Wireshark-dev: [Wireshark-dev] Last few license header questions

From: Evan Huus <eapache@xxxxxxxxx>
Date: Wed, 5 Feb 2014 15:54:16 -0500
The buildbot is down to reporting only 13 files with license header
problems. Let's get them cleaned up and finally have that step pass :)

---

'diameter/dictionary.dtd'
'wimaxasncp/dictionary.dtd'

These two DTD files both appear to be extracted from C code at some
point in the past. We already have an 'ignore' rule set up for the
whole "dtds/" directory, so I'm happy ignoring these as well unless
there is some way we should be embedding license information in them?

'epan/dissectors/pidl/mapi/request.cnf.c'
'epan/dissectors/pidl/mapi/response.cnf.c'

These two appear to be Ronnie's. They're not entirely C code so I'm
not sure if embedding a license header is possible. Ronnie, I assume
we're safe to add an exception for these as well?

'epan/reedsolomon.h'
'epan/reedsolomon.c'

These two simply say "May be used under the terms of the GNU public
license" which presumably means GPLv1. I have not been able to find
specific information on compatibility, but it appears that the GPLv1
does not have an "or later" clause. This means we need to
remove/replace/relicense it. I found another reed-solomon C library
(http://sourceforge.net/projects/rscode/) but it's only available
under the GPLv3+ which we also can't use. *sigh*. Anyone have any
ideas?

'epan/dtd_grammar.h'
'epan/dfilter/grammar.h'

These two are generated by lemon, so we can just add an ignore rule for them.

'doc/make-authors-short.pl'
'docbook/custom_layer_pdf.xsl'

This two appear to have been written by Ulf Lamping. The latter is
XML, not C, so (as with the DTD files above) I'm not sure what that
means for its status.

'doc/make-authors-format.pl'
'doc/perlnoutf.pl'

These two appear to have been written by Graeme Hewson.

'autogen.sh'

The very first version was written by Gilbert Ramirez according to
git, but it's undergone so many changes since then I'm not sure how we
should attribute it...

---

So it looks like the only potentially problematic one is going to be
the reedsolomon code. The rest involve either tracking down the
original author, or agreeing (eg for autogen.sh) that we should just
add a GPLv2+ header and be done with it.

Thoughts?

Evan