def source_to_unicode(txt, errors="replace", skip_encoding_cookie=True):
"""Converts a bytes string with python source code to unicode.
Unicode strings are passed through unchanged. Byte strings are checked
for the python source file encoding cookie to determine encoding.
txt can be either a bytes buffer or a string containing the source
code.
"""
if isinstance(txt, str):
return txt
if isinstance(txt, bytes):
buf = io.BytesIO(txt)
else:
buf = txt
try:
encoding, _ = detect_encoding(buf.readline)
except SyntaxError:
encoding = "ascii"
buf.seek(0)
text = io.TextIOWrapper(buf, encoding, errors=errors, line_buffering=True)
text.mode = "r"
if skip_encoding_cookie:
return u"".join(strip_encoding_cookie(text))
else:
return text.read()
def source_to_unicode(txt, errors="replace", skip_encoding_cookie=True):
"""Converts a bytes string with python source code to unicode.
Unicode strings are passed through unchanged. Byte strings are checked
for the python source file encoding cookie to determine encoding.
txt can be either a bytes buffer or a string containing the source
code.
"""
if isinstance(txt, str):
return txt
if isinstance(txt, bytes):
buf = io.BytesIO(txt)
else:
buf = txt
try:
encoding, _ = detect_encoding(buf.readline)
except SyntaxError:
encoding = "ascii"
buf.seek(0)
text = io.TextIOWrapper(buf, encoding, errors=errors, line_buffering=True)
text.mode = "r"
if skip_encoding_cookie:
return u"".join(strip_encoding_cookie(text))
else:
return text.read()
def get_encoding(obj):
"""Get encoding for python source file defining obj
Returns None if obj is not defined in a sourcefile.
"""
ofile = find_file(obj)
# run contents of file through pager starting at line where the object
# is defined, as long as the file isn't binary and is actually on the
# filesystem.
if ofile is None:
return None
elif ofile.endswith((".so", ".dll", ".pyd")):
return None
elif not os.path.isfile(ofile):
return None
else:
# Print only text files, not extension binaries. Note that
# getsourcelines returns lineno with 1-offset and page() uses
# 0-offset, so we must adjust.
with io.open(ofile, "rb") as buf: # Tweaked to use io.open for Python 2
encoding, _ = detect_encoding(buf.readline)
return encoding
def get_encoding(obj):
"""Get encoding for python source file defining obj
Returns None if obj is not defined in a sourcefile.
"""
ofile = find_file(obj)
# run contents of file through pager starting at line where the object
# is defined, as long as the file isn't binary and is actually on the
# filesystem.
if ofile is None:
return None
elif ofile.endswith(('.so', '.dll', '.pyd')):
return None
elif not os.path.isfile(ofile):
return None
else:
# Print only text files, not extension binaries. Note that
# getsourcelines returns lineno with 1-offset and page() uses
# 0-offset, so we must adjust.
with io.open(ofile, 'rb') as buf: # Tweaked to use io.open for Python 2
encoding, _ = detect_encoding(buf.readline)
return encoding
