CKTIS / Python
re.findall
Collects all non-overlapping matches, changing the result shape according to capturing groups.
Version: Python 3.14; repository image, production runtime not attested
Syntax
re.findall(pattern, string, flags=0)Collects all non-overlapping matches, changing the result shape according to capturing groups.
A single capturing group returns its matched text rather than the full match.
findall scans left to right, collecting nonoverlapping matches, including empty ones. Without captures it returns whole matches. With one capture it returns captured values and with several it returns tuples. In the example parentheses remove id= from each reported value; the second pattern reports the entire fields. After an empty match a nonempty match can start at the same position, so .*? over a produces an empty string, a, then an empty string.
import re
s = "x1 x2"
assert re.findall(r"x\d", s) == ["x1", "x2"]
assert re.findall(r"x(\d)", s) == ["1", "2"]
assert re.findall(r"(x)(\d)", s) == [("x", "1"), ("x", "2")]
assert re.findall(r"z*", "") == [""]
assert re.findall(".*?", "a") == ["", "a", ""]
Examples
import re
print(re.findall(r"id=(\d+)", "id=3 id=17"))
print(re.findall(r"id=\d+", "id=3 id=17"))
Result
['3', '17'] ['id=3', 'id=17']
findall scans left to right, collecting nonoverlapping matches, including empty ones. Without captures it returns whole matches. With one capture it returns captured values and with several it returns tuples. In the example parentheses remove id= from each reported value; the second pattern reports the entire fields. After an empty match a nonempty match can start at the same position, so .*? over a produces an empty string, a, then an empty string.