Front page | perl.perl5.porters |
Postings from April 2011
Re: PATCH to perlunicode.pod
Thread Previous
|
Thread Next
From:
Karl Williamson
Date:
April 13, 2011 16:32
Subject:
Re: PATCH to perlunicode.pod
Message ID:
4DA6323D.7060705@khwilliamson.com
On 04/13/2011 01:07 PM, Tom Christiansen wrote:
> Karl Williamson<public@khwilliamson.com> wrote
> on Wed, 13 Apr 2011 09:44:39 MDT:
>
>> I've now pushed to blead, at Jesses's suggestion, all the pod changes.
>> These include things in perlre that address this confusion.
>
> I started out noticing that perlunicode was talking about warnings
> categories in all caps when it shouldn't.
>
> 1296 Perl will warn (using the warning category SURROGATE which is a
> 1297 sub-category of UTF8) if an attempt is made
> 1345 (using the warning category NONCHAR which is a sub-category of UTF8) if
> 1354 NON_UNICODE which is a sub-category of UTF8) if an attempt is made to
>
> I had just planned to generate a patch for those, but then when I got
> in there, I ended up doing a light proofread. The patch below looks
> bigger than it is.
Thanks for your patch. I used all CAPS because I was mimicking some
other pod. Wonder which one(s)...
I agree with most of your edits, but don't know how many can be applied
to 5.14, as it is frozen. I got in a little bit of trouble for making
light edits to pods that I was checking through to make sure 5.14 was
adequately covered, and found things that I thought should be fixed that
weren't directly related to 5.14 issues. As a result I have a bunch of
changes that I haven't submitted (including one especially that I'm
steamed at how long we have lied to people about, because I was a victim
many years ago of believing it, and didn't realize until recently that
the docs are wrong--people shouldn't have to join p5p and get familiar
with the code to find out these things).
Jesse agrees though that the 5.14 code changes should be documented, so
your edits to those should be acceptable; and I believe that regressions
in the new docs should also be acceptable. I know immediately if some
changes fit these categories or not, but others I don't.
>
> Specific things I changed include:
>
> * Specified CWCM instead of making people sift through
> perluniprops; hope that's right.
Actually that's not what I meant, and that is a 5.14 regression, so it's
changeable. Perhaps you know how to rephrase what I did mean, which is
perluniprops adds a note for each property (and each of its aliases)
that matches differently if /i is specified.
>
> * Updated the tr18 ref to the current v13, and made sure that
> we were indeed still current with that. v13 includd the \N{}
> recommendation.
Not really 5.14 related, but your wording about \p \P and \N is much
better than what is there currently.
>
> * Added a footnote to Unicode::LineBreak in the tr18 table.
I wanted to do that myself, but it isn't 5.14 related.
>
> * Tried to fixed the brain-buster about all not unbearded denizens
> not shaving themselves except for the barber's wife.
>
> Sorry. I mean this one:
>
> In C<quotemeta> or its inline equivalent C<\Q>, all characters whose
> code points are above 127 are not quoted in UTF-8 encoded strings, but
> all are quoted in UTF-8 strings.
>
> That (still) makes no sense to me. Here's the wording I came up with that
> reflects what I *thought* it was trying to say:
>
> In C<quotemeta> or its inline equivalent C<\Q>, no characters
> code points above 127 are quoted in UTF-8 encoded strings, but in
> byte encoded strings, code points between 128-255 are always quoted.
That is a 5.14 regression, and I had it wrong. Either the original
email or ticket was wrong about this, or I misread it, and didn't do the
legwork of checking it out. Now that I have, I see that your rewording
is correct.
>
> See previous mail.
>
> Karl and others should probably make sure I am not misstating anything
> in the proffered patch below.
>
> --tom
>
Thread Previous
|
Thread Next