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
35
from bzrlib.iterablefile import IterableFile
37
37
# XXX: some redundancy is allowing to write stanzas in isolation as well as
38
# through a writer object.
38
# through a writer object.
40
40
class RioWriter(object):
41
41
def __init__(self, to_file):
45
45
def write_stanza(self, stanza):
47
self._to_file.write('\n')
48
48
stanza.write(self._to_file)
49
49
self._soft_nl = True
52
52
class RioReader(object):
53
53
"""Read stanzas from a file as a sequence
55
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
56
56
lines (with newlines.)
58
58
def __init__(self, from_file):
93
93
class Stanza(object):
94
94
"""One stanza for rio.
96
Each stanza contains a set of named fields.
96
Each stanza contains a set of named fields.
98
98
Names must be non-empty ascii alphanumeric plus _. Names can be repeated
99
99
within a stanza. Names are case-sensitive. The ordering of fields is
117
117
def add(self, tag, value):
118
118
"""Append a name and value to the stanza."""
119
assert valid_tag(tag), \
120
("invalid tag %r" % tag)
119
if not valid_tag(tag):
120
raise ValueError("invalid tag %r" % (tag,))
121
121
if isinstance(value, str):
122
122
value = unicode(value)
123
123
elif isinstance(value, unicode):
128
128
raise TypeError("invalid type for rio value: %r of type %s"
129
129
% (value, type(value)))
130
130
self.items.append((tag, value))
133
def from_pairs(cls, pairs):
132
138
def __contains__(self, find_tag):
133
139
"""True if there is any field in this stanza with the given tag."""
134
140
for tag, value in self.items:
167
173
for tag, value in self.items:
168
assert isinstance(tag, str), type(tag)
169
assert isinstance(value, unicode)
171
175
result.append(tag + ': \n')
172
176
elif '\n' in value:
195
199
for tag, value in self.items:
197
result.append(tag + ': \n')
201
result.append(tag + u': \n')
199
203
# don't want splitlines behaviour on empty lines
200
val_lines = value.split('\n')
201
result.append(tag + ': ' + val_lines[0] + '\n')
204
val_lines = value.split(u'\n')
205
result.append(tag + u': ' + val_lines[0] + u'\n')
202
206
for line in val_lines[1:]:
203
result.append('\t' + line + '\n')
207
result.append(u'\t' + line + u'\n')
205
result.append(tag + ': ' + value + '\n')
209
result.append(tag + u': ' + value + u'\n')
206
210
return u''.join(result)
208
212
def write(self, to_file):
237
241
for tag, value in self.items:
242
_tag_re = re.compile(r'^[-a-zA-Z0-9_]+$')
243
246
def valid_tag(tag):
244
return bool(_tag_re.match(tag))
247
return _valid_tag(tag)
247
250
def read_stanza(line_iter):
248
251
"""Return new Stanza read from list of lines or a file
250
253
Returns one Stanza that was read, or returns None at end of file. If a
251
254
blank line follows the stanza, it is consumed. It's not an error for
252
255
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.
256
start of the input this is really an empty stanza and that is returned.
255
258
Only the stanza lines and the trailing blank (if any) are consumed
256
259
from the line_iter.
258
261
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)
263
return _read_stanza_utf8(line_iter)
264
266
def read_stanza_unicode(unicode_iter):
278
280
: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
283
return _read_stanza_unicode(unicode_iter)
320
286
def to_patch_lines(stanza, max_width=72):
373
340
for line in line_iter:
374
341
if line.startswith('# '):
377
assert line.startswith('#')
343
elif line.startswith('#'):
346
raise ValueError("bad line %r" % (line,))
379
347
if last_line is not None and len(line) > 2:
381
349
line = re.sub('\r', '', line)