Re: [Pnet-developers] An automatic cvs pull patch for ml-pnet

[Gopal V]

If memory serves me right, Rhys Weatherley wrote:
> At the cost of adding one more place that must be kept up to date by the 
> ml-pnet developers, for a target audience of exactly zero people.  The result 
> will be very fragile and I don't want to have to spend half of my time fixing 
> "the PULLCVS version doesn't work" bugs because I forgot to add a line to a 
> file I never, ever, use myself.  These kinds of bugs never turn up before the 
> release - always 2 days after when it is too late to fix them.

I don't think too many people would complain if we had an audience of 
exactly zero people ... :-)

No , you don't need to add a line to any file ... it picks up the cvs module
name from the location of the build file ... ie 
class/System.Data/System.Data.build will automatically translate to 
mcs/class/System.Data/ as the CVS module name ...

> But that's the whole point of ml-pnet!  To allow people to download and 
> compile the code from another project.  We don't need to make it easy.  Just 
> *possible*.

My PoV is that ml-pnet is a way of adding libraries for some functionality
to pnet without re-inventing some wheels lying around in CVS elsewhere ...

It's a bad move to tell "Company A" to download another project and
direct his/her attention to what "we don't have , but they do"..
(especially if their website mentions that you can run Eclipse on it !)

> Let's get the thing working, with a good selection of libraries, and then 
> worry about ease of use later when and if a real need demonstrates itself.

If you're too fussy about adding it to the Makefile.am , we could always
put into Makefile.pullcvs ... I've seen that in Mozilla which uses a 
makefile to pull the files from "it's own CVS" ..

Anyway attached is the last & final pullcvs patch ... let's see if someone
wants it .. otherwise it will join dfagen in my frozen/

Gopal
-- 
The difference between insanity and genius is measured by success