Allow starting with | in multiline match

[cometkim]

I'm parsing bcp47 language tag using reghex and work perfectly!

Can we accept starting with a | character instead pattern when using a multiline match?

const irregular = match('irregular')`
-  ${/en-GB-oed/}
+ | ${/en-GB-oed/}
  | ${/i-ami/}
  | ${/i-bnn/}
  | ${/i-default/}
  | ${/i-enochian/}
  | ${/i-hak/}
  | ${/i-klingon/}
  | ${/i-lux/}
  | ${/i-mingo/}
  | ${/i-navajo/}
  | ${/i-pwn/}
  | ${/i-tao/}
  | ${/i-tay/}
  | ${/i-tsu/}
  | ${/sgn-BE-FR/}
  | ${/sgn-BE-NL/}
  | ${/sgn-CH-DE/}
`;

[fabiospampinato]

IMHO #17 would be a better fix for the alignment problem you are seeing. Also if basically broken syntax becomes acceptable it will become harder to spot issues, which is already kind of a problem given that there's no syntax highlighting for the DSL.

[kitten]

So, currently I'm actually inclined to add neither 😅

Regarding the extra pipe, I agree, it's an unnecessary addition that will make it harder to read if someone comes into the DSL without being used to it.

The same goes for arrays though. It wouldn't immediately be obvious whether they abstract a sequence or alternations. Furthermore it opens up new problems like sequences in an alternations which currently just works, so that could leave RegHex's DSL in a spot where it'd jump between string DSL and in-JS DSL, which I'd prefer to avoid

[cometkim]

Since the *-in-JS solution uses JS module semantics, it becomes difficult to deal with the problem with Babel. When a user tries to import an external module into the DSL, somethings like code extraction and dependency tracking are may required aditionally. (ex. Linaria does)