def main(self):
"""
IN: self.root
OT: dict. {
lino: [(node_type, node_value), (...), ...], ...
}
lino: int. count from 1 but not consecutive. the linos are
already sorted by ascending order.
node_type: str. refer to _ast class types, e.g. "<class '_ast
.Import'>", "<class '_ast.FunctionDef'>", ...
node_value: str/dict. e.g. 'os.path.abspath', {'os': 'os'}, ...
"""
out = {}
for node in ast_walk(self.root):
if not hasattr(node, 'lineno'):
continue
if node.col_offset == -1:
# 说明这个节点是 docstring
continue
x = out.setdefault(node.lineno, [])
x.append((str(type(node)), self.eval_node(node)))
# sort linos
sorted_out = {k: out[k] for k in sorted(out.keys())}
return sorted_out
def get_lino_indent_dict(self):
"""
IN: self.file
self.root
OT: {lino: indent}
lino: int. count from 1 but not consecutive. the linos are
already sorted by ascending order.
indent: int. the column offset, assert all of them would be
integral multiple of 4, e.g. 0, 4, 8, 12, ... and no
exception.
"""
import re
from lk_utils.read_and_write_basic import read_file_by_line
reg = re.compile(r'^ *')
code_lines = read_file_by_line(self.file)
lino_indent = {}
for node in ast_walk(self.root):
if not hasattr(node, 'lineno'):
continue
line = code_lines[node.lineno - 1]
indent = reg.findall(line)[0]
lino_indent[node.lineno] = len(indent)
# sort linos
sorted_lino_indent = {
k: lino_indent[k]
for k in sorted(lino_indent.keys())
}
return sorted_lino_indent
def _get_class_names(self, file):
try:
with open(file) as fd:
parsed_source = ast_parse(fd.read())
class_names = [n.name for n in ast_walk(parsed_source) if isinstance(n, ClassDef)]
except IOError, e:
raise ClassLoaderError('Error - Couldn\'t open : \'{:}\'. Reason : {:}'.format(file, e.strerror))
def _get_class_names(self, file):
try:
with open(file) as fd:
parsed_source = ast_parse(fd.read())
class_names = [
node.name for node in ast_walk(parsed_source)
if isinstance(node, ClassDef)
]
except IOError, e:
raise ClassLoaderError(
'Error - Couldn\'t open : \'{0}\'. Reason : {1}'.format(
file, e.strerror))
def _get_class_names(self, filename):
class_names = []
try:
with open(filename) as fd:
parsed_source = ast_parse(fd.read().strip(self.ENCODING_DECLARATION))
class_names = [n.name for n in ast_walk(parsed_source) if isinstance(n, ClassDef)]
except IOError as e:
raise ClassLoaderError('Error - Couldn\'t open : \'{:}\'. Reason : {:}.'.format(filename, e.strerror))
except SyntaxError as e:
raise ClassLoaderError('Error - Couldn\'t parse \'{:}\'. Reason: {:}.'.format(filename, e))
return class_names
def get_lineno_indent_dict2(self):
"""
IN: self.root
OT: {lineno: indent}
lineno: int. 行号, 从 1 开始数
indent: int. 列缩进量, 是 4 的整数倍, 如 0, 4, 8, ...
注意有一个特殊值 -1
"""
out = {}
for node in ast_walk(self.root):
if not hasattr(node, 'lineno'):
continue
out[node.lineno] = node.col_offset
return out
def main(self):
"""
IN: self.root
OT: dict. {
lineno: [(node_type, node_value), (...), ...], ...
}
"""
out = {}
for node in ast_walk(self.root):
if not hasattr(node, 'lineno'):
continue
x = out.setdefault(node.lineno, [])
x.append((str(type(node)), self.eval_node(node)))
return out
def get_lino_indent_dict(self):
"""
refer: {lkdemo}/ast_demo.py
NOTICE: 使用 regex 计算行缩进, 不要用 node.col_offset.
为什么: 假设存在以下代码:
1 | def func():
2 | if a == 1:
3 | pass
对第 2 行 `if a == 1:`, 使用正则计算结果是 indent = 4, 而 node.col_offset
的值是 7. 原因在于, node.col_offset 计算的是变量 `a` 的列位置, 而非该行的缩进
位置.
IN: self.file
self.root
OT: {lino: indent}
lino: int. count from 1 but not consecutive. the linos are
already sorted by ascending order.
indent: int. the column offset, assert all of them would be
integral multiple of 4, e.g. 0, 4, 8, 12, ...
"""
from lk_utils.read_and_write_basic import read_file_by_line
lino_indent = {}
code_lines = read_file_by_line(self.file)
reg = re.compile(r'^ *')
for node in ast_walk(self.root):
if not (hasattr(node, 'lineno') and hasattr(node, 'col_offset')):
continue
if node.lineno in lino_indent:
continue
if node.col_offset == -1:
# 说明这个节点是 docstring
continue
lino_indent[node.lineno] = len(
reg.findall(code_lines[node.lineno - 1])[0])
# sort linos
sorted_lino_indent = {
k: lino_indent[k]
for k in sorted(lino_indent.keys())
}
return sorted_lino_indent
def get_lineno_indent_dict(self, file):
"""
IN: self.root
OT: {lineno: indent}
lineno: int. 行号, 从 1 开始数
indent: int. 列缩进量, 是 4 的整数倍, 如 0, 4, 8, ...
注意有一个特殊值 -1
"""
import re
from lk_utils.read_and_write_basic import read_file_by_line
reg = re.compile(r'^ *')
code_lines = read_file_by_line(file)
out = {}
for node in ast_walk(self.root):
if not hasattr(node, 'lineno'):
continue
line = code_lines[node.lineno - 1]
indent = reg.findall(line)[0]
out[node.lineno] = len(indent)
return out
def get_lino_indent_dict2(self):
"""
another way calculating indent by ast column_offset.
IN: self.file
self.root
OT: {lino: indent}
"""
lino_indent = {}
for node in ast_walk(self.root):
if not (hasattr(node, 'lineno') and hasattr(node, 'col_offset')):
continue
if node.lineno in lino_indent:
continue
lino_indent[node.lineno] = node.col_offset
# sort linos
sorted_lino_indent = {
k: lino_indent[k]
for k in sorted(lino_indent.keys())
}
return sorted_lino_indent