Front page | perl.perl5.porters |
Postings from June 2006
On Tue, Jun 20, 2006 at 09:17:35AM +0100, Steve Hay wrote: > demerphq wrote: > >Yes, core perl support for Win32 is much less than what it should be, > [...] > >The solution to this problem is to improve core support for OS > >abstraction. Something ive been advocating and sending patches in for > >years. And you know what? Generally ive been overruled by people who > >dont know jack shit about win32, or how its apis work. And thats the > >general impression many Win32 people who try to contribute back > >encounter[2]. They contribute a patch, the people who normally would > >apply it are unable or unwilling to assess it with an open mind and > >the patch ends up ignored. They dont bother wasting their time again. In the past few years there is only one Win32 patch that I am aware of that got ignored, which was from BrowserUK for Borland support, and that got ignored by accident due to bad timing - it hit the list just after Jarkko had released 5.8.1, and fell between the cracks. It was applied, once I spotted it. So my memory really doesn't square with your assertion. What examples did I forget? > I've been trying to look at all the Win32 patches that come in for quite > a while now and I don't recall having overruled or ignored any, except > possibly suggestions for including new (Win32) modules into the core, > which I don't feel I could unilaterally do without support from others. I don't remember these specifically. But also I don't remember any great debates about "should/shouldn't" include things. The big "problem" comes down to the need for the core to be conservative, given that we don't want to break things that already work as a side effect of making changes. In particular, signal handling and threads are both wonderfully non-portable, and the more you learn about how to do them, the more you discover that there is to learn, and you start to wonder (well, fear) just how fragile the current arrangement is. (Whilst having no good idea on any less fragile way to replace it) Nicholas Clark