Using a rest api I get the following forms of strings back:
/primerjs-0.0.3-3.tgz
/primerjs-0.0.3.tgz
/0.0.3-16
I want to grab just the 0.0.3 part from the above strings. I came up with the following regex expression:
(\d+\.)+\d*(?!tgz)
and I have tested it on an online regex tester and it seems to grab what I want it to. However the following code only prints ['0.']
text = '/primerjs-0.0.9.tgz'
m = re.findall(r"(\d+\.)+\d*(?!tgz)", text)
print m
what am I doing wrong?
Use a non-capturing group:
(?:\d+\.)+\d*(?!tgz)
See the regex demo
Or, use an alternative pattern:
[/-](\d+\.\d+\.\d+)
See another demo
Both will work well with re.findall with your examples. The first one has no capturing groups, so re.findall will output match values, and the second will output only captured values (Group 1 contents) since re.findall returns capture group contents if capturing groups are defined inside the pattern.
import re
rx = r'[/-](\d+\.\d+\.\d+)'
print(re.findall(rx, '/primerjs-0.0.3-3.tgz /primerjs-0.0.3.tgz /0.0.3-16'))
rx = r'(?:\d+\.)+\d*(?!tgz)'
print(re.findall(rx, '/primerjs-0.0.3-3.tgz /primerjs-0.0.3.tgz /0.0.3-16'))
Output:
['0.0.3', '0.0.3', '0.0.3']
['0.0.3', '0.0.3', '0.0.3']
using parenthesis create object groups. here I selected the 0 group that means give me the whole match
the code:
text = '/primerjs-0.0.9.tgz'
...: m = [x.group(0) for x in re.finditer(r"(\d+\.)+\d*(?!tgz)", text)]
...: print m[0]
'0.0.9'
a better approach is to use search instead of re.findall()
text = '/primerjs-0.0.9.tgz'
...: m = re.search(r"(\d+\.)+\d*(?!tgz)", text).group(0)
...: print m
'0.0.9'
you can even add named group for clarity:
text = '/primerjs-0.0.9.tgz'
...: m = re.search(r"(?P<version>(\d+\.)+\d*(?!tgz))", text).group('version')
...: print m
'0.0.9'
