1
# Copyright (C) 2005 by Canonical Ltd
3
# Distributed under the GNU General Public Licence v2
5
# \subsection{\emph{rio} - simple text metaformat}
7
# \emph{r} stands for `restricted', `reproducible', or `rfc822-like'.
9
# The stored data consists of a series of \emph{stanzas}, each of which contains
10
# \emph{fields} identified by an ascii name, with Unicode or string contents.
11
# The field tag is constrained to alphanumeric characters.
12
# There may be more than one field in a stanza with the same name.
14
# The format itself does not deal with character encoding issues, though
15
# the result will normally be written in Unicode.
17
# The format is intended to be simple enough that there is exactly one character
18
# stream representation of an object and vice versa, and that this relation
19
# will continue to hold for future versions of bzr.
21
# In comments, $\min(1,10)$
27
# XXX: some redundancy is allowing to write stanzas in isolation as well as
28
# through a writer object.
30
class RioWriter(object):
31
def __init__(self, to_file):
33
self._to_file = to_file
35
def write_stanza(self, stanza):
38
stanza.write(self._to_file)
42
class RioReader(object):
43
"""Read stanzas from a file as a sequence
45
to_file can be anything that can be enumerated as a sequence of
46
lines (with newlines.)
48
def __init__(self, from_file):
49
self._from_file = from_file
53
s = read_stanza(self._from_file)
59
def read_stanzas(from_file):
61
s = read_stanza(from_file)
68
"""One stanza for rio.
70
Each stanza contains a set of named fields.
72
Names must be non-empty ascii alphanumeric plus _. Names can be repeated
73
within a stanza. Names are case-sensitive. The ordering of fields is
76
Each field value must be either an int or a string.
81
def __init__(self, **kwargs):
82
"""Construct a new Stanza.
84
The keyword arguments, if any, are added in sorted order to the stanza.
88
for tag, value in sorted(kwargs.items()):
91
def add(self, tag, value):
92
"""Append a name and value to the stanza."""
93
assert valid_tag(tag), \
94
("invalid tag %r" % tag)
95
if isinstance(value, (str, unicode)):
97
## elif isinstance(value, (int, long)):
98
## value = str(value) # XXX: python2.4 without L-suffix
100
raise ValueError("invalid value %r" % value)
101
self.items.append((tag, value))
103
def __contains__(self, find_tag):
104
"""True if there is any field in this stanza with the given tag."""
105
for tag, value in self.items:
111
"""Return number of pairs in the stanza."""
112
return len(self.items)
114
def __eq__(self, other):
115
if not isinstance(other, Stanza):
117
return self.items == other.items
119
def __ne__(self, other):
120
return not self.__eq__(other)
123
return "Stanza(%r)" % self.items
125
def iter_pairs(self):
126
"""Return iterator of tag, value pairs."""
127
return iter(self.items)
130
"""Generate sequence of lines for external version of this file."""
132
# max() complains if sequence is empty
135
for tag, value in self.items:
136
assert isinstance(value, (str, unicode))
138
result.append(tag + ': \n')
140
# don't want splitlines behaviour on empty lines
141
val_lines = value.split('\n')
142
result.append(tag + ': ' + val_lines[0] + '\n')
143
for line in val_lines[1:]:
144
result.append('\t' + line + '\n')
146
result.append(tag + ': ' + value + '\n')
150
"""Return stanza as a single string"""
151
return ''.join(self.to_lines())
153
def write(self, to_file):
154
"""Write stanza to a file"""
155
to_file.writelines(self.to_lines())
158
"""Return the value for a field wih given tag.
160
If there is more than one value, only the first is returned. If the
161
tag is not present, KeyError is raised.
163
for t, v in self.items:
171
def get_all(self, tag):
173
for t, v in self.items:
178
_tag_re = re.compile(r'^[-a-zA-Z0-9_]+$')
180
return bool(_tag_re.match(tag))
183
def read_stanza(line_iter):
184
"""Return new Stanza read from list of lines or a file
186
Returns one Stanza that was read, or returns None at end of file. If a
187
blank line follows the stanza, it is consumed. It's not an error for
188
there to be no blank at end of file. If there is a blank file at the
189
start of the input this is really an empty stanza and that is returned.
191
Only the stanza lines and the trailing blank (if any) are consumed
198
for line in line_iter:
199
if line == None or line == '':
202
break # end of stanza
203
assert line[-1] == '\n'
205
if line[0] == '\t': # continues previous value
207
raise ValueError('invalid continuation line %r' % real_l)
208
accum_value += '\n' + line[1:-1]
209
else: # new tag:value line
211
stanza.add(tag, accum_value)
213
colon_index = line.index(': ')
215
raise ValueError('tag/value separator not found in line %r' % real_l)
216
tag = line[:colon_index]
217
assert valid_tag(tag), \
218
"invalid rio tag %r" % tag
219
accum_value = line[colon_index+2:-1]
220
if tag is not None: # add last tag-value
221
stanza.add(tag, accum_value)
223
else: # didn't see any content
added
removed
bzrlib/rio.py
