Re: code provenance

4 messages Options
Embed this post
Permalink
Markus Neteler-2

Re: code provenance

Reply Threaded More More options
Print post
Permalink
Hi Scott
(cc grass-psc to illustrate how (easy) the missing work is done)

On 4/18/07, Scott W Mitchell <[hidden email]> wrote:
...
> Looking into it today, I see from
> http://wiki.osgeo.org/index.php/GRASS_Provenance_Review
> that it's mostly done, but it looks like
> there's just the tedious but simple step of inserting license
> statements into a bunch of main.c files.  Am I interpreting the
> status correctly?

yes, it's just a bit time consuming. After several (hundred(s)) headers
I was shifting away...

> My term just ended, I'm getting the marking under control, and am
> available to help on this if that would be valuable.  Is it as simple
> as going through a list of files and adding the text?  Is that list
> already generated?

Nope, but it's (semi-)automatically done:

- be online with CVS connection possible
- have 'cvs2cl.pl' in path (http://www.red-bean.com/cvs2cl/cvs2cl.pl)
- edit tools/copywrite.pl for your preferred editor
- run:
    perl tools/copywrite.pl contributors.csv

It will run through the GRASS source code and search for main.c files which
are yet lacking a copyright statement. In the editor it will open 3 files:
- main.c itself where it inserts already a template header to be fine tuned
- description.html for verification of authors of pre-CVS times (CERL etc)
- local ChangeLog.tmp file generated on the fly

To do then in the editor:
- break long lines in this template header
- add missing authors from description.html, special attention to the
  first author (all CERL code needs manual addition of CERL author
  as first contributor since it was before CVS and doesn't appear in
  the ChangeLog.tmp file)
- remove old header is it is replaced by new, keep other
  relevant comments (don't polish too much!)

That's it. Save and the script will fetch the next candidate.

Once all is done it has to be done for each library, too. Here
I suggest to add it to the most relevant file of each library only,
not to all.

> Please let me know if I can help.

Absolutely!
Thanks
Markus


Scott Mitchell-5

Re: code provenance

Reply Threaded More More options
Print post
Permalink
On 18-Apr-07, at 18:03 , Markus Neteler wrote:

> Hi Scott
> (cc grass-psc to illustrate how (easy) the missing work is done)

Thanks for the instructions!  One more question - does this need to  
be done in the head of the grass6 CVS tree, in the release branch,  
both, or ____ ... ?

Cheers,
Scott


Markus Neteler-2

Re: code provenance

Reply Threaded More More options
Print post
Permalink
On 4/19/07, Scott Mitchell <[hidden email]> wrote:
> On 18-Apr-07, at 18:03 , Markus Neteler wrote:
>
> > Hi Scott
> > (cc grass-psc to illustrate how (easy) the missing work is done)
>
> Thanks for the instructions!  One more question - does this need to
> be done in the head of the grass6 CVS tree, in the release branch,
> both, or ____ ... ?

Only into CVS HEAD (currently 6.3-CVS).

thanks
markus


Markus Neteler-3

Re: Re: code provenance

Reply Threaded More More options
Print post
Permalink
In reply to this post by Scott Mitchell-5
On Thu, Apr 19, 2007 at 08:56:46AM -0400, Scott Mitchell wrote:
> On 18-Apr-07, at 18:03 , Markus Neteler wrote:
>
> >Hi Scott
> >(cc grass-psc to illustrate how (easy) the missing work is done)
>
> Thanks for the instructions!  One more question - does this need to  
> be done in the head of the grass6 CVS tree, in the release branch,  
> both, or ____ ... ?

Probably I already answered this:
I suggest to only do it in CVS HEAD.

Markus