On Tue, 28 Jul 2020 at 17:41, Stephan Witt <st.w...@gmx.net> wrote: > Am 28.07.2020 um 05:02 schrieb Thibaut Cuvelier <tcuvel...@lyx.org>: > > > On Mon, 27 Jul 2020 at 12:34, Pavel Sanda <sa...@lyx.org> wrote: > On Mon, Jul 27, 2020 at 04:56:48AM +0200, Thibaut Cuvelier wrote: > > commit e709a6626e731b460a98ea34d762a776de90b54c > > Author: Thibaut Cuvelier <tcuvel...@lyx.org> > > Date: Mon Jul 27 03:14:38 2020 +0200 > > > > Support the mathbbm font. > > > > Should just be able to read formulae, not add it from the GUI, as > it's mostly redundant with mathds and mathbb. > > Hi Thibaut, > > I suppose that the issue raised by Kornel is not related, as discussed > below? > > why is this font addition needed? > > In order to support more math formulae: \mathbbm supports a few more > doublestroke symbols than the existing fonts (like lower-case characters). > It's not exactly the most useful math font in the world, but not supporting > it made problems in some of my private test cases. (Actually, I mostly need > it to convert properly into MathML, but rendering it properly in LyX was > not much harder.) > > And how is it related to the changes in src/output_docbook.cpp? > > It's not related at all, that was a squashing mistake… That part is just > to ensure that the <abstract> tag is output when it should (the code makes > some inference about which paragraphs belong to the abstract, but it does > not go into the details to see if these are just comments, for instance): > having an empty <abstract> is a validation error in DocBook. > > > The issue raised by Kornel is present on Mac too. git bisect says: >
I reverted the commit so that the master branch still behaves properly. On Tue, Jul 28, 2020 at 05:02:13AM +0200, Thibaut Cuvelier wrote: > > > Hi Thibaut, > > > > > > > I suppose that the issue raised by Kornel is not related, as discussed > > below? > > Not all, but the fact that display of characters is screwed in Math Manual > (at least in linux) is caused by this commit. > > Does the rendering of math manual looks correct on your computer with this > patch? > No, it shows the same patterns as Stephan. > > why is this font addition needed? > > > > > > > In order to support more math formulae: \mathbbm supports a few more > > doublestroke symbols than the existing fonts (like lower-case > characters). > > It's not exactly the most useful math font in the world, but not > supporting > > it made problems in some of my private test cases. (Actually, I mostly > need > > it to convert properly into MathML, but rendering it properly in LyX was > > not much harder.) > > I see, true, we can't do properly lower case. Unfortunately they are not > fixed > with this patch here either. > Either more work refinement is needed or different approach (like adding > the > missing glyphs to dsrom10). > The \mathds command in LaTeX does not generate many lower case characters, only h and k: http://www.ctex.org/documents/packages/math/dsdoc.pdf (Section 3, end of page 3). I'm not sure it would work to add more symbols to dsrom10, unless they generate calls to other LaTeX commands than \mathds for the unsupported symbols. > In general we have policy that patches should be sent to the list first so > people > can react before changes are committed. You don't need to that if you > maintain > some specific area (like docbook in your case, no one can check that > anyway) or > do some tiny uncontroversial change (like typos, etc), but geneneric > additions > like this patch should definitely follow post-before-commit rule... Sorry, I thought this was on the benign side…
-- lyx-devel mailing list lyx-devel@lists.lyx.org http://lists.lyx.org/mailman/listinfo/lyx-devel
