def tree_entries_from_data(data):
"""Reads the binary representation of a tree and returns tuples of Tree items
:param data: data block with tree data (as bytes)
:return: list(tuple(binsha, mode, tree_relative_path), ...)"""
ord_zero = ord('0')
space_ord = ord(' ')
len_data = len(data)
i = 0
out = list()
while i < len_data:
mode = 0
# read mode
# Some git versions truncate the leading 0, some don't
# The type will be extracted from the mode later
while byte_ord(data[i]) != space_ord:
# move existing mode integer up one level being 3 bits
# and add the actual ordinal value of the character
mode = (mode << 3) + (byte_ord(data[i]) - ord_zero)
i += 1
# END while reading mode
# byte is space now, skip it
i += 1
# parse name, it is NULL separated
ns = i
while byte_ord(data[i]) != 0:
i += 1
# END while not reached NULL
# default encoding for strings in git is utf8
# Only use the respective unicode object if the byte stream was encoded
name = data[ns:i]
try:
name = name.decode(defenc)
except UnicodeDecodeError:
pass
# END handle encoding
# byte is NULL, get next 20
i += 1
sha = data[i:i + 20]
i = i + 20
out.append((sha, mode, name))
# END for each byte in data stream
return out
def tree_to_stream(entries, write):
"""Write the give list of entries into a stream using its write method
:param entries: **sorted** list of tuples with (binsha, mode, name)
:param write: write method which takes a data string"""
ord_zero = ord('0')
bit_mask = 7 # 3 bits set
for binsha, mode, name in entries:
mode_str = b''
for i in xrange(6):
mode_str = bchr(((mode >> (i * 3)) & bit_mask) + ord_zero) + mode_str
# END for each 8 octal value
# git slices away the first octal if its zero
if byte_ord(mode_str[0]) == ord_zero:
mode_str = mode_str[1:]
# END save a byte
# here it comes: if the name is actually unicode, the replacement below
# will not work as the binsha is not part of the ascii unicode encoding -
# hence we must convert to an utf8 string for it to work properly.
# According to my tests, this is exactly what git does, that is it just
# takes the input literally, which appears to be utf8 on linux.
if isinstance(name, text_type):
name = name.encode(defenc)
write(b''.join((mode_str, b' ', name, b'\0', binsha)))
def tree_to_stream(entries, write):
"""Write the give list of entries into a stream using its write method
:param entries: **sorted** list of tuples with (binsha, mode, name)
:param write: write method which takes a data string"""
ord_zero = ord('0')
bit_mask = 7 # 3 bits set
for binsha, mode, name in entries:
mode_str = b''
for i in xrange(6):
mode_str = bchr((
(mode >> (i * 3)) & bit_mask) + ord_zero) + mode_str
# END for each 8 octal value
# git slices away the first octal if its zero
if byte_ord(mode_str[0]) == ord_zero:
mode_str = mode_str[1:]
# END save a byte
# here it comes: if the name is actually unicode, the replacement below
# will not work as the binsha is not part of the ascii unicode encoding -
# hence we must convert to an utf8 string for it to work properly.
# According to my tests, this is exactly what git does, that is it just
# takes the input literally, which appears to be utf8 on linux.
if isinstance(name, text_type):
name = name.encode(defenc)
write(b''.join((mode_str, b' ', name, b'\0', binsha)))