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
|
##############################################################################
#
# Copyright (c) 2001-2004 Zope Foundation and Contributors.
# All Rights Reserved.
#
# This software is subject to the provisions of the Zope Public License,
# Version 2.1 (ZPL). A copy of the ZPL should accompany this distribution.
# THIS SOFTWARE IS PROVIDED "AS IS" AND ANY AND ALL EXPRESS OR IMPLIED
# WARRANTIES ARE DISCLAIMED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF TITLE, MERCHANTABILITY, AGAINST INFRINGEMENT, AND FITNESS
# FOR A PARTICULAR PURPOSE.
#
##############################################################################
"""Buffers
"""
from io import BytesIO
# copy_bytes controls the size of temp. strings for shuffling data around.
COPY_BYTES = 1 << 18 # 256K
# The maximum number of bytes to buffer in a simple string.
STRBUF_LIMIT = 8192
class FileBasedBuffer(object):
remain = 0
def __init__(self, file, from_buffer=None):
self.file = file
if from_buffer is not None:
from_file = from_buffer.getfile()
read_pos = from_file.tell()
from_file.seek(0)
while True:
data = from_file.read(COPY_BYTES)
if not data:
break
file.write(data)
self.remain = int(file.tell() - read_pos)
from_file.seek(read_pos)
file.seek(read_pos)
def __len__(self):
return self.remain
def __nonzero__(self):
return self.remain > 0
__bool__ = __nonzero__ # py3
def append(self, s):
file = self.file
read_pos = file.tell()
file.seek(0, 2)
file.write(s)
file.seek(read_pos)
self.remain = self.remain + len(s)
def get(self, numbytes=-1, skip=False):
file = self.file
if not skip:
read_pos = file.tell()
if numbytes < 0:
# Read all
res = file.read()
else:
res = file.read(numbytes)
if skip:
self.remain -= len(res)
else:
file.seek(read_pos)
return res
def skip(self, numbytes, allow_prune=0):
if self.remain < numbytes:
raise ValueError("Can't skip %d bytes in buffer of %d bytes" % (
numbytes, self.remain)
)
self.file.seek(numbytes, 1)
self.remain = self.remain - numbytes
def newfile(self):
raise NotImplementedError()
def prune(self):
file = self.file
if self.remain == 0:
read_pos = file.tell()
file.seek(0, 2)
sz = file.tell()
file.seek(read_pos)
if sz == 0:
# Nothing to prune.
return
nf = self.newfile()
while True:
data = file.read(COPY_BYTES)
if not data:
break
nf.write(data)
self.file = nf
def getfile(self):
return self.file
def _close(self):
# named _close because ReadOnlyFileBasedBuffer is used as
# wsgi file.wrapper, and its protocol reserves "close"
if hasattr(self.file, 'close'):
self.file.close()
self.remain = 0
class TempfileBasedBuffer(FileBasedBuffer):
def __init__(self, from_buffer=None):
FileBasedBuffer.__init__(self, self.newfile(), from_buffer)
def newfile(self):
from tempfile import TemporaryFile
return TemporaryFile('w+b')
class BytesIOBasedBuffer(FileBasedBuffer):
def __init__(self, from_buffer=None):
if from_buffer is not None:
FileBasedBuffer.__init__(self, BytesIO(), from_buffer)
else:
# Shortcut. :-)
self.file = BytesIO()
def newfile(self):
return BytesIO()
class ReadOnlyFileBasedBuffer(FileBasedBuffer):
# used as wsgi.file_wrapper
def __init__(self, file, block_size=32768):
self.file = file
self.block_size = block_size # for __iter__
def prepare(self, size=None):
if hasattr(self.file, 'seek') and hasattr(self.file, 'tell'):
start_pos = self.file.tell()
self.file.seek(0, 2)
end_pos = self.file.tell()
self.file.seek(start_pos)
fsize = end_pos - start_pos
if size is None:
self.remain = fsize
else:
self.remain = min(fsize, size)
elif hasattr(self.file, 'close'):
# called by task if self.filelike has no seek/tell
self.close = self.file.close
return self.remain
def get(self, numbytes=-1, skip=False):
# never read more than self.remain (it can be user-specified)
if numbytes == -1 or numbytes > self.remain:
numbytes = self.remain
file = self.file
if not skip:
read_pos = file.tell()
res = file.read(numbytes)
if skip:
self.remain -= len(res)
else:
file.seek(read_pos)
return res
def __iter__(self): # called by task if self.filelike has no seek/tell
return self
def next(self):
val = self.file.read(self.block_size)
if not val:
raise StopIteration
return val
__next__ = next # py3
def append(self, s):
raise NotImplementedError
class OverflowableBuffer(object):
"""
This buffer implementation has four stages:
- No data
- String-based buffer
- StringIO-based buffer
- Temporary file storage
The first two stages are fastest for simple transfers.
"""
overflowed = False
buf = None
strbuf = b'' # Bytes-based buffer.
def __init__(self, overflow):
# overflow is the maximum to be stored in a StringIO buffer.
self.overflow = overflow
def __len__(self):
buf = self.buf
if buf is not None:
return buf.__len__()
else:
return self.strbuf.__len__()
def __nonzero__(self):
return self.__len__() > 0
__bool__ = __nonzero__ # py3
def _create_buffer(self):
strbuf = self.strbuf
if len(strbuf) >= self.overflow:
self._set_large_buffer()
else:
self._set_small_buffer()
buf = self.buf
if strbuf:
buf.append(self.strbuf)
self.strbuf = b''
return buf
def _set_small_buffer(self):
self.buf = BytesIOBasedBuffer(self.buf)
self.overflowed = False
def _set_large_buffer(self):
self.buf = TempfileBasedBuffer(self.buf)
self.overflowed = True
def append(self, s):
buf = self.buf
if buf is None:
strbuf = self.strbuf
if len(strbuf) + len(s) < STRBUF_LIMIT:
self.strbuf = strbuf + s
return
buf = self._create_buffer()
buf.append(s)
sz = len(buf)
if not self.overflowed:
if sz >= self.overflow:
self._set_large_buffer()
def get(self, numbytes=-1, skip=False):
buf = self.buf
if buf is None:
strbuf = self.strbuf
if not skip:
return strbuf
buf = self._create_buffer()
return buf.get(numbytes, skip)
def skip(self, numbytes, allow_prune=False):
buf = self.buf
if buf is None:
if allow_prune and numbytes == len(self.strbuf):
# We could slice instead of converting to
# a buffer, but that would eat up memory in
# large transfers.
self.strbuf = b''
return
buf = self._create_buffer()
buf.skip(numbytes, allow_prune)
def prune(self):
"""
A potentially expensive operation that removes all data
already retrieved from the buffer.
"""
buf = self.buf
if buf is None:
self.strbuf = b''
return
buf.prune()
if self.overflowed:
sz = len(buf)
if sz < self.overflow:
# Revert to a faster buffer.
self._set_small_buffer()
def getfile(self):
buf = self.buf
if buf is None:
buf = self._create_buffer()
return buf.getfile()
def _close(self):
buf = self.buf
if buf is not None:
buf._close()
|
