Re: [tap] More uncertainty about junk lines

[Andrew Rodland <[e-mail]>]

On Fri, Jun 24, 2011 at 9:20 AM, Ovid <[e-mail]> wrote:

> >________________________________
> >From: Andrew Rodland <[e-mail]>
> >To: "[e-mail]" <[e-mail]>; "[e-mail]" <
> [e-mail]>
> >Sent: Friday, 24 June 2011, 15:08
> >Subject: [tap] More uncertainty about junk lines
> >
> >
> >Once again I'm working on my parser and trying to deal with some
> ambiguities, and I'm realizing that I may be mistaken about the handling of
> junk lines. What does "any line that does not parse must be ignored" in the
> draft spec really mean? Does it mean lines that couldn't possibly be TAP in
> any position (in which case we can still throw an error on out-of-sequence
> TAP) or does it mean any line that can't parse in *this* position, in which
> case there are no true parse errors at all? In other words, is this TAP:
> >1..3
> >ok 1
> >ok 2
> >1..1
> >ok 3
> >valid with an ignored fourth line, or does it have an error on the fourth
> line?
>
>
> That's a very annoying edge case you have there, but I would argue that
> anything which *could* be parsed as a TAP line MUST be parsed as such, thus
> rendering the above invalid. Otherwise, we'd have no idea if you were
> spitting out bad TAP or something else was spitting out junk which happened
> to look like TAP.
>
> Thus:
>
> * If something *can* be parsed as a TAP line, do so, even if it renders the
> TAP document invalid.
> * If something *can't* be parsed as a TAP line, ignore it.
>

I agree entirely. I implemented the opposite behavior in TAP::Spec::Parser
and I've been uncomfortable about it ever since I have. Only allowing
"non-TAP lines" to be ignored as junk seems a lot more sane, in terms of
allowing us to reject malformed TAP and in terms of keeping parsers
manageable. I'm going to change the behavior in TAP::Spec::Parser, write a
few tests for it, and then set about clarifying the spec, maybe at YAPC.

Andrew