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
# \subsection{\emph{rio} - simple text metaformat}
19
# \emph{r} stands for `restricted', `reproducible', or `rfc822-like'.
21
# The stored data consists of a series of \emph{stanzas}, each of which contains
22
# \emph{fields} identified by an ascii name, with Unicode or string contents.
23
# The field tag is constrained to alphanumeric characters.
24
# There may be more than one field in a stanza with the same name.
26
# The format itself does not deal with character encoding issues, though
27
# the result will normally be written in Unicode.
29
# The format is intended to be simple enough that there is exactly one character
30
# stream representation of an object and vice versa, and that this relation
31
# will continue to hold for future versions of bzr.
35
from bzrlib.iterablefile import IterableFile
37
# XXX: some redundancy is allowing to write stanzas in isolation as well as
38
# through a writer object.
40
class RioWriter(object):
41
def __init__(self, to_file):
43
self._to_file = to_file
45
def write_stanza(self, stanza):
47
self._to_file.write('\n')
48
stanza.write(self._to_file)
52
class RioReader(object):
53
"""Read stanzas from a file as a sequence
55
to_file can be anything that can be enumerated as a sequence of
56
lines (with newlines.)
58
def __init__(self, from_file):
59
self._from_file = from_file
63
s = read_stanza(self._from_file)
70
def rio_file(stanzas, header=None):
71
"""Produce a rio IterableFile from an iterable of stanzas"""
73
if header is not None:
77
if first_stanza is not True:
79
for line in s.to_lines():
82
return IterableFile(str_iter())
85
def read_stanzas(from_file):
87
s = read_stanza(from_file)
94
"""One stanza for rio.
96
Each stanza contains a set of named fields.
98
Names must be non-empty ascii alphanumeric plus _. Names can be repeated
99
within a stanza. Names are case-sensitive. The ordering of fields is
102
Each field value must be either an int or a string.
105
__slots__ = ['items']
107
def __init__(self, **kwargs):
108
"""Construct a new Stanza.
110
The keyword arguments, if any, are added in sorted order to the stanza.
114
for tag, value in sorted(kwargs.items()):
117
def add(self, tag, value):
118
"""Append a name and value to the stanza."""
119
if not valid_tag(tag):
120
raise ValueError("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
128
raise TypeError("invalid type for rio value: %r of type %s"
129
% (value, type(value)))
130
self.items.append((tag, value))
132
def __contains__(self, find_tag):
133
"""True if there is any field in this stanza with the given tag."""
134
for tag, value in self.items:
140
"""Return number of pairs in the stanza."""
141
return len(self.items)
143
def __eq__(self, other):
144
if not isinstance(other, Stanza):
146
return self.items == other.items
148
def __ne__(self, other):
149
return not self.__eq__(other)
152
return "Stanza(%r)" % self.items
154
def iter_pairs(self):
155
"""Return iterator of tag, value pairs."""
156
return iter(self.items)
159
"""Generate sequence of lines for external version of this file.
161
The lines are always utf-8 encoded strings.
164
# max() complains if sequence is empty
167
for tag, value in self.items:
169
result.append(tag + ': \n')
171
# don't want splitlines behaviour on empty lines
172
val_lines = value.split('\n')
173
result.append(tag + ': ' + val_lines[0].encode('utf-8') + '\n')
174
for line in val_lines[1:]:
175
result.append('\t' + line.encode('utf-8') + '\n')
177
result.append(tag + ': ' + value.encode('utf-8') + '\n')
181
"""Return stanza as a single string"""
182
return ''.join(self.to_lines())
184
def to_unicode(self):
185
"""Return stanza as a single Unicode string.
187
This is most useful when adding a Stanza to a parent Stanza
193
for tag, value in self.items:
195
result.append(tag + ': \n')
197
# don't want splitlines behaviour on empty lines
198
val_lines = value.split('\n')
199
result.append(tag + ': ' + val_lines[0] + '\n')
200
for line in val_lines[1:]:
201
result.append('\t' + line + '\n')
203
result.append(tag + ': ' + value + '\n')
204
return u''.join(result)
206
def write(self, to_file):
207
"""Write stanza to a file"""
208
to_file.writelines(self.to_lines())
211
"""Return the value for a field wih given tag.
213
If there is more than one value, only the first is returned. If the
214
tag is not present, KeyError is raised.
