1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
|
#! /usr/bin/env python
# (C) 2005 Canonical Ltd
# based on an idea by Matt Mackall
# modified to squish into bzr by Martin Pool
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
"""Packed file revision storage.
A Revfile holds the text history of a particular source file, such
as Makefile. It can represent a tree of text versions for that
file, allowing for microbranches within a single repository.
This is stored on disk as two files: an index file, and a data file.
The index file is short and always read completely into memory; the
data file is much longer and only the relevant bits of it,
identified by the index file, need to be read.
Each text version is identified by the SHA-1 of the full text of
that version. It also has a sequence number within the file.
The index file has a short header and then a sequence of fixed-length
records:
* byte[20] SHA-1 of text (as binary, not hex)
* uint32 sequence number this is based on, or -1 for full text
* uint32 flags: 1=zlib compressed
* uint32 offset in text file of start
* uint32 length of compressed delta in text file
* uint32[3] reserved
total 48 bytes.
The header is also 48 bytes for tidyness and easy calculation.
Both the index and the text are only ever appended to; a consequence
is that sequence numbers are stable references. But not every
repository in the world will assign the same sequence numbers,
therefore the SHA-1 is the only universally unique reference.
This is meant to scale to hold 100,000 revisions of a single file, by
which time the index file will be ~4.8MB and a bit big to read
sequentially.
Some of the reserved fields could be used to implement a (semi?)
balanced tree indexed by SHA1 so we can much more efficiently find the
index associated with a particular hash. For 100,000 revs we would be
able to find it in about 17 random reads, which is not too bad.
"""
# TODO: Something like pread() would make this slightly simpler and
# perhaps more efficient.
# TODO: Could also try to mmap things...
import sys, zlib, struct, mdiff, stat, os, sha
from binascii import hexlify
factor = 10
_RECORDSIZE = 48
_HEADER = "bzr revfile v1\n"
_HEADER = _HEADER + ('\xff' * (_RECORDSIZE - len(_HEADER)))
_NO_BASE = 0xFFFFFFFFL
# fields in the index record
I_SHA = 0
I_BASE = 1
I_FLAGS = 2
I_OFFSET = 3
I_LEN = 4
class RevfileError(Exception):
pass
class Revfile:
def __init__(self, basename):
# TODO: Option to open readonly
# TODO: Lock file while open
# TODO: advise of random access
self.basename = basename
idxname = basename + '.irev'
dataname = basename + '.drev'
idx_exists = os.path.exists(idxname)
data_exists = os.path.exists(dataname)
if idx_exists != data_exists:
raise RevfileError("half-assed revfile")
if not idx_exists:
self.idxfile = open(idxname, 'w+b')
self.datafile = open(dataname, 'w+b')
print 'init empty file'
self.idxfile.write(_HEADER)
self.idxfile.flush()
else:
self.idxfile = open(idxname, 'r+b')
self.datafile = open(dataname, 'r+b')
h = self.idxfile.read(_RECORDSIZE)
if h != _HEADER:
raise RevfileError("bad header %r in index of %r"
% (h, self.basename))
def revision(self, rev):
base = self.index[rev][0]
start = self.index[base][1]
end = self.index[rev][1] + self.index[rev][2]
f = open(self.datafile())
f.seek(start)
data = f.read(end - start)
last = self.index[base][2]
text = zlib.decompress(data[:last])
for r in range(base + 1, rev + 1):
s = self.index[r][2]
b = zlib.decompress(data[last:last + s])
text = mdiff.bpatch(text, b)
last = last + s
return text
def add_full_text(self, t):
"""Add a full text to the file.
This is not compressed against any reference version.
Returns the index for that text."""
idx = len(self)
self.datafile.seek(0, 2) # to end
self.idxfile.seek(0, 2)
assert self.idxfile.tell() == _RECORDSIZE * (idx + 1)
data_offset = self.datafile.tell()
assert isinstance(t, str) # not unicode or anything wierd
self.datafile.write(t)
self.datafile.flush()
entry = sha.new(t).digest()
entry += struct.pack(">IIII12x", 0xFFFFFFFFL, 0, data_offset, len(t))
assert len(entry) == _RECORDSIZE
self.idxfile.write(entry)
self.idxfile.flush()
return idx
def _get_full_text(self, idx):
idxrec = self[idx]
assert idxrec[I_FLAGS] == 0
assert idxrec[I_BASE] == _NO_BASE
l = idxrec[I_LEN]
if l == 0:
return ''
self.datafile.seek(idxrec[I_OFFSET])
text = self.datafile.read(l)
if len(text) != l:
raise RevfileError("short read %d of %d "
"getting text for record %d in %r"
% (len(text), l, idx, self.basename))
return text
def __len__(self):
"""Return number of revisions."""
l = os.fstat(self.idxfile.fileno())[stat.ST_SIZE]
if l % _RECORDSIZE:
raise RevfileError("bad length %d on index of %r" % (l, self.basename))
if l < _RECORDSIZE:
raise RevfileError("no header present in index of %r" % (self.basename))
return int(l / _RECORDSIZE) - 1
def __getitem__(self, idx):
"""Index by sequence id returns the index field"""
self._seek_index(idx)
return self._read_next_index()
def _seek_index(self, idx):
self.idxfile.seek((idx + 1) * _RECORDSIZE)
def _read_next_index(self):
rec = self.idxfile.read(_RECORDSIZE)
if not rec:
raise IndexError("end of index file")
elif len(rec) != _RECORDSIZE:
raise RevfileError("short read of %d bytes getting index %d from %r"
% (len(rec), idx, self.basename))
return struct.unpack(">20sIIII12x", rec)
def addrevision(self, text, changeset):
t = self.tip()
n = t + 1
if not n % factor:
data = zlib.compress(text)
base = n
else:
prev = self.revision(t)
data = zlib.compress(mdiff.bdiff(prev, text))
base = self.index[t][0]
offset = 0
if t >= 0:
offset = self.index[t][1] + self.index[t][2]
self.index.append((base, offset, len(data), changeset))
entry = struct.pack(">llll", base, offset, len(data), changeset)
open(self.indexfile(), "a").write(entry)
open(self.datafile(), "a").write(data)
def dump(self, f=sys.stdout):
f.write('%-8s %-40s %-8s %-8s %-8s %-8s\n'
% tuple('idx sha1 base flags offset len'.split()))
f.write('-------- ---------------------------------------- ')
f.write('-------- -------- -------- --------\n')
for i, rec in enumerate(self):
f.write("#%-7d %40s " % (i, hexlify(rec[0])))
if rec[1] == _NO_BASE:
f.write("(none) ")
else:
f.write("#%-7d " % rec[1])
f.write("%8x %8d %8d\n" % (rec[2], rec[3], rec[4]))
def main(argv):
r = Revfile("testrev")
try:
cmd = argv[1]
except IndexError:
sys.stderr.write("usage: revfile dump\n"
" revfile add\n"
" revfile get IDX\n")
return 1
if cmd == 'add':
new_idx = r.add_full_text(sys.stdin.read())
print 'added idx %d' % new_idx
elif cmd == 'dump':
r.dump()
elif cmd == 'get':
try:
idx = int(argv[2])
except IndexError:
sys.stderr.write("usage: revfile get IDX\n")
return 1
if idx < 0 or idx >= len(r):
sys.stderr.write("invalid index %r\n" % idx)
return 1
sys.stdout.write(r._get_full_text(idx))
else:
sys.stderr.write("unknown command %r\n" % cmd)
return 1
if __name__ == '__main__':
import sys
sys.exit(main(sys.argv) or 0)
|