On Wed, 26 Jan 2005, John-Mark Bell wrote:
And I now have such a response and it is _not_ freely
Seeing as my original version of this hasn't appeared, here's a
better-worded version (i.e. any inconsistencies between this and the
version that's sat in some mail queue somewhere are entirely my fault.
_This_ is the definitive version):
Castle have now changed their minds over this (to some extent, at least).
Simplest, I think, is to quote from the "horses mouth". John Ballance
UnicodeLib is distributed as part of the current C-Tools product.
The !Unicode resource is data used by UnicodeLib elements.
a) It may not be Freely distributed.
b) It may be distributed alongside product requiring it which was
created using a legitimate copy of C-Tools according to the terms of
the C-Tools license.
The Iconv module falls under clause b of the quote above. Therefore, I
have now bundled the !Unicode resource with Iconv - follow the link on the
NetSurf download page to get the latest version of the module, along with
the extra resource files.
The bundled !Unicode contains a small number of extra encoding files over
and above those supplied by Castle. These were created by myself and add
support for a couple of extra codepages. They are the
Microsoft.CP1256 and Microsoft.CP874 files within the !Unicode resource.
Therefore, it's worth RO5 users updating to this version as well as those
who don't have the "latest and greatest" !Unicode. Additionally, issues
with these specific encodings should be directed at me.
In addition, there are some 42 encodings that Iconv knows about for which,
there are no corresponding encoding data files in the !Unicode resource.
This is something that may be looked at in future, but I see no pressing
need to do anything about it at present. If anyone wants to do something
about this (or are just curious as to what's missing) then they simply
need to ask, as I've a complete list here.
Hopefully, that clarifies the situation.