Another suggestion would be to leave them as is unless you're running with
'-e' (though again with an option to turn them on in that case if you so
desire).

On Tue, Aug 16, 2016 at 9:25 PM, perl6 via RT <perl6-bugs-follo...@perl.org>
wrote:

> Greetings,
>
> This message has been automatically generated in response to the
> creation of a trouble ticket regarding:
>         "[RFC] Proposal to move grammar related part of error messages
> behind a runtime flag",
> a summary of which appears below.
>
> There is no need to reply to this message right now.  Your ticket has been
> assigned an ID of [perl #128969].
>
> Please include the string:
>
>          [perl #128969]
>
> in the subject line of all future correspondence about this issue. To do
> so,
> you may reply to this message.
>
>                         Thank you,
>                         perl6-bugs-follo...@perl.org
>
> -------------------------------------------------------------------------
> Paraphrasing from IRC where my first comment starts here:
> http://irclog.perlgeek.de/perl6/2016-08-17#i_13035789
>
> My feeling has always been that Perl 6's errors are good, but I'm
> sympathetic to whoever it was that complained that the list of "expecting
> any of" things wasn't all that helpful in the general case. Personally, I
> have never once gotten any useful information out of them, but the messages
> that precede them are very helpful. E.g., perl6 -ne '.say if /asdf \s+ \/'
> gives the error message:
>
> ===SORRY!===
> Regex not terminated.
> at -e:1
> ------> .say if /asdf \s+ \/⏏<EOL>
> Unable to parse regex; couldn't find final '/'
> at -e:1
> ------> .say if /asdf \s+ \/⏏<EOL>
>     expecting any of:
>         infix stopper
>
> The first two parts are really helpful, but the "expecting any of" I
> ignore. I understand that people doing fancy grammar related things
> probably do find them very useful. However, my impression is that most
> people "in the wild" writing Perl (5 or 6) are frequently doing command
> line text manipulation and things like that (akin to my example). And for
> them, the first part of the error message is extremely useful and the
> second part isn't (IMHO).
>
> So I propose turning the second part (i.e, "expecting any of" and things
> like that) off in the general case and adding a flag to turn it on
> (--grammar-errors?), like we already have for --ll-exceptions.
>

Reply via email to