def _EndRecData(fpin):
"""Return data from the "End of Central Directory" record, or None.
The data is a list of the nine items in the ZIP "End of central dir"
record followed by a tenth item, the file seek offset of this record."""
fpin.seek(-22, 2) # Assume no archive comment.
filesize = fpin.tell() + 22 # Get file size
data = fpin.readall()
start = len(data)-2
assert start > 0
if data[0:4] == stringEndArchive and data[start:] == "\000\000":
endrec = runpack(structEndArchive, data)
return EndRecStruct(endrec, "", filesize - 22)
# Search the last END_BLOCK bytes of the file for the record signature.
# The comment is appended to the ZIP file and has a 16 bit length.
# So the comment may be up to 64K long. We limit the search for the
# signature to a few Kbytes at the end of the file for efficiency.
# also, the signature must not appear in the comment.
END_BLOCK = min(filesize, 1024 * 4)
fpin.seek(filesize - END_BLOCK, 0)
data = fpin.readall()
start = data.rfind(stringEndArchive)
if start >= 0: # Correct signature string was found
endrec = runpack(structEndArchive, data[start:start+22])
comment = data[start+22:]
if endrec[7] == len(comment): # Comment length checks out
# Append the archive comment and start offset
return EndRecStruct(endrec, comment, filesize - END_BLOCK + start)
return # Error, return None
def _GetContents(self, fp):
endrec = _EndRecData(fp)
if not endrec:
raise BadZipfile, "File is not a zip file"
size_cd = endrec.stuff[5] # bytes in central directory
offset_cd = endrec.stuff[6] # offset of central directory
self.comment = endrec.comment
x = endrec.filesize - size_cd
concat = x - offset_cd
self.start_dir = offset_cd + concat
fp.seek(self.start_dir, 0)
total = 0
while total < size_cd:
centdir = fp.read(46)
total = total + 46
if centdir[0:4] != stringCentralDir:
raise BadZipfile, "Bad magic number for central directory"
centdir = runpack(structCentralDir, centdir)
filename = fp.read(centdir[_CD_FILENAME_LENGTH])
# Create ZipInfo instance to store file information
x = RZipInfo(filename)
x.extra = fp.read(centdir[_CD_EXTRA_FIELD_LENGTH])
x.comment = fp.read(centdir[_CD_COMMENT_LENGTH])
total = (total + centdir[_CD_FILENAME_LENGTH]
+ centdir[_CD_EXTRA_FIELD_LENGTH]
+ centdir[_CD_COMMENT_LENGTH])
x.header_offset = centdir[_CD_LOCAL_HEADER_OFFSET] + concat
# file_offset must be computed below...
(x.create_version, x.create_system, x.extract_version, x.reserved,
x.flag_bits, x.compress_type, t, d,
crc, x.compress_size, x.file_size) = centdir[1:12]
x.CRC = r_uint(crc) & 0xffffffff
x.dostime = t
x.dosdate = d
x.volume, x.internal_attr, x.external_attr = centdir[15:18]
# Convert date/time code to (year, month, day, hour, min, sec)
x.date_time = ( (d>>9)+1980, (d>>5)&0xF, d&0x1F,
t>>11, (t>>5)&0x3F, (t&0x1F) * 2 )
self.filelist.append(x)
self.NameToInfo[x.filename] = x
for data in self.filelist:
fp.seek(data.header_offset, 0)
fheader = fp.read(30)
if fheader[0:4] != stringFileHeader:
raise BadZipfile, "Bad magic number for file header"
fheader = runpack(structFileHeader, fheader)
# file_offset is computed here, since the extra field for
# the central directory and for the local file header
# refer to different fields, and they can have different
# lengths
data.file_offset = (data.header_offset + 30
+ fheader[_FH_FILENAME_LENGTH]
+ fheader[_FH_EXTRA_FIELD_LENGTH])
fname = fp.read(fheader[_FH_FILENAME_LENGTH])
if fname != data.orig_filename:
raise RuntimeError, \
'File name in directory "%s" and header "%s" differ.' % (
