Hi Rob, On Tue, 23 Jul 2013 18:15:07 +0100 Rob Dixon <rob.di...@gmx.com> wrote:
> On 23/07/2013 14:39, Shlomi Fish wrote: > > Hi Rob, > > > > I recommend against using XML::XPath because it's been undermaintained, is > > slower than XML::LibXML's XPath support, may be more incomplete and I > > believe it has poorer support for XML namespaces. > > > > Instead one should use > > https://metacpan.org/module/XML::LibXML::XPathContext , which is part of > > XML-LibXML, which for completeness' sake I'd like to note that I am its > > primary maintainer, so I may be biased. But recommending to use XPath for > > that ( http://en.wikipedia.org/wiki/XPath ) is a good idea. > > Hello Shlomi > > Thank you for declaring your interest in the subject of your evangelism. I apologise if you felt that I was needlessly promoting my modules here. That was not my intention. I recommended against using XML::XPath because it is commonly known as inferior, and has not had a release since 2003 and was suggesting XML::LibXML instead, because it is known to be a better alternative for most of XML::XPath's use-cases (including the one in the original thread). I'd like people to avoid recommending XML::XPath in this day and age. > As you say, XML::LibXML is *your* module (or at least, you are > maintaining it) just as perl-begin.org is *your* website, and common > wisdom says that in such circumstances you should refrain from promoting > either of them altogether. Why do you feel that I've been "promoting" XML::LibXML in this thread? Why does the fact that I'm affiliated with it, prevent me from recommending it over a different alternative, which I believe (and can prove) that is inferior? Just for the record, XML::LibXML is https://en.wikipedia.org/wiki/Free_and_open-source_software (under the same terms as Perl itself) and I encourage other people to use it, build upon it, and contribute to it without the need to compensate me in any way. > If you insist on doing so anyway then I wish > you would make your declarations a *lot* more prominent, i.e. the > *first* thing you say in your post, rather than a subsidiary clause in a > secondary paragraph. Why do you feel I should do so? What was wrong with the disclaimer as it stood? > > I try to avoid recommending XS-based modules when I sense that the OP > may have trouble digging himself out of a hole when a CPAN module has > failed to install. Well, XML::XPath depends on XML::Parser which is an XS module (and not a core one): * https://metacpan.org/source/MSERGEANT/XML-XPath-1.13/Makefile.PL > XML::XPath works fine here. It is plenty fast enough, > and the data doesn't use namespaces. You have no reason to disparage it. > The original poster may need to use namespaces, and he may run into a bug that has crept in XML::XPath since its last release in 2003, and the data may be larger than you are trying it on. As a result, I can no longer recommend it in the general case. Regards, Shlomi Fish -- ----------------------------------------------------------------- Shlomi Fish http://www.shlomifish.org/ Optimising Code for Speed - http://shlom.in/optimise He who reinvents the wheel, may actually invent a much better wheel. Please reply to list if it's a mailing list post - http://shlom.in/reply . -- To unsubscribe, e-mail: beginners-unsubscr...@perl.org For additional commands, e-mail: beginners-h...@perl.org http://learn.perl.org/
