On 2015-11-12, Kornel Benko wrote: > Am Donnerstag, 12. November 2015 um 00:13:54, schrieb Kornel Benko > <kor...@lyx.org> >> Am Mittwoch, 11. November 2015 um 22:49:19, schrieb Guenter Milde >> <mi...@users.sf.net> >> > On 2015-11-11, Kornel Benko wrote: >> > > Am Mittwoch, 11. November 2015 um 00:11:26, schrieb Günter Milde
>> > Instead of changing the encoding for and back, the patch below prevents >> > encoding changes whenever the TeX engine is XeTeX or LuaTeX: >> > >> > +1 no needless encoding switches >> > +1 runparams.encoding matches the correct encoding at any time >> > +1 less complicated code >> > >> > As a bonus, LuaTeX & TeX fonts now also works for files with two or more >> > languages with different encodings - with inputenc == "auto", ... > Testing the exports gives now 126 failed tests. This is an improvement > of 6 cases. Improved tests are of type dvi3_texF and pdf5_texF. > I'd say, please commit Günter. Done. As a proof of concept, I also changed lib/doc/el/Intro.lyx to use DejaVu as non-TeX font. A similar approach might fix also other "*systemF" tests. We should, however, get a consensus on the "multiscript non-TeX font" for our manuals. It should be free, widely available, and have a good coverage for non-Latin scripts. DejaVu is just my suggestion, an alternative would be "Linux Libertine" or "Free Serif". Could you post a list of failing tests, please? (Maybe we could have an URL for test results (on the ftp server or wiki), so that people not able to run the test could check the latest results.) Günter
