极简规则自动补全 Lisp 式括号

括号补全.py

'''
想做对于 Lisp-like 的括号自动补全, 用 py 脚本处理
	先做了测试看(虽然我没学会 clj 没的测), 默认读取 file 输出 new_file
现在的定义的规则如下(其中空行被排除)
	每行代码开头添加 '('
	每行代码结尾, 判断缩进个数比下一行多 n 个
		当 (n >= 0), 在当前行末尾添加 (n+1) 个 ')'
	每行代码结尾, 如果 '(' 数 a, ')' 数 b, 下一行多 c 个缩进
		若 (a-b-c) 为正, 在该行末尾添加对应个 ')'
	可以有多层的缩进, 对应手动添加一次括号 '('
Jan 16 重新构造代码, 增加单行括号数统计和补全
	添加单行注释, 识别用下面这条正则, 不能处理行尾注释
		r'^\s$|^\s*;'
	用下面的正则大致尝试匹配单行中的单个值, 否则多余的括号会报错
		r'\t+(\S)+\n'
$ cat file
curry0
	curry1 (curry2
			curry3
		curry4
	curry5
curry00
	curry01
	curry02
	curry03
$ cat fi
(curry0
	(curry1 (curry2
			curry3)
		curry4)
	curry5)
(curry00
	curry01
	curry02
	curry03)
$ cat fibo.0
defun fibonacci (N
; google
	d
	if (or (zerop N) (= N 1
		1
	+ (fibonacci (- N 1)) (fibonacci (- N 2
$ cat fibo
(defun fibonacci (N)
; google
	d
	(if (or (zerop N) (= N 1))
		1)
	(+ (fibonacci (- N 1)) (fibonacci (- N 2)))
'''
#!/usr/bin/python3

import sys
import re

def count_indents (line):
	line = str (line)
	count = 0
	for char in line:
		if char == '\t':
			count += 1
			line = line[1:]
		else:
			break
	return count

def add_head (line, current_count, next_count):
	if re.match (r'\t+(\S)+\n|^\t+\[.*\]$', line):
		if current_count >= next_count:
			return line
		else:
			line = line[:current_count] + '(' + line[current_count:]
			return line
	else:
		line = line[:current_count] + '(' + line[current_count:]
		return line

def add_back (line, current_count, next_count):
	tabs = current_count - next_count
	if tabs > 0:
		line = line[:-1] + ')'*tabs + '\n'
	return line

def complete_back (line, current_count, next_count):
	tabs = next_count - current_count
	pre_bracket = 0
	after_bracket = 0
	for char in line:
		if char == '(':
			pre_bracket += 1
		if char == ')':
			after_bracket += 1
	num = pre_bracket - after_bracket - tabs
	if num > 0:
		line = line[:-1] + ')'*(num) + '\n'
		return line
	return line

file_name = sys.argv[1]
new_name = file_name[:-2]
texts = []
indents = []
comments = []
comment_index = []
index = 0

input_text = open (file_name, 'r')
for i in input_text:
	line = str (i)
	if re.match(r'^\s$|^\s*;', line):
		comments.append (line)
		comment_index.append (index)
	else:
		texts.append (line)
		index += 1
		indents.append (count_indents (line))
length = len (texts)
texts.append ('')
indents.append (0)
input_text.close ()

output_texts = ''
for i in range (length):
	for j in range (len (comment_index)):
		if comment_index[j] == i:
			output_texts += comments[j]
	texts[i] = add_head (texts[i], indents[i], indents[i+1])
	texts[i] = add_back (texts[i], indents[i], indents[i+1])
	texts[i] = complete_back (texts[i], indents[i], indents[i+1])
	output_texts += texts[i]

output_text = open (new_name, 'w')
output_text.write (output_texts)
output_text.close ()
print ('Done:', new_name)