Hi! I use $x↦2x$ in xits, and it prints x→2x instead of x↦2x. The same file works with latin modern. It seems that xits-math contains U+21A6 (↦) symbol (at least, kcharselect displays this character in xits-math font), so it should be related to some meta-information.
On Sat, Jun 12, 2010 at 01:19:08PM +0400, Yury G. Kudryashov wrote:
Hi!
I use $x↦2x$ in xits, and it prints x→2x instead of x↦2x. The same file works with latin modern. It seems that xits-math contains U+21A6 (↦) symbol (at least, kcharselect displays this character in xits-math font), so it should be related to some meta-information.
ConTeXt try to compose U+21A6 from base part and the regular right arrow (U+2192), since that base part seems to be some un-encoded, Latin Modern specific glyph, you get only U+2192 with XITS. Ideally, it should be done the other way around, fake U+21A6 in Latin Modern virtual font and then use the proper code point everywhere (may be this what is actually done but I misinterpreted it). -- Khaled Hosny Arabic localiser and member of Arabeyes.org team Free font developer
Khaled Hosny wrote:
On Sat, Jun 12, 2010 at 01:19:08PM +0400, Yury G. Kudryashov wrote:
Hi!
I use $x↦2x$ in xits, and it prints x→2x instead of x↦2x. The same file works with latin modern. It seems that xits-math contains U+21A6 (↦) symbol (at least, kcharselect displays this character in xits-math font), so it should be related to some meta-information.
ConTeXt try to compose U+21A6 from base part and the regular right arrow (U+2192), since that base part seems to be some un-encoded, Latin Modern specific glyph, you get only U+2192 with XITS. Where can I find this in the source code?
Ideally, it should be done the other way around, fake U+21A6 in Latin Modern virtual font and then use the proper code point everywhere (may be this what is actually done but I misinterpreted it).
On Sat, Jun 12, 2010 at 09:07:17PM +0400, Yury G. Kudryashov wrote:
Khaled Hosny wrote:
On Sat, Jun 12, 2010 at 01:19:08PM +0400, Yury G. Kudryashov wrote:
Hi!
I use $x↦2x$ in xits, and it prints x→2x instead of x↦2x. The same file works with latin modern. It seems that xits-math contains U+21A6 (↦) symbol (at least, kcharselect displays this character in xits-math font), so it should be related to some meta-information.
ConTeXt try to compose U+21A6 from base part and the regular right arrow (U+2192), since that base part seems to be some un-encoded, Latin Modern specific glyph, you get only U+2192 with XITS. Where can I find this in the source code?
I don't know, I just checked the log, it reports a missing character, whose code is in the private use area, and the glyphs embedded in the PDF (FontForge is handy here) and I found it embeds U+2192 and the other glyph from LM, but only U+2192 from XITS. -- Khaled Hosny Arabic localiser and member of Arabeyes.org team Free font developer
On 12-6-2010 5:22, Khaled Hosny wrote:
On Sat, Jun 12, 2010 at 01:19:08PM +0400, Yury G. Kudryashov wrote:
Hi!
I use $x↦2x$ in xits, and it prints x→2x instead of x↦2x. The same file works with latin modern. It seems that xits-math contains U+21A6 (↦) symbol (at least, kcharselect displays this character in xits-math font), so it should be related to some meta-information.
ConTeXt try to compose U+21A6 from base part and the regular right arrow (U+2192), since that base part seems to be some un-encoded, Latin Modern specific glyph, you get only U+2192 with XITS.
Ideally, it should be done the other way around, fake U+21A6 in Latin Modern virtual font and then use the proper code point everywhere (may be this what is actually done but I misinterpreted it).
indeed the idea is to have only the right unicode chars (we already fake some in virtual lm) as with much in math we started from what tex used to do Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
On 12-6-2010 11:09, Hans Hagen wrote:
Ideally, it should be done the other way around, fake U+21A6 in Latin Modern virtual font and then use the proper code point everywhere (may be this what is actually done but I misinterpreted it).
indeed the idea is to have only the right unicode chars (we already fake some in virtual lm)
the beta has some more virtual lm chars so this should work ok now Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
participants (3)
-
Hans Hagen -
Khaled Hosny -
Yury G. Kudryashov