Test, validate and match regular expressions
Matches content immediately following an equals sign (=) up to, but not including, the next tab or newline character. Uses a positive lookbehind to skip the '=' delimiter, and lazy matching combined with lookaheads to stop at tab or newline boundaries without consuming them. Ideal for extracting values from key=value pairs such as spreadsheet formulas, TSV data, or configuration files.