Feature description:
Add a toggle to parse the search query as a regular expression.
Problem solved:
Being unable to find what you’re looking for when searching for common words in big libraries due to hundreds or thousands of unrelated results.
Not being able to find different spellings of the same song with one search query.
Not being able to search for exactly what you’re looking for.
Brought benefits:
Allows users that know regex to narrow down search results efficiently.
Allows users with big libraries to only find what they’re looking for instead of hundreds or thousands of unrelated results.
Could speed up search dramatically by quickly discarding possibly thousands of results with the first characters of the query.
Other application solutions:
None that I’m aware of.
Additional description and context:
Example 1:
I’m looking for all versions of the song Black is the Colour.
However that’s often spelled either
Black is the colour (72 results) or
Black is the color (42 results)
Currently I have to search each of these variants on their own or shorten the search string to Black is the colo (which might no longer be distinct enough with shorter search strings).
With regex support I could simply search Black is the colou?r and find both variants in one go.
Example 2:
I’m looking for the song Liar by Argent but have forgotten the artist.
Searching Liar yields 574 (mostly unrelated) results.
With regex support I could search for ^Liar$ and should only see the 65 songs in my collection that are called exactly Liar.
Even worse is Hurt by Johnny Cash.
Searching Hurt:
1318 results currently while ^Hurt$ ought to only yield the 41 tracks in my collection that are called exactly Hurt.
There are many more cases where being able to specify exactly what you are looking for can be beneficial. If there is no easy way for you to parse search strings as regular expressions, disregard this request. I just think it would be a very powerful feature for advanced users with big libraries.