mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-06 12:20:52 +01:00
27c7158166
Summary: There is a module called `2to3` which you can target for future specifically to remove these, the directory of `caffe2` has the most redundant imports: ```2to3 -f future -w caffe2``` Pull Request resolved: https://github.com/pytorch/pytorch/pull/45033 Reviewed By: seemethere Differential Revision: D23808648 Pulled By: bugra fbshipit-source-id: 38971900f0fe43ab44a9168e57f2307580d36a38
97 lines
2.9 KiB
Python
97 lines
2.9 KiB
Python
## @package parser
|
|
# Module caffe2.python.docs.parser
|
|
|
|
|
|
|
|
|
|
import re
|
|
|
|
|
|
class Parser(object):
|
|
# List of tuples (regex_str, lambda(regex_match, formatter))
|
|
# If a lambda returns True it will be called repeatedly with replacement
|
|
# otherwise it will only be called on text that hasn't been parsed yet.
|
|
regexes = [
|
|
# Code blocks of various formats
|
|
('````(.+?)````',
|
|
lambda m, f: f.addCode(m.group(1))
|
|
),
|
|
('```(.+?)```',
|
|
lambda m, f: f.addCode(m.group(1))
|
|
),
|
|
(r'((( {2})+)(\S.*)(\n\s*\n|\n))+',
|
|
lambda m, f: f.addCode(m.group(0))
|
|
),
|
|
(r'([^\.])\n',
|
|
lambda m, f: f.addRaw('{c} '.format(c=m.group(1))) or True
|
|
),
|
|
('`(.+?)`',
|
|
lambda m, f: f.addCode(m.group(1), True)
|
|
),
|
|
# Make links clickable
|
|
('http[s]?://(?:[a-zA-Z]|[0-9]|[$-_@.&+]'
|
|
r'|[!*\(\),]|(?:%[0-9a-fA-F][0-9a-fA-F]))+',
|
|
lambda m, f: f.addLink(m.group(0), m.group(0))
|
|
),
|
|
(r'\*\*(.+?)\*\*',
|
|
lambda m, f: f.addEmphasis(m.group(1), 2)
|
|
),
|
|
(r'\*(.+?)\*',
|
|
lambda m, f: f.addEmphasis(m.group(1), 1)
|
|
),
|
|
]
|
|
|
|
def __init__(self, text, formatter):
|
|
self.text = text
|
|
self.lines = []
|
|
self.formatter = formatter
|
|
|
|
def parseText(self):
|
|
UNPARSED = 0
|
|
PARSED = 1
|
|
parsed_block = [(UNPARSED, self.text)]
|
|
for regex, func in self.regexes:
|
|
index = 0
|
|
while index < len(parsed_block):
|
|
label, text = parsed_block[index]
|
|
|
|
# Already been parsed
|
|
if (label == PARSED):
|
|
index += 1
|
|
continue
|
|
|
|
match = re.search(regex, text)
|
|
if match:
|
|
parsed_block.pop(index)
|
|
start = match.start(0)
|
|
end = match.end(0)
|
|
|
|
f = self.formatter.clone()
|
|
merge = func(match, f)
|
|
|
|
if merge:
|
|
merged = text[:start] + f.dump() + text[end:]
|
|
parsed_block.insert(index, (UNPARSED, merged))
|
|
else:
|
|
if text[:start]:
|
|
parsed_block.insert(index,
|
|
(UNPARSED, text[:start]))
|
|
|
|
index += 1
|
|
parsed_block.insert(index, (PARSED, f.dump()))
|
|
|
|
index += 1
|
|
if text[end:]:
|
|
parsed_block.insert(index,
|
|
(UNPARSED, text[end:]))
|
|
|
|
else:
|
|
index += 1
|
|
|
|
self.lines += [i for _, i in parsed_block]
|
|
self.text = ' '.join(self.lines)
|
|
|
|
def parse(self):
|
|
self.parseText()
|
|
return self.text
|