在Python中使用re.sub获取组匹配

Question

In Perl, it is possible for me to do a substitution and capture a group match at the same time. eg

my $string = "abcdef123";
$string =~ s/(\d+)//;
my $groupMatched = $1; # $groupMatched is 123

In Python, I can do the substitution using re.sub function as follows. However, I cannot find a way to capture the \\d+ group match without invoking another function re.match and performing an additional operation.

string = "abcdef123"
string = re.sub("(\d+)", "", string)

Does anyone know how I can capture the "\\d+" matched value as a separate variable from the same re.sub operation? I tried the following command and it doesn't work.

print r'\1'

Answer 1

You can cheat and pass a function to re.sub :

results = []
def capture_and_kill(match):
    results.append(match)
    return ""
string = "abcdef123"
string = re.sub("(\d+)", capture_and_kill, string)
results[0].group(1)
# => '123'

Answer 2

You can do the following:

sub_str = re.search("(\d+)", str).group(1)

Will find the "123" part.

Then you replace it:

str = str.replace(sub_str, "")

Note that if you have more than [0-9] sequence you'll need to use findall and iterate manually on all matches.

Answer 3

below code tested under python 3.6.

test = "abcdef123"
resp = re.sub(r'\w+[A-Za-z](\d+)',r'\1',test)
print (resp)

123

Answer 4

To build on Marouns answer you can also do the following:

str = "foobarzoo"
match = re.search(r"foo(bar)", str)

Then replace it with the full match "foobar":

str = str.replace(match.group(0), "")

The first group (within the brackets) "bar" is then accessible by:

match.group(1)

Using this method str will then finally equal "zoo"