13
13
# You should have received a copy of the GNU General Public License
14
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
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
19
17
# \subsection{\emph{rio} - simple text metaformat}
21
19
# \emph{r} stands for `restricted', `reproducible', or `rfc822-like'.
23
21
# The stored data consists of a series of \emph{stanzas}, each of which contains
24
22
# \emph{fields} identified by an ascii name, with Unicode or string contents.
25
# The field tag is constrained to alphanumeric characters.
23
# The field tag is constrained to alphanumeric characters.
26
24
# There may be more than one field in a stanza with the same name.
28
26
# The format itself does not deal with character encoding issues, though
29
27
# the result will normally be written in Unicode.
31
29
# The format is intended to be simple enough that there is exactly one character
32
30
# stream representation of an object and vice versa, and that this relation
33
31
# will continue to hold for future versions of bzr.
38
from .iterablefile import IterableFile
35
from bzrlib.iterablefile import IterableFile
43
37
# XXX: some redundancy is allowing to write stanzas in isolation as well as
44
# through a writer object.
38
# through a writer object.
46
40
class RioWriter(object):
47
41
def __init__(self, to_file):
51
45
def write_stanza(self, stanza):
53
self._to_file.write(b'\n')
54
48
stanza.write(self._to_file)
55
49
self._soft_nl = True
58
52
class RioReader(object):
59
53
"""Read stanzas from a file as a sequence
61
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
62
56
lines (with newlines.)
64
58
def __init__(self, from_file):
123
117
def add(self, tag, value):
124
118
"""Append a name and value to the stanza."""
125
if not valid_tag(tag):
126
raise ValueError("invalid tag %r" % (tag,))
127
if isinstance(value, bytes):
128
value = value.decode('ascii')
129
elif isinstance(value, text_type):
119
assert valid_tag(tag), \
120
("invalid tag %r" % tag)
121
if isinstance(value, str):
122
value = unicode(value)
123
elif isinstance(value, unicode):
125
## elif isinstance(value, (int, long)):
126
## value = str(value) # XXX: python2.4 without L-suffix
132
128
raise TypeError("invalid type for rio value: %r of type %s"
133
129
% (value, type(value)))
134
130
self.items.append((tag, value))
137
def from_pairs(cls, pairs):
142
132
def __contains__(self, find_tag):
143
133
"""True if there is any field in this stanza with the given tag."""
144
134
for tag, value in self.items:
168
158
def to_lines(self):
169
159
"""Generate sequence of lines for external version of this file.
171
161
The lines are always utf-8 encoded strings.
173
163
if not self.items:
174
164
# max() complains if sequence is empty
177
for text_tag, text_value in self.items:
178
tag = text_tag.encode('ascii')
179
value = text_value.encode('utf-8')
181
result.append(tag + b': \n')
167
for tag, value in self.items:
168
assert isinstance(tag, str), type(tag)
169
assert isinstance(value, unicode)
171
result.append(tag + ': \n')
183
173
# don't want splitlines behaviour on empty lines
184
val_lines = value.split(b'\n')
185
result.append(tag + b': ' + val_lines[0] + b'\n')
174
val_lines = value.split('\n')
175
result.append(tag + ': ' + val_lines[0].encode('utf-8') + '\n')
186
176
for line in val_lines[1:]:
187
result.append(b'\t' + line + b'\n')
177
result.append('\t' + line.encode('utf-8') + '\n')
189
result.append(tag + b': ' + value + b'\n')
179
result.append(tag + ': ' + value.encode('utf-8') + '\n')
192
182
def to_string(self):
193
183
"""Return stanza as a single string"""
194
return b''.join(self.to_lines())
184
return ''.join(self.to_lines())
196
186
def to_unicode(self):
197
187
"""Return stanza as a single Unicode string.
205
195
for tag, value in self.items:
207
result.append(tag + u': \n')
197
result.append(tag + ': \n')
209
199
# don't want splitlines behaviour on empty lines
210
val_lines = value.split(u'\n')
211
result.append(tag + u': ' + val_lines[0] + u'\n')
200
val_lines = value.split('\n')
201
result.append(tag + ': ' + val_lines[0] + '\n')
212
202
for line in val_lines[1:]:
213
result.append(u'\t' + line + u'\n')
203
result.append('\t' + line + '\n')
215
result.append(tag + u': ' + value + u'\n')
205
result.append(tag + ': ' + value + '\n')
216
206
return u''.join(result)
218
208
def write(self, to_file):
247
237
for tag, value in self.items:
242
_tag_re = re.compile(r'^[-a-zA-Z0-9_]+$')
252
243
def valid_tag(tag):
253
return _valid_tag(tag)
244
return bool(_tag_re.match(tag))
256
247
def read_stanza(line_iter):
257
248
"""Return new Stanza read from list of lines or a file
259
250
Returns one Stanza that was read, or returns None at end of file. If a
260
251
blank line follows the stanza, it is consumed. It's not an error for
261
252
there to be no blank at end of file. If there is a blank file at the
262
start of the input this is really an empty stanza and that is returned.
