On 12/2/2018 3:44 PM, Hans Åberg wrote:

...

On 2 Dec 2018, at 10:19, Henning Hraban Ramm wrote:

STIX fonts don’t contain super and subscript numbers.

They are in STIX Two, so I can get them using: \definefontfamily[stix-two][mono][Stix Two Text] \definefontfamily[stix-two][math][Stix Two Math]

\setupbodyfont[stix-two,10pt]

But it is an entirely different font design, so a font fallback to STIX would be desirable. Then the problem is that these super and subscript symbols are spread on a number of different ranges. —It seems one cannot have a general fallback in the case the glyph is missing.

In addition, they get translated in the typesetting, so when copied in the PDF and pasted into a text file, they loose being super and subscript.

in general these super and subscript are rather useless because (1) only a subset of characters is available and (2) it depends on selectively applying the subs and sups feature anyway, i can provide a solution to automatically add these scripts to a font; just for fun i implemented something % \definefontfeature[default][default][scripts=yes] % \definefontfeature[default][default][scripts=.6] \definefontfeature [default] [default] [scripts={factor=.6,up=.3,down=.2,extend=1.5}] but as i use some new trickery it won't be in the beta for a while. Han s ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------

On 12/2/2018 6:33 PM, Hans Åberg wrote:

...

On 2 Dec 2018, at 16:55, Hans Hagen wrote:

On 12/2/2018 3:44 PM, Hans Åberg wrote:

...

...

On 2 Dec 2018, at 10:19, Henning Hraban Ramm wrote:

STIX fonts don’t contain super and subscript numbers. They are in STIX Two, so I can get them using: \definefontfamily[stix-two][mono][Stix Two Text] \definefontfamily[stix-two][math][Stix Two Math] \setupbodyfont[stix-two,10pt] But it is an entirely different font design, so a font fallback to STIX would be desirable. Then the problem is that these super and subscript symbols are spread on a number of different ranges. —It seems one cannot have a general fallback in the case the glyph is missing. In addition, they get translated in the typesetting, so when copied in the PDF and pasted into a text file, they loose being super and subscript. in general these super and subscript are rather useless because (1) only a subset of characters is available and (2) it depends on selectively applying the subs and sups feature

They are usable in plain text files, though I think, originally added to Unicode as a font encoding compatibility. Right now, I just use it in generated output files to indicate variable relabeling, so from that point of view, a font fallback would suffice, as there is no need to copy and paste from the PDF them as part of computer code examples. One might want to copy and work them by hand, though.

However, after starting to use them, I thought they might be usable in input code, too. Then one would like to paste examples in the manual for copy and paste. The super and superscript parentheses could be used as replacement for TeX ^ and _, the latter which in turn comes from some pre-ASCII arrow symbols. Unicode does not have general super and superscript symbols, and for computer input, it would be better with a matching pair than a single symbol.

I have used the text substitutions on MacOS for some time now, and it works fine to write math styles and symbols directly in a plain text file. I have also seen these super and superscript numbers used for SI units for example, on the Internet. So it may become more common in the future.

...

anyway, i can provide a solution to automatically add these scripts to a font; just for fun i implemented something

% \definefontfeature[default][default][scripts=yes] % \definefontfeature[default][default][scripts=.6]

\definefontfeature [default] [default] [scripts={factor=.6,up=.3,down=.2,extend=1.5}]

but as i use some new trickery it won't be in the beta for a while.

Maybe as a part of general features: A font fallback that selects the character from a sequence of fonts if not available, which could involve ranges, too, but not necessarily. And being able to have untranslated throughput.

Using the available fallback mechanism that already can be done. This super/supscript stuff is a bit more tricky because when a superscript 2 is in the font, and a superscript 3 isn't, using a fake ^3 alongside that ^2 quite certainly will look bad because the 2 is normally of a different design that a scaled down 2, so one then has to replace them all. Then there is the question of scale and move (up/down). Although I have some experiemntal (font) magic on my computer it will probably take till next year before it will show up in a beta. Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------

On 2 Dec 2018, at 19:01, Hans Hagen wrote:

This super/supscript stuff is a bit more tricky because when a superscript 2 is in the font, and a superscript 3 isn't, using a fake ^3 alongside that ^2 quite certainly will look bad because the 2 is normally of a different design that a scaled down 2, so one then has to replace them all. Then there is the question of scale and move (up/down).

This issue shows up with the STIX and STIX2 fonts, as the former has the superscript digits ¹ ² ³, but not the other which came to Unicode later in a different range.

Although I have some experiemntal (font) magic on my computer it will probably take till next year before it will show up in a beta.

I use the code below, which seems fine, as the two fonts are close to each other. But it seems that one cannot add [force=yes] to ensure they all come from the same font as on the command \definefontfallback. This command would otherwise be good to pick replacements for certain glyphs. \definefallbackfamily [mainface] [serif] [stixtwotext] [range={00B2-00B3,00B9,02070-0208F}] \definefallbackfamily [mainface] [mono] [stixtwotext] [range={00B2-00B3,00B9,02070-0208F}] \definefallbackfamily [mainface] [math] [stixtwomath] [range={00B2-00B3,00B9,02070-0208F}] \definefontfamily [mainface] [serif] [stix] \definefontfamily [mainface] [mono] [stix] \definefontfamily [mainface] [math] [stix]

On 6 Dec 2018, at 15:23, Wolfgang Schuster wrote:

The force key is valid for the \definefallbackfamily command, e.g.

\definefallbackfamily [...] [...] [range={...},force=yes]

I tried to test it with the code below, but then it selects the stixtwomath glyphs. So possibly the force override is used only if the font has the glyph, instead of giving an error. \definefallbackfamily [mainface] [math] [stix] [range={00B2-00B3,00B9,02070-0208F},force=yes] \definefontfamily [mainface] [math] [stixtwomath]