216
for t, v in self.items:
224
def get_all(self, tag):
226
for t, v in self.items:
232
"""Return a dict containing the unique values of the stanza.
235
for tag, value in self.items:
239
_tag_re = re.compile(r'^[-a-zA-Z0-9_]+$')
241
return bool(_tag_re.match(tag))
244
def read_stanza(line_iter):
245
"""Return new Stanza read from list of lines or a file
247
Returns one Stanza that was read, or returns None at end of file. If a
248
blank line follows the stanza, it is consumed. It's not an error for
249
there to be no blank at end of file. If there is a blank file at the
250
start of the input this is really an empty stanza and that is returned.
252
Only the stanza lines and the trailing blank (if any) are consumed
255
The raw lines must be in utf-8 encoding.
257
unicode_iter = (line.decode('utf-8') for line in line_iter)
258
return read_stanza_unicode(unicode_iter)
261
def read_stanza_unicode(unicode_iter):
262
"""Read a Stanza from a list of lines or a file.
264
The lines should already be in unicode form. This returns a single
265
stanza that was read. If there is a blank line at the end of the Stanza,
266
it is consumed. It is not an error for there to be no blank line at
267
the end of the iterable. If there is a blank line at the beginning,
268
this is treated as an empty Stanza and None is returned.
270
Only the stanza lines and the trailing blank (if any) are consumed
271
from the unicode_iter
273
:param unicode_iter: A iterable, yeilding Unicode strings. See read_stanza
274
if you have a utf-8 encoded string.
275
:return: A Stanza object if there are any lines in the file.
282
# TODO: jam 20060922 This code should raise real errors rather than
283
# using 'assert' to process user input, or raising ValueError
284
# rather than a more specific error.
286
for line in unicode_iter:
287
if line is None or line == '':
290
break # end of stanza
292
if line[0] == '\t': # continues previous value
294
raise ValueError('invalid continuation line %r' % real_l)
295
accum_value += '\n' + line[1:-1]
296
else: # new tag:value line
298
stanza.add(tag, accum_value)
300
colon_index = line.index(': ')
302
raise ValueError('tag/value separator not found in line %r'
304
tag = str(line[:colon_index])
305
if not valid_tag(tag):
306
raise ValueError("invalid rio tag %r" % (tag,))
307
accum_value = line[colon_index+2:-1]
309
if tag is not None: # add last tag-value
310
stanza.add(tag, accum_value)
312
else: # didn't see any content
316
def to_patch_lines(stanza, max_width=72):
317
"""Convert a stanza into RIO-Patch format lines.
319
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
320
It resists common forms of damage such as newline conversion or the removal
321
of trailing whitespace, yet is also reasonably easy to read.
323
:param max_width: The maximum number of characters per physical line.
324
:return: a list of lines
327
raise ValueError(max_width)
328
max_rio_width = max_width - 4
330
for pline in stanza.to_lines():
331
for line in pline.split('\n')[:-1]:
332
line = re.sub('\\\\', '\\\\\\\\', line)
334
partline = line[:max_rio_width]
335
line = line[max_rio_width:]
336
if len(line) > 0 and line[0] != [' ']:
338
break_index = partline.rfind(' ', -20)
340
break_index = partline.rfind('-', -20)
343
break_index = partline.rfind('/', -20)
345
line = partline[break_index:] + line
346
partline = partline[:break_index]
349
partline = re.sub('\r', '\\\\r', partline)
353
elif re.search(' $', partline):
356
lines.append('# ' + partline + '\n')
362
def _patch_stanza_iter(line_iter):
367
return map[match.group(0)]
370
for line in line_iter:
371
if line.startswith('# '):
373
elif line.startswith('#'):
376
raise ValueError("bad line %r" % (line,))
377
if last_line is not None and len(line) > 2:
379
line = re.sub('\r', '', line)
380
line = re.sub('\\\\(.|\n)', mapget, line)
381
if last_line is None:
385
if last_line[-1] == '\n':
388
if last_line is not None:
392
def read_patch_stanza(line_iter):
393
"""Convert an iterable of RIO-Patch format lines into a Stanza.
395
RIO-Patch is a RIO variant designed to be e-mailed as part of a patch.
396
It resists common forms of damage such as newline conversion or the removal
397
of trailing whitespace, yet is also reasonably easy to read.
401
return read_stanza(_patch_stanza_iter(line_iter))
added
removed
bzrlib/rio.py
