The visualisation of the tag characters and \XeTeXglyph

[jsbien]

The editor of a paper of mine has some problem with typesetting the visualisation of the tag characters. I hoped they can be access directly with \XeTeXglyph but it seems I'm wrong. "a tag" is at slot 1114605, but \XeTeXglyph does not accept such an argument. Any suggestions?

[psb1558]

The Unicode tag characters have the type "default ignorable," so they are invisible and zero-width, like (for example) the zero-width non-joiner. That makes them hard to manage in word processors, so Junicode has a parallel set of visible tags. These have no code points but instead are accessible via entities consisting of an ampersand, two underscores, the tag identifier, and a semicolon. So &__s; gives you the visible tag character corresponding to \char"0E0073:

Apply ss10 to this to convert the visible tag to the invisible Unicode tag and resolve the character+tag+tag sequences listed in the Junicode Manual Chapter 5.

My thought, when setting up this system, was that users could use the visible tags for ease of editing (not generally necessary but perfectly okay in XeTeX), and then apply ss10 to the whole document for presentation.

In sum: instead of \char"0E0073, use &__s; without ss10 to get the visible tag.

[jsbien]

I know this and this works for me but for unknown reasons wasn't working elsewhere. The editor is quite fluent in font tools, so he circumvented the problem by extracting the required glyphs from Junicode into a new font. From my point of view the problem is solved.

[psb1558]

I'm glad your editor found a workaround, but it sounds like an awkward one. Problems with Junicode are often caused by an old version of the font hanging around in the TeX installation. If not, and your editor is willing to share his code, I can probably get it working the way it should. I've never had any difficulty with these tags.

[jsbien]

I also never had problems with the tags.
His setup is quite complicated. I tried to reproduce the problem but gave up. He uses its own document class, which he provided to me, but uses also a lot of some fonts I'm not familiar with. At present his priority is to ship the book to the printer. Perhaps later he will be interested in investigating the problem.
BTW, recent Emacs has glyphless-display-mode which looks quite useful.

[jsbien]

FYI, I accepted a brute force work-arounds in the official paper https://doi.org/10.36824/2022-graf-bien and made a properly typeset postprint https://www.researchgate.net/publication/390326449_16th_century_Latin_printed_brevigraphs_in_Unicode---a_computer_resource.

[psb1558]

Both versions came out looking good!

[jsbien]

Thanks!
BTW, I tried to close the issue as outdated, but it doesn't work.

[psb1558]

I'll try it.

[psb1558]

Doesn't work for me either. I suppose it wants an answer marked as correct.