Re: Question about writing a CPAN module based on Inline::C

2012-03-12 Thread Michael Roberts
On 3/12/2012 9:55 AM, David Mertens wrote: > > Second, there's no reason you couldn't have a Module.pm file that is > edited by any sort of script at build time, and in which you have > special placeholders. This might lead to a codebase that is more > legible than the standard heredoc-based .pm.PL

Re: Question about writing a CPAN module based on Inline::C

2012-03-12 Thread David Mertens
On Mon, Mar 12, 2012 at 12:38 AM, Michael Roberts wrote: > On 3/12/2012 12:43 AM, David Oswald wrote: > >> Perhaps I should look (at some point) at moving some of the > >> configuration logic out of Makefile.PL and into a CPP.pm.PL file for > >> Inline::CPP. CPP.pm.PL could read the existing CPP.

Re: Question about writing a CPAN module based on Inline::C

2012-03-11 Thread Michael Roberts
On 3/12/2012 12:43 AM, David Oswald wrote: >> Perhaps I should look (at some point) at moving some of the >> configuration logic out of Makefile.PL and into a CPP.pm.PL file for >> Inline::CPP. CPP.pm.PL could read the existing CPP.pm and write out >> the changes much like the way Makefile.PL does

Re: Question about writing a CPAN module based on Inline::C

2012-03-11 Thread David Oswald
>> Wow - that's exactly what I need!  Does MakeMaker just automatically run >> anything named *.PL that it finds?  That's perfect! >> > > So... it's a bit tedious to find, but the documentation for > ExtUtils::MakeMaker ( > http://search.cpan.org/~mschwern/ExtUtils-MakeMaker-6.62/lib/ExtUtils/MakeM

Re: Question about writing a CPAN module based on Inline::C

2012-03-11 Thread David Mertens
On Sun, Mar 11, 2012 at 2:13 PM, Michael Roberts wrote: > On 3/11/2012 8:41 AM, David Mertens wrote: > > The trick in your situation, which you may not know, if that you can have > scripts that dynamically generates your module files at build time. Instead > of Module.pm, you would have Module.p

Re: Question about writing a CPAN module based on Inline::C

2012-03-11 Thread Michael Roberts
On 3/11/2012 8:41 AM, David Mertens wrote: > The trick in your situation, which you may not know, if that you can > have scripts that dynamically generates your module files at build > time. Instead of Module.pm, you would have Module.pm.PL > ,.which will be run by ExtUtils::Ma

Re: Question about writing a CPAN module based on Inline::C

2012-03-11 Thread Michael Roberts
That's exactly what I think *must* be the only solution. Well, if it's not the best solution, at least I know at least one other fool is using it, right? Thanks! Michael On 3/11/2012 12:24 AM, David Oswald wrote: > I don't know if this is necessarily the best strategy in your case (it > may be,

Re: Question about writing a CPAN module based on Inline::C

2012-03-11 Thread David Mertens
Michael, ExtUtils::nvcc (which I wrote) helps users of Inline::C by providing functions that returns the necessary configuration options for using nvcc as your compiler. The trick in your situation, which you may not know, if that you can have scripts that dynamically generates your module files

Re: Question about writing a CPAN module based on Inline::C

2012-03-11 Thread Sisyphus
- Original Message - From: "Michael Roberts" To: Sent: Sunday, March 11, 2012 3:45 PM Subject: Question about writing a CPAN module based on Inline::C I'm toying with the idea of writing a new Image::Magick wrapper that, you know, actually compiles on systems like mine. Here's my