Emacs Regexp syntax clarification

Question

I am a bit confused on some aspects of Emacs regexps. Specifically with the paragraph on this page

1. ‘\W’ matches a character which is not within a word
2. ‘\B’ matches characters which are not a word boundary
3. ‘\`’ matches the beginning of the whole input
4. ‘\'’ matches the end of the whole input

Can someone give examples of where each of these might apply?

Also suppose for a generic regexp R, if I augment it as ^R$ what does that mean? Does it mean it matches the expression at the beginning and end of sentences. Elsewhere on this site I have seen the regexp ^\s-$ being used to detect blank lines.

I understand each of ^ and $ in isolation but not when used together

Answer 1

^ matches an empty string at a beginning of a line, $ at an end of a line. Therefore, ^R$ matches a line that contains the single character R.

To experiment, you can try

(string-match "\\W" "abc d")

The backslash needs to be doubled in double quotes.

It returns 3, because the space is not within a word (and the first character has index 0).

\B matches inside a word, so for example, the "b" would match in " abc".

I'm not sure how to explain the very beginning and end of the input. string-match will always return 0 for \` and the length of the string for \'. So, if the string contains no newlines, ^ and \` are equivalent, as are $ and \'. If the string contains newlines, ^ matches at each line's start and $ at each line's end, while \` still matches at the very beginning of the string only, and similarly for \' and the very end.