253
start of the input this is really an empty stanza and that is returned.
264
255
Only the stanza lines and the trailing blank (if any) are consumed
265
256
from the line_iter.
267
258
The raw lines must be in utf-8 encoding.
269
return _read_stanza_utf8(line_iter)
260
unicode_iter = (line.decode('utf-8') for line in line_iter)
261
return read_stanza_unicode(unicode_iter)
272
264
def read_stanza_unicode(unicode_iter):
286
278
:return: A Stanza object if there are any lines in the file.
289
return _read_stanza_unicode(unicode_iter)
285
# TODO: jam 20060922 This code should raise real errors rather than
286
# using 'assert' to process user input, or raising ValueError
287
# rather than a more specific error.
289
for line in unicode_iter:
290
if line is None or line == '':
293
break # end of stanza
294
assert line.endswith('\n')
296
if line[0] == '\t': # continues previous value
298
raise ValueError('invalid continuation line %r' % real_l)
299
accum_value += '\n' + line[1:-1]
300
else: # new tag:value line
302
stanza.add(tag, accum_value)
304
colon_index = line.index(': ')
306
raise ValueError('tag/value separator not found in line %r'
308
tag = str(line[:colon_index])
309
assert valid_tag(tag), \
310
"invalid rio tag %r" % tag
311
accum_value = line[colon_index+2:-1]
313
if tag is not None: # add last tag-value
314
stanza.add(tag, accum_value)
316
else: # didn't see any content
292
320
def to_patch_lines(stanza, max_width=72):
299
327
:param max_width: The maximum number of characters per physical line.
300
328
:return: a list of lines
303
raise ValueError(max_width)
304
331
max_rio_width = max_width - 4
306
333
for pline in stanza.to_lines():
307
for line in pline.split(b'\n')[:-1]:
308
line = re.sub(b'\\\\', b'\\\\\\\\', line)
334
for line in pline.split('\n')[:-1]:
335
line = re.sub('\\\\', '\\\\\\\\', line)
309
336
while len(line) > 0:
310
337
partline = line[:max_rio_width]
311
338
line = line[max_rio_width:]
312
if len(line) > 0 and line[:1] != [b' ']:
339
if len(line) > 0 and line[0] != [' ']:
314
break_index = partline.rfind(b' ', -20)
341
break_index = partline.rfind(' ', -20)
315
342
if break_index < 3:
316
break_index = partline.rfind(b'-', -20)
343
break_index = partline.rfind('-', -20)
318
345
if break_index < 3:
319
break_index = partline.rfind(b'/', -20)
346
break_index = partline.rfind('/', -20)
320
347
if break_index >= 3:
321
348
line = partline[break_index:] + line
322
349
partline = partline[:break_index]
323
350
if len(line) > 0:
325
partline = re.sub(b'\r', b'\\\\r', partline)
352
partline = re.sub('\r', '\\\\r', partline)
326
353
blank_line = False
327
354
if len(line) > 0:
329
elif re.search(b' $', partline):
356
elif re.search(' $', partline):
331
358
blank_line = True
332
lines.append(b'# ' + partline + b'\n')
359
lines.append('# ' + partline + '\n')
334
lines.append(b'# \n')
338
365
def _patch_stanza_iter(line_iter):
339
map = {b'\\\\': b'\\',
342
369
def mapget(match):
343
370
return map[match.group(0)]
346
373
for line in line_iter:
347
if line.startswith(b'# '):
374
if line.startswith('# '):
349
elif line.startswith(b'#'):
377
assert line.startswith('#')
352
raise ValueError("bad line %r" % (line,))
353
379
if last_line is not None and len(line) > 2:
355
line = re.sub(b'\r', b'', line)
356
line = re.sub(b'\\\\(.|\n)', mapget, line)
381
line = re.sub('\r', '', line)
382
line = re.sub('\\\\(.|\n)', mapget, line)
357
383
if last_line is None:
360
386
last_line += line
361
if last_line[-1:] == b'\n':
387
if last_line[-1] == '\n':
364
390
if last_line is not None: