Question

Supose I have this string:

a= "hello world hella warld"

and I want to match all coincidences whit the regex:

b='(?P<hel>hell[oa])\s*(?P<wrl>w[oa]rld)'

I can use re.findall(b,a) and get:

[('hello', 'world'),('hella','warld')]

but I really want to get:

[{'hel':'hello','wrl':'world'},{'hel':'hella','wrl':'warld'}]

Mi queston is there some native or easy way to get this in Python?

Second Question:

I wrote a function to get dictionaries:

def findalldict(regex,line):
    matches = []
    match = 1
    c = line
    while match != None and len(c)>1:
        match =re.search(regex,c)
        if match:
            matches.append(match.groupdict())
            c =c[match.end():]
    return matches

but I’m no sure if it’s correct, can you guys see any mistake? or you know a better way to accomplish this?

Answer 1

You can use finditer instead of findall to get an iterator of MatchObjects:

>>> regex = re.compile('(?P<hel>hell[oa])\s*(?P<wrl>w[oa]rld)')
>>> line = "hello world hella warld"
>>> [m.groupdict() for m in regex.finditer(line)]
[{'hel': 'hello', 'wrl': 'world'}, {'hel': 'hella', 'wrl': 'warld'}]