On Sep 3, 2004, at 3:01 PM, guenter geiger wrote:

On Fri, 3 Sep 2004, Hans-Christoph Steiner wrote:

...

So I make a quick web page to document what seems to me the standard layout of files and directories for working with the Pd sources.

http://puredata.org/docs/developer/devlayout

Comments?

Hi hc.

can you explain a bit more about the background of the layout ?

When making the package build systems for Windows and MacOS X, the Makefiles need to reference things in all of those directories. Some developers have been putting things in different places, making it difficult to have one centralized build system. So I am putting this doc out there to try to get everyone on the same page in terms of how the Pd directories are laid out.

A good example is needing to reference either the pd src directory or the bin binary (i.e. -bundle_loader ../../../pd/bin/pd) or packages/darwin_app/Makefile calling make in each of the different directories and copying files from them.

.hc

________________________________________________________________________ ____

"[W]e have invented the technology to eliminate scarcity, but we are deliberately throwing it away to benefit those who profit from scarcity." -John Gilmore

On Sep 3, 2004, at 5:01 PM, Frank Barknecht wrote:

Hallo, Hans-Christoph Steiner hat gesagt: // Hans-Christoph Steiner wrote:

...

So I make a quick web page to document what seems to me the standard layout of files and directories for working with the Pd sources.

http://puredata.org/docs/developer/devlayout

Comments?

I don't like version numbers in direcory names at all, I think this is a wrong thing to do (at least on the conceptual level). Versions are what tagging and branching in CVS should cope with, IMO. The rest is fine with me.

(I suppose, the old "build" directories should move to packages/build or similar.)

The pd directories with version numbers are the ones from Miller. They untar.gz that way, so it makes sense to leave them named that way and use a symlink. The pd directories checked out from CVS don't need version numbers, that I agree with. I changed the doc to represent that.

.hc

________________________________________________________________________ ____

There is no way to peace, peace is the way. -A.J. Muste

On Sep 4, 2004, at 5:38 AM, guenter geiger wrote:

On Fri, 3 Sep 2004, Hans-Christoph Steiner wrote:

...

On Sep 3, 2004, at 5:01 PM, Frank Barknecht wrote:

...

Hallo, Hans-Christoph Steiner hat gesagt: // Hans-Christoph Steiner wrote:

...

So I make a quick web page to document what seems to me the standard layout of files and directories for working with the Pd sources.

http://puredata.org/docs/developer/devlayout

Comments?

I don't like version numbers in direcory names at all, I think this is a wrong thing to do (at least on the conceptual level). Versions are what tagging and branching in CVS should cope with, IMO. The rest is fine with me.

(I suppose, the old "build" directories should move to packages/build or similar.)

The pd directories with version numbers are the ones from Miller. They untar.gz that way, so it makes sense to leave them named that way and use a symlink. The pd directories checked out from CVS don't need version numbers, that I agree with. I changed the doc to represent that.

If you want to establish a standard, isn't that counter productive having all these different pd sources lying around ? It would be bad if developers would start to depend on them. I think this should not be specified at all, only the existance of the pd directory.

You could add a note that the pd directory might be a symlink to a directory containing the pd sources, but where this directory has to be should not be specified, because it doesn't matter in this case.

Good point, I changed the doc to represent that:

http://puredata.org/docs/developer/devlayout

.hc

http://at.or.at/hans/