Bug #58879 for HTML-Tree: as_HTML change breaks existing code

Sun Jun 27 17:12:54 2010 sprout [...] cpan.org - Ticket created

Subject:	as_HTML change breaks existing code
Date:	Sun, 27 Jun 2010 14:12:43 -0700
To:	bug-HTML-Tree [...] rt.cpan.org
From:	Father Chrysostomos <sprout [...] cpan.org>

HTML-Tree 3.23_1 I agree that it doesn’t make sense for as_HTML to tack "\n" on the end, but I also think it’s too late to change it now. The amount of code using HTML::Tree is huge (see http://deps.cpantesters.org/depended-on-by.pl?dist=HTML-Tree-3.23) and I know that at least some of it breaks with this change. Maybe you could provide an option to omit the "\n". Or perhaps the version number could be increased to 4, and the new behaviour enabled by HTML::Element->new( -api => 4 ) or some such.

Sun Aug 01 23:10:09 2010 Jeff.Fearn [...] gmail.com - Broken in 3.23_1 added

Sun Aug 01 23:10:09 2010 Jeff.Fearn [...] gmail.com - Correspondence added

I'm generally against supporting invalid behaviour, even if it's widely used. You just end up with a mess that makes adding new features impossible. Now I know that sometimes the cost of doing things properly is too high, can you give an examples of the cost of this change being too high to bear?

Sun Aug 01 23:10:10 2010 The RT System itself - Status changed from 'new' to 'open'

Sun Aug 01 23:10:10 2010 Jeff.Fearn [...] gmail.com - Reference to ticket #41739 added

Sun Aug 01 23:22:43 2010 sprout [...] cpan.org - Correspondence added

Subject:	Re: [rt.cpan.org #58879] as_HTML change breaks existing code
Date:	Sun, 1 Aug 2010 20:22:33 -0700
To:	bug-HTML-Tree [...] rt.cpan.org
From:	Father Chrysostomos <sprout [...] cpan.org>

On Aug 1, 2010, at 8:10 PM, Jeff Fearn via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=58879 > > > I'm generally against supporting invalid behaviour, even if it's widely > used. You just end up with a mess that makes adding new features impossible. > > Now I know that sometimes the cost of doing things properly is too high, > can you give an examples of the cost of this change being too high to bear?

It‘s never going to be too high for *me* to bear. It’s just time-consuming to correct ‘working’ code. I have some code that does chop() on the return value. Since HTML::Tree is so widely used, I can’t imagine I’m the only one. Then again, maybe I am. :-)

Sun Sep 19 02:38:16 2010 Jeff.Fearn [...] gmail.com - Correspondence added

On Mon Aug 02 13:22:43 2010, sprout@cpan.org wrote: Show quoted text

> > On Aug 1, 2010, at 8:10 PM, Jeff Fearn via RT wrote: >

> > <URL: https://rt.cpan.org/Ticket/Display.html?id=58879 > > > > > I'm generally against supporting invalid behaviour, even if it's

> widely

> > used. You just end up with a mess that makes adding new features

> impossible.

> > > > Now I know that sometimes the cost of doing things properly is too

> high,

> > can you give an examples of the cost of this change being too high

> to bear? > > It‘s never going to be too high for *me* to bear. It’s just time- > consuming to correct ‘working’ code. > > I have some code that does chop() on the return value. Since > HTML::Tree is so widely used, I can’t imagine I’m the only one. Then > again, maybe I am. :-) >

This change was in response to a bug opened because adding a new line is invalid behaviour. While it may affect some use cases I think the original bug is correct and this behaviour should be changed to not output a spurious new line. You have the chop there because it's wrong to add the new line, so ironically enough this fix is only likely to affect people who have worked this issue in their own code :/ Here's hoping most of them used chomp instead of chop :) Cheers, Jeff.

Sun Sep 19 02:38:18 2010 Jeff.Fearn [...] gmail.com - Status changed from 'open' to 'rejected'