Re: [Jsonpath] Fwd: I-D Action: draft-bormann-jsonpath-iregexp-04.txt

[Tim Bray <[e-mail]>]

On Mon, Apr 25, 2022 at 9:55 AM Glyn Normington <
[e-mail]> wrote:

> I have one question about I-Regexp. Must implementations raise errors for
> regular expressions which do not conform to the I-Regexp syntax? That
> wasn't clear to me from the spec.
>
Hmm, good question.  At the moment, we've got Carsten's specification of
the syntax and then we outsource the semantics to XSD with a MUST.  That's
all we say about implementation behavior. Speaking under my lazy-person
hat, I worry about getting on a slippery slope if we start trying to
specify implementation behavior.

Hmm, I guess we do also at least imply that evaluating an I-Regexp against
a piece of text should produce a boolean result.

I guess the minimal thing we could do would be say "I-Regexp
implementations MUST report violations of the syntax specified in this
document."

But that could be a significant extra effort for developers, so I probably
would be against it, but only by about 51/49 and wouldn't scream too much
if others thought it a good idea.

>
>