Mercurial > public > mercurial-scm > hg-stable
comparison mercurial/utils/cborutil.py @ 39439:a40d3da89b7d
cborutil: remove readindefinitebytestringtoiter()
This was implemented as part of implementing streaming encoding.
It was never used outside of tests.
Now that we have a full CBOR decoder, it can be used for incremental
decoding of indefinite-length byte strings.
This also removes the last use of the vendored cbor2 package from this
module.
Differential Revision: https://phab.mercurial-scm.org/D4433
| author | Gregory Szorc <gregory.szorc@gmail.com> |
|---|---|
| date | Fri, 31 Aug 2018 15:54:17 -0700 |
| parents | aeb551a3bb8a |
| children | babad5ebaf0a |
comparison
equal
deleted
inserted
replaced
| 39438:aeb551a3bb8a | 39439:a40d3da89b7d |
|---|---|
| 7 | 7 |
| 8 from __future__ import absolute_import | 8 from __future__ import absolute_import |
| 9 | 9 |
| 10 import struct | 10 import struct |
| 11 import sys | 11 import sys |
| 12 | |
| 13 from ..thirdparty.cbor.cbor2 import ( | |
| 14 decoder as decodermod, | |
| 15 ) | |
| 16 | 12 |
| 17 # Very short very of RFC 7049... | 13 # Very short very of RFC 7049... |
| 18 # | 14 # |
| 19 # Each item begins with a byte. The 3 high bits of that byte denote the | 15 # Each item begins with a byte. The 3 high bits of that byte denote the |
| 20 # "major type." The lower 5 bits denote the "subtype." Each major type | 16 # "major type." The lower 5 bits denote the "subtype." Each major type |
| 216 | 212 |
| 217 if not fn: | 213 if not fn: |
| 218 raise ValueError('do not know how to encode %s' % type(v)) | 214 raise ValueError('do not know how to encode %s' % type(v)) |
| 219 | 215 |
| 220 return fn(v) | 216 return fn(v) |
| 221 | |
| 222 def readindefinitebytestringtoiter(fh, expectheader=True): | |
| 223 """Read an indefinite bytestring to a generator. | |
| 224 | |
| 225 Receives an object with a ``read(X)`` method to read N bytes. | |
| 226 | |
| 227 If ``expectheader`` is True, it is expected that the first byte read | |
| 228 will represent an indefinite length bytestring. Otherwise, we | |
| 229 expect the first byte to be part of the first bytestring chunk. | |
| 230 """ | |
| 231 read = fh.read | |
| 232 decodeuint = decodermod.decode_uint | |
| 233 byteasinteger = decodermod.byte_as_integer | |
| 234 | |
| 235 if expectheader: | |
| 236 initial = decodermod.byte_as_integer(read(1)) | |
| 237 | |
| 238 majortype = initial >> 5 | |
| 239 subtype = initial & SUBTYPE_MASK | |
| 240 | |
| 241 if majortype != MAJOR_TYPE_BYTESTRING: | |
| 242 raise decodermod.CBORDecodeError( | |
| 243 'expected major type %d; got %d' % (MAJOR_TYPE_BYTESTRING, | |
| 244 majortype)) | |
| 245 | |
| 246 if subtype != SUBTYPE_INDEFINITE: | |
| 247 raise decodermod.CBORDecodeError( | |
| 248 'expected indefinite subtype; got %d' % subtype) | |
| 249 | |
| 250 # The indefinite bytestring is composed of chunks of normal bytestrings. | |
| 251 # Read chunks until we hit a BREAK byte. | |
| 252 | |
| 253 while True: | |
| 254 # We need to sniff for the BREAK byte. | |
| 255 initial = byteasinteger(read(1)) | |
| 256 | |
| 257 if initial == BREAK_INT: | |
| 258 break | |
| 259 | |
| 260 length = decodeuint(fh, initial & SUBTYPE_MASK) | |
| 261 chunk = read(length) | |
| 262 | |
| 263 if len(chunk) != length: | |
| 264 raise decodermod.CBORDecodeError( | |
| 265 'failed to read bytestring chunk: got %d bytes; expected %d' % ( | |
| 266 len(chunk), length)) | |
| 267 | |
| 268 yield chunk | |
| 269 | 217 |
| 270 class CBORDecodeError(Exception): | 218 class CBORDecodeError(Exception): |
| 271 """Represents an error decoding CBOR.""" | 219 """Represents an error decoding CBOR.""" |
| 272 | 220 |
| 273 if sys.version_info.major >= 3: | 221 if sys.version_info.major >= 3: |