优先考虑多个正则表达式模式

Question

I want to try a second pattern only if the first one didn't match. The use of the Match object is the same in both cases. Is there any way to make it shorter?

for line in cli.splitlines():
    res = re.search('\s*(.*) (.*);', line)
    if res:
        pairs_found[res.groups()[0]] = res.groups()[1]
    else:
        res = re.search('(\w*) (\d*) {', line)
        if res:
            pairs_found[res.groups()[0]] = res.groups()[1]

Answer 1

The following precompiles the regexes into patterns and uses the behaviour of or to evaluate to the first match:

pat1 = re.compile('\s*(.*) (.*);')
pat2 = re.compile('(\w*) (\d*) {')

# or forced into one line:
# pat1, pat2 = map(re.compile, ['\s*(.*) (.*);', '(\w*) (\d*) {'])

for line in cli.splitlines():
    res = pat1.search(line) or pat2.search(line)
    if res:
        pairs_found[res.groups()[0]] = res.groups()[1]

In Python >= 3.8, you can use an assignment expression to shorten it even more:

    if res := (pat1.search(line) or pat2.search(line)):
        pairs_found[res.groups()[0]] = res.groups()[1]

Answer 2

Does this help?

for line in cli.splitlines():
    res = re.search('\s*(.*) (.*);', line)
    if not res: res = re.search('(\w*) (\d*) {', line)
    if res: pairs_found[res.groups()[0]] = res.groups()[1]