Testing regular expressions and understanding hits

Enter sample, flag and test text separately

Enter the pattern without enclosing /.../ characters and select the JavaScript flags in your own field. g searches for multiple hits, i ignores upper case, m changes line anchors and s also allows the point to capture line breaks. Which newer flags are available depends on the current browser. Syntax errors appear instead of silently rewriting the pattern.

Recognize hit groups and positions

The result indicates the full hit, numbered groups and named groups. Start and end are zero-based UTF-16 positions; an emoji can therefore occupy more than one position. Empty hits are forwarded in a controlled manner. A maximum of 500 hits are issued so that extensive text tests remain clear.

Control replacements before deployment

Activate the replacement by, for example, $1 for the first group, $& for the entire hit, or $ for a nominated group. Without a global flag, only the first hit is replaced. For more than 500 hits, a potentially misleading partial replacement is not issued. The resulting text file can be downloaded.

JavaScript is not PCRE or RE2

The evaluation uses the RegExp engine of your browser. Search patterns for PHP, Nginx, Apache or Search Console may have different rules. For these tasks you use the linked specialized tools. The syntax reference documented JavaScript RegExp.

Problematic patterns do not slow down the page

The exam runs in a separate worker and is aborted after three seconds. You can also stop them manually. The pattern is limited to 2,000 characters, the test text to 100,000 characters. Very elaborate traceability can still lead to the time limit; then reduce the text or formulate the pattern more clearly.

Frequent questions