a small question about our regex parts

[usta]

@captn3m0 @BiNZGi
I have seen our regex lines finishes with always $ but it looks like not correct to me, is it ruby thing or a typo ?
For example linux kernel regex is like
regex: ^v(?0|[1-9]\d*).(?0|[1-9]\d*)(.(?0|[1-9]\d*))?$
but when i try to use this regex on online regex tools it couldnt select correct fields :

On the other hand when i remove $ at the end it selects correct fields :

[BiNZGi]

I am not a regex expert and cannot say much about your issue. @captn3m0 is the right man for that 😄

[usta]

@BiNZGi i am also not an expert but each time i play with it i get more and more fun , it is like a video game which you love it because of you couldnt beat it :D ❤️

[captn3m0]

The $ in a regex denotes end of the line. If you are copy-pasting arbitary text, where the matching text (tag name) might be anywhere, the $ will break things since it expects the matching part to end the string.

This is important since we want to match the "complete" tag in all cases, and not a partial tag. As an example from your screenshot, we want to match v5.18, but not v5.18-rc1. The code that powers our automation lists the tags from the upstream project, and checks them one by one, which is akin to listing all the tags with a newline between them.

So the correct way to use our regexes would be to:

1. Make a list of all the tags. I have a small function for this which I run as get_remote_tags

function get_remote_tags() {
    D=$(mktemp -d)
    cd "$D"
    git init --bare
    git fetch --quiet --tags --filter=blob:none "$1"
    git tag -l
}

2. Copy-pasting the list of tags one line each against the regex.

See this link for example against the linux kernel: https://rubular.com/r/SUaaKX6Z1bGok3

If you remove the $, note that all of the following get selected:

v3.15
v3.15-rc1
v3.15-rc2
v3.15-rc3
v3.15-rc4
v3.15-rc5
v3.15-rc6
v3.15-rc7
v3.15-rc8

which is not what we want