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.
37
from bzrlib import osutils
38
35
from bzrlib.iterablefile import IterableFile
40
37
# XXX: some redundancy is allowing to write stanzas in isolation as well as
41
# through a writer object.
38
# through a writer object.
43
40
class RioWriter(object):
44
41
def __init__(self, to_file):
55
52
class RioReader(object):
56
53
"""Read stanzas from a file as a sequence
58
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
59
56
lines (with newlines.)
61
58
def __init__(self, from_file):
96
93
class Stanza(object):
97
94
"""One stanza for rio.
99
Each stanza contains a set of named fields.
96
Each stanza contains a set of named fields.
101
98
Names must be non-empty ascii alphanumeric plus _. Names can be repeated
102
99
within a stanza. Names are case-sensitive. The ordering of fields is
131
128
raise TypeError("invalid type for rio value: %r of type %s"
132
129
% (value, type(value)))
133
130
self.items.append((tag, value))
136
def from_pairs(cls, pairs):
141
132
def __contains__(self, find_tag):
142
133
"""True if there is any field in this stanza with the given tag."""
143
134
for tag, value in self.items:
202
193
for tag, value in self.items:
204
result.append(tag + u': \n')
195
result.append(tag + ': \n')
206
197
# don't want splitlines behaviour on empty lines
207
val_lines = value.split(u'\n')
208
result.append(tag + u': ' + val_lines[0] + u'\n')
198
val_lines = value.split('\n')
199
result.append(tag + ': ' + val_lines[0] + '\n')
209
200
for line in val_lines[1:]:
210
result.append(u'\t' + line + u'\n')
201
result.append('\t' + line + '\n')
212
result.append(tag + u': ' + value + u'\n')
203
result.append(tag + ': ' + value + '\n')
213
204
return u''.join(result)
215
206
def write(self, to_file):
244
235
for tag, value in self.items:
239
_tag_re = re.compile(r'^[-a-zA-Z0-9_]+$')
249
240
def valid_tag(tag):
250
return _valid_tag(tag)
241
return bool(_tag_re.match(tag))
253
244
def read_stanza(line_iter):
254
245
"""Return new Stanza read from list of lines or a file
256
247
Returns one Stanza that was read, or returns None at end of file. If a
257
248
blank line follows the stanza, it is consumed. It's not an error for
258
249
there to be no blank at end of file. If there is a blank file at the
259
start of the input this is really an empty stanza and that is returned.
250
start of the input this is really an empty stanza and that is returned.
261
252
Only the stanza lines and the trailing blank (if any) are consumed
262
253
from the line_iter.
264
255
The raw lines must be in utf-8 encoding.
266
return _read_stanza_utf8(line_iter)
257
unicode_iter = (line.decode('utf-8') for line in line_iter)
258
return read_stanza_unicode(unicode_iter)
269
261
def read_stanza_unicode(unicode_iter):
283
275
:return: A Stanza object if there are any lines in the file.
286
return _read_stanza_unicode(unicode_iter)
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
289
316
def to_patch_lines(stanza, max_width=72):