[SUGGEST] Reform eclipse and eclipse related ports

[Mark Linimon]

On Tue, Oct 18, 2005 at 06:07:25PM +0200, Roman Neuhauser wrote:
> > That highlights my point that IMHO we need two different functionalities:
> > 'search' and 'browse'.  'make search' is barely adequate for searching.
> 
> What are you missing from make search? I'll try and add it if it's
> within reasonable bounds of complexity.

e.g. searching when you don't know the exact spelling of the name.

It's the "dictionary problem" -- how do you find the meaning of a word
when you're not sure what the word is?

> > We have nothing at all for browsing (unless you count reading an entire
> > list of ports in hierarchy as 'browsing', which even an old command-line
> > kind of guy like me thinks is crude).
> 
> Can you define 'browsing'?

"show me the ports that have something to do with the Internet"; "show
me a list of plugins that work with Apache2".
 
> How will the Wes' colleagues find it? You need to be able to find
> a port to install it. If a port is required to make sense of the
> structure, we need a bootstrap mechanism, like something in the
> base.  Like, ls.

Don't be silly.  Neither portupgrade nor cvsup are in base.  People
pkg_install them and then they're good to go.  ls is _not_, by any
stretch of the imagination, an adequate tool unless you already have
a pretty good idea of what it is you're looking for.

I really do not believe that anything with a UI belongs in base, and
I believe that 'search' and 'browse', to be useful to the largest number
of people, need to be UI-based; whether that's as applications, or from
web pages, or more likely, both.

> I would certainly prefer if we considered the fs structure to be the
> primary interface (and treated it accordingly). I'm weird, I know.

It's always going to be the 'primary' interface but if we don't build
tools on top of it, it's simply going to limit the ports tree's usefulness
to people who aren't as hardcore as you or I are.  fwiw, I rely extensively
on a little script that I wrote that greps things out of the contents of
Makefiles.  I am familiar with these kinds of tools.

But, I mean, honestly, I think I can state without much fear of
contradiction that I have as good a working knowledge of what's in the
ports tree as anyone.  Even so, the other day I went looking to see if
there was any port specifically geared to synchronizing file systems on
two peer machines (rather than, e.g., geared to just backing up a file
system).  It was really painful to do that, and it shouldn't have been.
None of the existing tools are even vaguely adequate for that.

mcl