From: Tony G. <Ton...@Su...> - 2006-02-14 17:53:34
|
Stefan Seefeld <se...@sy...> writes: ... >> xmlroff depends on PangoXSL for the definition of some additional >> PangoAttribute types, and it actually uses one of those types. >> PangoXSL is not a replacement for Pango. If anything, it is a layer on top >> of >> Pango. > > Thanks for the explanation ! Is anybody else using / depending on pangoxsl ? Nothing else. > I'm just wondering whether for the purpose of packaging it would be possible > to bundle pangoxsl and xmlroff, so users don't have to care. There's the slight issue of licenses. LGPL in PangoXSL could be seen as an continuation of LGPL in PangoPDF, which was necessary because PangoPDF was a HBPP (half-baked Pango port, to further institutionalise someone's passing comment). PangoPDF needed to be LGPL because it did more than use Pango's public interfaces. PangoXSL defines several structs which are very similar to the structs for other PangoAttribute types that are defined in Pango's pango-attributes.h. I really don't know whether the fact that PangoXSL uses only what's public in pango-attributes.h would allow PangoXSL to use a different license or whether it uses enough of pango-attributes.h that PangoXSL is definitely a derivative work. IANAL, but this paragraph from Section 5 of the LGPL gives some hope that PangoXSL could be relicensed under a different license since PangoXSL really only uses or reuses data structure layouts: If such an object file uses only numerical parameters, data structure layouts and accessors, and small macros and small inline functions (ten lines or less in length), then the use of the object file is unrestricted, regardless of whether it is legally a derivative work. (Executables containing this object code plus portions of the Library will still fall under Section 6.) If it could be relicensed, then it could be included in the xmlroff distribution. Regards, Tony. |