pyparsing自定义解析规则

1.Word(token)

用于匹配由允许的字符集组成的单词,常见的错误是使用特定字符串Word("expr"）匹配"expr"

- L {alphas}  字母
 - L {nums}  数字
 - L {alphanums} 数字字母混合

2.Suppress

忽略表达式中内容

import pyparsing as pp

source = "a , b, c, d"
wd = pp.Word(pp.alphas)
wd_list = wd + pp.ZeroOrMore(','+ wd)
print wd_list.parseString(source)
# result ['a', ',', 'b', ',', 'c', ',', 'd']

# ZeroOrMore
wd_list = wd +pp.ZeroOrMore(pp.Suppress(',')+wd)

print wd_list.parseString(source)
# ['a', 'b', 'c', 'd']

3. Group 

使用group将返回的结果，使匹配的合成一个字符串

from pyparsing import *

wd = Word(alphas)
comma = Literal(",")
greetee = OneOrMore(wd)
end = oneOf("! ?")
greeting = wd + comma + greetee + end
# result::['Hello', ',', 'World', '!']
print greeting.parseString("Hello,World!")
wd = Group(Word(alphas))
comma = Literal(",")
greetee = OneOrMore(wd)
end = oneOf("! ?")
greeting = wd + comma + greetee + end
# [['Hello'], ',', ['World'], '!']
print greeting.parseString("Hello,World!")

from pyparsing import *

wd = Word(alphas)
comma = Literal(",").suppress()
greetee = OneOrMore(wd)
end = oneOf("! ?").suppress()
greeting = wd + comma + greetee + end
#['Hello', 'World']
print greeting.parseString("Hello,World!")

 4. setResultsName 给每一个token匹配起一个漂亮的名字

给匹配的token起一个名字，方便在解析后的ParseResults对象中像字典一样调用

from pyparsing import *

integer = Word(nums)
date_str = (integer("year")+'/'+integer("month")+'/'+integer("day"))
# integer("year") equivalent to interger.setResultsName("year")
data = date_str.parseString('2019/04/17')

# year,type:<type 'str'>,value:2019
print('year,type:%s,value:%s' %(type(data.year),data.year))

5 setParseAction 对每个解析的token进行处理

处理的方法可以自定义，其中三个参数见下

- s   = the original string being parsed (see note below) # 原字符串
- loc = the location of the matching substring # 匹配的token所处位置
- toks = a list of the matched tokens    # 匹配的token列表

比如想对上例中的日期转为int数字，可以自定义一个parseAction，如下：

from pyparsing import *

integer = Word(nums).setParseAction(lambda s,lo,tokens:int(tokens[0]))

date_str = (integer("year")+'/'+integer("month")+'/'+integer("day"))
data = date_str.parseString('2019/04/17')

# year,type:<type 'int'>,value:2019
print('year,type:%s,value:%s' %(type(data.year),data.year))

6. parseString 解析传入的字符串

   str：第一个参数传入需要解析字符串

　parseAll: 第二个参数是否为完全匹配解析。1.解析配置的模式必须与字符串一致，否则会报错。2.匹配的tokens放置在tokens列表中，在上面定义parseAction时，使用tokens[0]，因为token中只有一个匹配的token，但在此种模式下，tokens中可能存在多个token

7. delimitedList 只需要传入一个匹配格式，就可以 Word,Word....若干个匹配，默认每个Word使用逗号断开

om pyparsing import Word, alphas, alphanums, Combine, oneOf, Optional, delimitedList, Group, Keyword

testdata = """
  int func1(float *vec, int len, double arg1);
  int func2(float **arr, float *vec, int len, double arg1, double arg2);
  """
# function retun type is alphas and function name is number,alphas and _
ident = Word(alphas, alphanums + "_")
# define var: var type and before var name *.
vartype = Combine( oneOf("float double int char") + Optional(Word("*")), adjacent = False)
# return type and var name or  * var name
arglist = delimitedList(Group(vartype("type") + ident("name")))

functionCall = Keyword("int") + ident("name") + "(" + arglist("args") + ")" + ";"

for fn,s,e in functionCall.scanString(testdata):
    print(fn.name)
    for a in fn.args:
        print(" - %(name)s (%(type)s)" % a)

# output:
# func1
#  - vec (float*)
#  - len (int)
#  - arg1 (double)
# func2
#  - arr (float**)
#  - vec (float*)
#  - len (int)
#  - arg1 (double)
#  - arg2 (double)