On 1/21/19 12:02 PM, Thomas (HFM) Wyant wrote: > Dear Perl Porters, > > > While researching correspondence re my recent blog post on sanitizing > numeric input > (http://blogs.perl.org/users/tom_wyant/2019/01/untrusted-numeric-input.html) > I discovered the following in perlrecharclass as the last paragraph in > "POSIX Character Classes": > > > It is proposed to change this behavior in a future release of Perl so that > whether or not Unicode rules are in effect would not change the behavior: > Outside of locale, the POSIX classes would behave like their ASCII-range > counterparts. If you wish to comment on this proposal, send email to > "perl5-porters@perl.org". > > > In response to this offer, I would like to share the following thoughts: > > > * Is this still the plan? If this is no longer the plan, the paragraph > should be removed. If it is still the plan, I believe more publicity > would be helpful -- say, on https://www.perl.com/. I was able to track > this paragraph as far back as Perl 5.14.0, May 2011, so if it is still > the plan it has been hanging fire for almost 8 years. This is not the plan anymore, and I have removed the paragraph in 4f5c9941bb6f93a967e4cc3ef19c9d39351f0ad3 > > > * I would prefer that the referred-to change NOT be made. We already > have a good half-dozen ways to deal with the "restrict to ASCII range" > problem, most of which are Perl-version-dependent. The justification for > adding yet another is unclear to me. Specifically, what does it get me > that 'use re /a;' (introduced in 5.13.10) does not? > > > * If there are compelling reasons to pursue this, I would appreciate it > if those reasons were made explicit. > > > * If this change is pursued, please try to do it in a way that minimizes > the need for Perl code to be aware of the version of Perl it runs under. > This seems to me to mean tying it to 'use 5.xxxx;', but there may well > be better ways. > > > With thanks for shepherding Perl into its 4th decade (!!!) > > > Tom Wyant >Thread Previous