Question about whitespaces

[Zeta611]

Is there a way to match against tab or newline characters?

Thanks for the great tool!

[art-w]

Hello! The tab characters are replaced by 4 spaces during preprocessing so I'm afraid it's not possible to search for them... which usecase did you have in mind? (you can however search \t for code that uses this escape notation in chars/strings https://sherlocode.com/?q=%5Ct )

For newlines, there is some secret syntax:

• \^ to match the beginning of the line ( https://sherlocode.com/?q=%5C%5Elet )
• \$ to match the end of the line ( https://sherlocode.com/?q=in%5C%24 )

You may also enjoy the undocumented syntax \b for "word boundary" : let's say you are searching for the keyword in, but the matches contain wrong stuff like begin or inside... You could try adding spaces around in, but then it wouldn't find )in or in\n. This is where the \b word boundary is useful : it doesn't match a character, but checks that a word starts/ends at this location, so searching \bin\b would only find real appearance of the word "in" https://sherlocode.com/?q=%5Cbin%5Cb

[Zeta611]

Hi, sorry for the late response.

The use case for matching whitespaces, including newline characters, was for searching patterns that span multiple lines of code. It doesn't seem possible to match patterns using pattern1\$\^pattern2, to match a line ending with pattern1 and starts with pattern2 at the line that follows.

Nonetheless, the word boundary matching \b pattern you mentioned seems to be very useful!

[art-w]

Ha right, sorry no it's not currently possible to match on multiple lines (but it would be nice indeed :) )