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., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
17
17
# \subsection{\emph{rio} - simple text metaformat}
19
19
# \emph{r} stands for `restricted', `reproducible', or `rfc822-like'.
21
21
# The stored data consists of a series of \emph{stanzas}, each of which contains
22
22
# \emph{fields} identified by an ascii name, with Unicode or string contents.
23
# The field tag is constrained to alphanumeric characters.
23
# The field tag is constrained to alphanumeric characters.
24
24
# There may be more than one field in a stanza with the same name.
26
26
# The format itself does not deal with character encoding issues, though
27
27
# the result will normally be written in Unicode.
29
29
# The format is intended to be simple enough that there is exactly one character
30
30
# stream representation of an object and vice versa, and that this relation
31
31
# will continue to hold for future versions of bzr.
35
from bzrlib import osutils
35
36
from bzrlib.iterablefile import IterableFile
37
38
# XXX: some redundancy is allowing to write stanzas in isolation as well as
38
# through a writer object.
39
# through a writer object.
40
41
class RioWriter(object):
41
42
def __init__(self, to_file):
52
53
class RioReader(object):
53
54
"""Read stanzas from a file as a sequence
55
to_file can be anything that can be enumerated as a sequence of
56
to_file can be anything that can be enumerated as a sequence of
56
57
lines (with newlines.)
58
59
def __init__(self, from_file):
93
94
class Stanza(object):
94
95
"""One stanza for rio.
96
Each stanza contains a set of named fields.
97
Each stanza contains a set of named fields.
98
99
Names must be non-empty ascii alphanumeric plus _. Names can be repeated
99
100
within a stanza. Names are case-sensitive. The ordering of fields is
117
118
def add(self, tag, value):
118
119
"""Append a name and value to the stanza."""
119
assert valid_tag(tag), \
120
("invalid tag %r" % tag)
120
if not valid_tag(tag):
121
raise ValueError("invalid tag %r" % (tag,))
121
122
if isinstance(value, str):
122
123
value = unicode(value)
123
124
elif isinstance(value, unicode):
128
129
raise TypeError("invalid type for rio value: %r of type %s"
129
130
% (value, type(value)))
130
131
self.items.append((tag, value))
134
def from_pairs(cls, pairs):
132
139
def __contains__(self, find_tag):
133
140
"""True if there is any field in this stanza with the given tag."""
134
141
for tag, value in self.items:
167
174
for tag, value in self.items:
168
assert isinstance(tag, str), type(tag)
169
assert isinstance(value, unicode)
171
176
result.append(tag + ': \n')
172
177
elif '\n' in value:
195
200
for tag, value in self.items:
197
result.append(tag + ': \n')
202
result.append(tag + u': \n')
199
204
# don't want splitlines behaviour on empty lines
200
val_lines = value.split('\n')
201
result.append(tag + ': ' + val_lines[0] + '\n')
205
val_lines = value.split(u'\n')
206
result.append(tag + u': ' + val_lines[0] + u'\n')
202
207
for line in val_lines[1:]:
203
result.append('\t' + line + '\n')
208
result.append(u'\t' + line + u'\n')
205
result.append(tag + ': ' + value + '\n')
210
result.append(tag + u': ' + value + u'\n')
206
211
return u''.join(result)
208
213
def write(self, to_file):
237
242
for tag, value in self.items:
242
_tag_re = re.compile(r'^[-a-zA-Z0-9_]+$')
243
247
def valid_tag(tag):
244
return bool(_tag_re.match(tag))
248
return _valid_tag(tag)
247
251
def read_stanza(line_iter):
248
252
"""Return new Stanza read from list of lines or a file
250
254
Returns one Stanza that was read, or returns None at end of file. If a
251
255
blank line follows the stanza, it is consumed. It's not an error for
252
256
there to be no blank at end of file. If there is a blank file at the
253
start of the input this is really an empty stanza and that is returned.
257
start of the input this is really an empty stanza and that is returned.
255
259
Only the stanza lines and the trailing blank (if any) are consumed
256
260
from the line_iter.
258
262
The raw lines must be in utf-8 encoding.
260
unicode_iter = (line.decode('utf-8') for line in line_iter)
261
return read_stanza_unicode(unicode_iter)
264
return _read_stanza_utf8(line_iter)
264
267
def read_stanza_unicode(unicode_iter):
278
281
:return: A Stanza object if there are any lines in the file.
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
284
return _read_stanza_unicode(unicode_iter)
320
287
def to_patch_lines(stanza, max_width=72):
373
341
for line in line_iter:
374
342
if line.startswith('# '):
377
assert line.startswith('#')
344
elif line.startswith('#'):
347
raise ValueError("bad line %r" % (line,))
379
348
if last_line is not None and len(line) > 2:
381
350
line = re.sub('\r', '', line)