data.orig_filename, fname)
fp.seek(self.start_dir, 0)
def unserialize_code(coderepr, start=0):
coderepr = coderepr[start:]
header = coderepr[:header_lgt]
argcount, nlocals, stacksize, code_len = runpack(FMT, header)
assert code_len >= 0
names_pos = code_len + header_lgt
code = coderepr[header_lgt:names_pos]
num = runpack('i', coderepr[names_pos:names_pos + int_lgt])
names, end_names = unserialize_names(coderepr[names_pos + int_lgt:], num)
const_start = names_pos + int_lgt + end_names
consts, pos = unserialize_consts(coderepr[const_start:])
pos = start + const_start + pos
return Code(argcount, nlocals, stacksize, code, consts, names), pos
def unserialize_consts(constrepr):
pos = int_lgt
consts_w = []
num = runpack('i', constrepr[:int_lgt])
for i in range(num):
next_const, pos = unserialize_const(constrepr, pos)
consts_w.append(next_const)
return consts_w, pos
def unserialize_const(c, start):
assert start >= 0
if c[start] == 'd':
end = start + int_lgt + 1
intval = runpack('i', c[start + 1:end])
return Int(intval), end
elif c[start] == 's':
value, end = unserialize_str(c, start + 1)
return Str(value), end
elif c[start] == 'n':
return spli_None, start + 1
elif c[start] == 'c':
return unserialize_code(c, start + 1)
else:
raise NotSupportedFormat(c[start])
def fromstring(space, s):
from pypy.module.micronumpy.interp_numarray import SingleDimArray
length = len(s)
if length % FLOAT_SIZE == 0:
number = length/FLOAT_SIZE
else:
raise OperationError(space.w_ValueError, space.wrap(
"string length %d not divisable by %d" % (length, FLOAT_SIZE)))
a = SingleDimArray(number)
start = 0
end = FLOAT_SIZE
i = 0
while i < number:
part = s[start:end]
a.storage[i] = runpack('d', part)
i += 1
start += FLOAT_SIZE
end += FLOAT_SIZE
return space.wrap(a)
def fromstring(space, s):
from pypy.module.micronumpy.interp_numarray import NDimArray
length = len(s)
if length % FLOAT_SIZE == 0:
number = length/FLOAT_SIZE
else:
raise OperationError(space.w_ValueError, space.wrap(
"string length %d not divisable by %d" % (length, FLOAT_SIZE)))
dtype = space.fromcache(W_Float64Dtype)
a = NDimArray(number, [number], dtype=dtype)
start = 0
end = FLOAT_SIZE
i = 0
while i < number:
part = s[start:end]
a.dtype.setitem(a.storage, i, dtype.box(runpack('d', part)))
i += 1
start += FLOAT_SIZE
end += FLOAT_SIZE
return space.wrap(a)
def fn():
return runpack('i', data)
def fn():
a, b, c, d = runpack('iiii', data)
return a * 1000 + b * 100 + c * 10 + d
def fn():
return runpack('sll',
'a\x00\x00\x00\x03\x00\x00\x00\x04\x00\x00\x00')[1]
def fn():
return runpack('i', data)
def fn():
a, b, c, d = runpack('iiii', data)
return a * 1000 + b * 100 + c * 10 + d
def runpack_str(self, s):
return self.box(runpack(self.format_code, s))
def fn():
return runpack('sll', 'a'+pad+'\x03'+pad+'\x04'+pad)[1]
def fn():
return runpack('sll', 'a\x00\x00\x00\x03\x00\x00\x00\x04\x00\x00\x00')[1]
def _PyFloat_Unpack8(space, ptr, le):
input = rffi.charpsize2str(ptr, 8)
if rffi.cast(lltype.Signed, le):
return runpack.runpack("<d", input)
else:
return runpack.runpack(">d", input)
def unserialize_str(data, start):
end_lgt = start + int_lgt
lgt = runpack('i', data[start:end_lgt])
assert lgt >= 0
end_str = end_lgt + lgt
return data[end_lgt:end_str], end_str
def runpack_str(self, s):
return self.box(runpack(self.format_code, s))
def unserialize_str(data, start):
end_lgt = start + int_lgt
lgt = runpack('i', data[start:end_lgt])
assert lgt >= 0
end_str = end_lgt + lgt
return data[end_lgt:end_str], end_str
