to be no blank at end of file. If there is a blank file at the
270
start of the input this is really an empty stanza and that is returned.
189
start of the input this is really an empty stanza and that is returned.
272
191
Only the stanza lines and the trailing blank (if any) are consumed
273
192
from the line_iter.
275
The raw lines must be in utf-8 encoding.
277
return _read_stanza_utf8(line_iter)
280
def read_stanza_unicode(unicode_iter):
281
"""Read a Stanza from a list of lines or a file.
283
The lines should already be in unicode form. This returns a single
284
stanza that was read. If there is a blank line at the end of the Stanza,
285
it is consumed. It is not an error for there to be no blank line at
286
the end of the iterable. If there is a blank line at the beginning,
287
this is treated as an empty Stanza and None is returned.
289
Only the stanza lines and the trailing blank (if any) are consumed
290
from the unicode_iter
292
:param unicode_iter: A iterable, yeilding Unicode strings. See read_stanza
293
if you have a utf-8 encoded string.
294
:return: A Stanza object if there are any lines in the file.
297
return _read_stanza_unicode(unicode_iter)
300
def to_patch_lines(stanza, max_width=72):
301
"""Convert a stanza into RIO-Patch format lines.
303
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
304
It resists common forms of damage such as newline conversion or the removal
305
of trailing whitespace, yet is also reasonably easy to read.
307
:param max_width: The maximum number of characters per physical line.
308
:return: a list of lines
311
raise ValueError(max_width)
312
max_rio_width = max_width - 4
314
for pline in stanza.to_lines():
315
for line in pline.split(b'\n')[:-1]:
316
line = re.sub(b'\\\\', b'\\\\\\\\', line)
318
partline = line[:max_rio_width]
319
line = line[max_rio_width:]
320
if len(line) > 0 and line[:1] != [b' ']:
322
break_index = partline.rfind(b' ', -20)
324
break_index = partline.rfind(b'-', -20)
327
break_index = partline.rfind(b'/', -20)
329
line = partline[break_index:] + line
330
partline = partline[:break_index]
333
partline = re.sub(b'\r', b'\\\\r', partline)
337
elif re.search(b' $', partline):
340
lines.append(b'# ' + partline + b'\n')
342
lines.append(b'# \n')
346
def _patch_stanza_iter(line_iter):
347
map = {b'\\\\': b'\\',
352
return map[match.group(0)]
355
198
for line in line_iter:
356
if line.startswith(b'# '):
358
elif line.startswith(b'#'):
361
raise ValueError("bad line %r" % (line,))
362
if last_line is not None and len(line) > 2:
364
line = re.sub(b'\r', b'', line)
365
line = re.sub(b'\\\\(.|\n)', mapget, line)
366
if last_line is None:
370
if last_line[-1:] == b'\n':
373
if last_line is not None:
377
def read_patch_stanza(line_iter):
378
"""Convert an iterable of RIO-Patch format lines into a Stanza.
380
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
381
It resists common forms of damage such as newline conversion or the removal
382
of trailing whitespace, yet is also reasonably easy to read.
386
return read_stanza(_patch_stanza_iter(line_iter))
390
from ._rio_pyx import (
392
_read_stanza_unicode,
395
except ImportError as e:
396
osutils.failed_to_load_extension(e)
397
from ._rio_py import (
399
_read_stanza_unicode,
199
if line == None or line == '':
202
break # end of stanza
203
assert line[-1] == '\n'
205
if line[0] == '\t': # continues previous value
207
raise ValueError('invalid continuation line %r' % real_l)
208
accum_value += '\n' + line[1:-1]
209
else: # new tag:value line
211
stanza.add(tag, accum_value)
213
colon_index = line.index(': ')
215
raise ValueError('tag/value separator not found in line %r' % real_l)
216
tag = line[:colon_index]
217
assert valid_tag(tag), \
218
"invalid rio tag %r" % tag
219
accum_value = line[colon_index+2:-1]
220
if tag is not None: # add last tag-value
221
stanza.add(tag, accum_value)
223
else: # didn't see any content