如何从字符串的末尾提取 substring？

Question

I am trying to extract the timestamp from the end of my file name.

(Example file name: ABC_xyz_march_2020.xlsx or xyz_mno__20_07_2019.xlsx , xyz_spa_20-07-2019.xlsx , xyz-mar_2019.csv , ABC-dec-5.csv , etc) (output: march_2020 , 20_07_2019 , 20-07-2019 , mar_2019 , dec-5 , etc)

I am using the split() function of string but I do not extract the month in words.

Can anyone suggest a different approach?

Answer 1

I don't often reach for regular expressions, but sometimes that's the right tool for the job:

import re

PATTERNS = [
    re.compile(pattern)
    for pattern in (
        r"([A-Za-z]+_\d{4})\..*",  # {month_name}_{year}
        r"(\d{2}_\d{2}_\d{4})\..*",  # {day}_{month}_{year}
        r"(\d{2}-\d{2}-\d{4})\..*",  # {day}-{month}-{year}
        r"([A-Za-z]+-\d{1,2})\..*",  # {month_name}-day
    )
]

inputs = [
    "ABC_xyz_march_2020.xlsx",
    "xyz_mno__20_07_2019.xlsx",
    "xyz_spa_20-07-2019.xlsx",
    "xyz-mar_2019.csv",
    "ABC-dec-5.csv",
]

outputs = [
    "march_2020",
    "20_07_2019",
    "20-07-2019",
    "mar_2019",
    "dec-5",
]

for filename, expected_output in zip(inputs, outputs):
    for pattern in PATTERNS:
        match = pattern.search(filename)
        if not match:
            continue
        matched_date = match.group(1)
        if matched_date != expected_output:
            raise ValueError(
                f"In date {filename=}, got {matched_date=} instead of {expected_output=}"
            )
        print(f"Looked at {filename=} and found {matched_date=}")
        break

This builds a list of regular expression objects. Then for each input filename, it tries to match the filename against each regexp until one matches. Error handling (or deciding what to do if no patterns match) is left to the reader.