« back to all changes in this revision
Viewing changes to bzrlib/knit.py
- browse files at revision 1926
- compare with another revision
- compare with revision 5147.5.4
- download diff
- download tarball
- view history from revision 1926
- Committer: Canonical.com Patch Queue Manager
- Date: 2006-08-15 14:16:58 UTC
- mfrom: (1911.2.9 cache-revision-encoding)
- Revision ID: pqm@pqm.ubuntu.com-20060815141658-4a1356532d7aa155
(jam) cache utf-8 encoding/decoding
added
removed
bzrlib/knit.py
74
74
import warnings
75
75
76
76
import bzrlib
77
import bzrlib.errors as errors
77
from bzrlib import (
78
cache_utf8,
79
errors,
80
)
78
81
from bzrlib.errors import FileExists, NoSuchFile, KnitError, \
79
82
InvalidRevisionId, KnitCorrupt, KnitHeaderError, \
80
83
RevisionNotPresent, RevisionAlreadyPresent
82
85
from bzrlib.trace import mutter
83
86
from bzrlib.osutils import contains_whitespace, contains_linebreaks, \
84
87
sha_strings
85
from bzrlib.versionedfile import VersionedFile, InterVersionedFile
86
88
from bzrlib.symbol_versioning import DEPRECATED_PARAMETER, deprecated_passed
87
89
from bzrlib.tsort import topo_sort
88
90
import bzrlib.weave
91
from bzrlib.versionedfile import VersionedFile, InterVersionedFile
89
92
90
93
91
94
# TODO: Split out code specific to this format into an associated object.
162
165
internal representation is of the format:
163
166
(revid, plaintext)
164
167
"""
168
decode_utf8 = cache_utf8.decode
165
169
lines = []
166
170
for line in content:
167
171
origin, text = line.split(' ', 1)
168
lines.append((origin.decode('utf-8'), text))
172
lines.append((decode_utf8(origin), text))
169
173
return KnitContent(lines)
170
174
171
175
def parse_line_delta_iter(self, lines):
182
186
internal representation is
183
187
(start, end, count, [1..count tuples (revid, newline)])
184
188
"""
189
decode_utf8 = cache_utf8.decode
185
190
result = []
186
191
lines = iter(lines)
187
192
next = lines.next
193
198
while remaining:
194
199
origin, text = next().split(' ', 1)
195
200
remaining -= 1
196
contents.append((origin.decode('utf-8'), text))
201
contents.append((decode_utf8(origin), text))
197
202
result.append((start, end, count, contents))
198
203
return result
199
204
202
207
203
208
see parse_fulltext which this inverts.
204
209
"""
205
return ['%s %s' % (o.encode('utf-8'), t) for o, t in content._lines]
210
encode_utf8 = cache_utf8.encode
211
return ['%s %s' % (encode_utf8(o), t) for o, t in content._lines]
206
212
207
213
def lower_line_delta(self, delta):
208
214
"""convert a delta into a serializable form.
209
215
210
216
See parse_line_delta which this inverts.
211
217
"""
218
encode_utf8 = cache_utf8.encode
212
219
out = []
213
220
for start, end, c, lines in delta:
214
221
out.append('%d,%d,%d\n' % (start, end, c))
215
for origin, text in lines:
216
out.append('%s %s' % (origin.encode('utf-8'), text))
222
out.extend(encode_utf8(origin) + ' ' + text
223
for origin, text in lines)
217
224
return out
218
225
219
226
1206
1213
return self._cache[version_id][5]
1207
1214
1208
1215
def _version_list_to_index(self, versions):
1216
encode_utf8 = cache_utf8.encode
1209
1217
result_list = []
1210
1218
for version in versions:
1211
1219
if version in self._cache:
1213
1221
result_list.append(str(self._cache[version][5]))
1214
1222
# -- end lookup () --
1215
1223
else:
1216
result_list.append('.' + version.encode('utf-8'))
1224
result_list.append('.' + encode_utf8(version))
1217
1225
return ' '.join(result_list)
1218
1226
1219
1227
def add_version(self, version_id, options, pos, size, parents):
1227
1235
(version_id, options, pos, size, parents).
1228
1236
"""
1229
1237
lines = []
1238
encode_utf8 = cache_utf8.encode
1230
1239
for version_id, options, pos, size, parents in versions:
1231
line = "\n%s %s %s %s %s :" % (version_id.encode('utf-8'),
1240
line = "\n%s %s %s %s %s :" % (encode_utf8(version_id),
1232
1241
','.join(options),
1233
1242
pos,
1234
1243
size,
1324
1333
"""
1325
1334
sio = StringIO()
1326
1335
data_file = GzipFile(None, mode='wb', fileobj=sio)
1336
1337
version_id_utf8 = cache_utf8.encode(version_id)
1327
1338
data_file.writelines(chain(
1328
["version %s %d %s\n" % (version_id.encode('utf-8'),
1339
["version %s %d %s\n" % (version_id_utf8,
1329
1340
len(lines),
1330
1341
digest)],
1331
1342
lines,
1332
["end %s\n" % version_id.encode('utf-8')]))
1343
["end %s\n" % version_id_utf8]))
1333
1344
data_file.close()
1334
1345
length= sio.tell()
1335
1346
1364
1375
rec = df.readline().split()
1365
1376
if len(rec) != 4:
1366
1377
raise KnitCorrupt(self._filename, 'unexpected number of elements in record header')
1367
if rec[1].decode('utf-8')!= version_id:
1378
if cache_utf8.decode(rec[1]) != version_id:
1368
1379
raise KnitCorrupt(self._filename,
1369
1380
'unexpected version, wanted %r, got %r' % (
1370
1381
version_id, rec[1]))
1379
1390
record_contents = df.readlines()
1380
1391
l = record_contents.pop()
1381
1392
assert len(record_contents) == int(rec[2])
1382
if l.decode('utf-8') != 'end %s\n' % version_id:
1393
if l != 'end %s\n' % cache_utf8.encode(version_id):
1383
1394
raise KnitCorrupt(self._filename, 'unexpected version end line %r, wanted %r'
1384
1395
% (l, version_id))
1385
1396
df.close()
Loggerhead is a web-based interface for Breezy
Version: 2.0.1