mirror of
https://github.com/explosion/spaCy.git
synced 2024-09-22 11:59:14 +03:00
55 lines
1.6 KiB
Python
Executable File
55 lines
1.6 KiB
Python
Executable File
#!/usr/bin/env python
|
|
import sys
|
|
import re
|
|
import os
|
|
import ast
|
|
|
|
|
|
src_dirname = sys.argv[1]
|
|
dst_dirname = sys.argv[2]
|
|
prefix = "test_"
|
|
|
|
|
|
for filename in os.listdir(src_dirname):
|
|
match = re.match(re.escape(prefix) + r"(.+)\.py", filename)
|
|
if not match:
|
|
continue
|
|
|
|
name = match.group(1)
|
|
source = open(os.path.join(src_dirname, filename)).readlines()
|
|
tree = ast.parse("".join(source))
|
|
|
|
for item in tree.body:
|
|
if isinstance(item, ast.FunctionDef) and item.name.startswith(prefix):
|
|
|
|
# only ast.expr and ast.stmt have line numbers, see:
|
|
# https://docs.python.org/2/library/ast.html#ast.AST.lineno
|
|
line_numbers = [x.lineno for x in ast.iter_child_nodes(item)
|
|
if isinstance(x, ast.expr) or
|
|
isinstance(x, ast.stmt)]
|
|
|
|
body = source[min(line_numbers)-1:max(line_numbers)]
|
|
|
|
# make sure we are inside an indented function body
|
|
assert all([re.match(r"\s", l[0]) for l in body])
|
|
|
|
offset = 0
|
|
for line in body:
|
|
match = re.search(r"[^\s]", line)
|
|
if match:
|
|
offset = match.start(0)
|
|
|
|
# remove indentation
|
|
assert offset > 0
|
|
|
|
for i in range(len(body)):
|
|
body[i] = body[i][offset:] if len(body[i]) > offset else "\n"
|
|
|
|
# make sure empty lines contain a newline
|
|
assert all([l[-1] == "\n" for l in body])
|
|
|
|
code_filename = "%s.%s" % (name, item.name[len(prefix):])
|
|
|
|
with open(os.path.join(dst_dirname, code_filename), "w") as f:
|
|
f.write("".join(body))
|