Datasets:
jblitzar
/
github-python-corpus

Dataset card Data Studio Files
xet
Community
text
stringlengths
1
93.6k
img = img.convert('RGB')
if self.transform is not None:
img = self.transform(img)
return img, self.images[i][1]
if __name__ == '__main__':
main()
# <FILESEP>
"""
The preprocessing tokens in C are particularly hard, and
this character stream makes it slightly easier.
"""
class CharacterStream(object):
def __init__(self, generator, line=1, filename=""):
self.comments = True
self.generator = discard_comments(self, logical_characters(self, generator))
self.line = line
self.filename = filename
self.character = '\n' # Marks beginning of new line
# Doesn't go through line incrementing, so
# it is sufficient for denoting beginning of the first line.
def get_next(self):
assert self.character, "error in tokenizing"
character = self.character
self.character = pull(self.generator)
return character
def is_space(self):
return self.character in spaces
def skip_spaces(self):
while self.character in spaces:
self.get_next()
def skip_spaces_and_newlines(self):
while self.character in spaces_and_newlines:
self.get_next()
@property
def position(self):
return self.line, self.filename
def logical_characters(stream, sequence):
backslash = False
for ch in sequence:
if backslash:
backslash = False
if ch == "\n":
stream.line += 1
continue
yield "\\"
if ch != "\\":
yield ch
continue
elif ch != "\\":
if ch == "\n":
stream.line += 1
yield ch
else:
backslash = True
if backslash:
yield "\\"
def discard_comments(stream, sequence):
"C tokenizer is fat!"
state = 0
for ch in sequence:
if state == 0:
if ch != '/':
yield ch
elif stream.comments:
state = 1
else:
yield ch
elif state == 1:
if ch == '/':
state = 2
elif ch == '*':
state = 3
else:
yield '/'
yield ch
state = 0
elif state == 2 and ch == '\n':
yield ch
state = 0
elif state == 3 and ch == '*':
state = 4
elif state == 4:
if ch == '/':
yield ' '
state = 0
elif ch != '*':
state = 3
def pull(generator):