On Jun 12, 2016, at 7:28 PM, Hans Hagen <pragma@wxs.nl> wrote:

this is why long ago (so maybe forgotten) \pdfincludechars was introduced (because in widgets one also needs to have glyphs present for editing)

Never known, so not ‘forgotten’.

Thanks for the tip...

(i remember using the 'put wanted glyph typeset white one meter in the margin outside mediabox' trick but that could confuse viewers)

… however, it doesn’t work with the test-bera example.

viz.

> \T1/fve/m/n/17.28 .

<recently read> \font

l.31 \showthe\font

./test-berafont.tex:32: pdfTeX error (font): command cannot be used with virtua

l font.

?

l.32 \pdfincludechars\font

{\'{ }a}

./test-berafont.tex:32: ==> Fatal error occurred, no output PDF file produced!

Transcript written on test-berafont.log.

The same thing happens with the hebart example.

> \HE8/cmr/m/n/12 .

<recently read> \font

l.39 \showthe\font

?

./hebart.tex:40: pdfTeX error (font): command cannot be used with virtual font.

l.40 \pdfincludechars\font

{\char32'\char39}

./hebart.tex:40: ==> Fatal error occurred, no output PDF file produced!

Besides, how is one supposed to know which extra glyphs need to be added?

What I don’t understand is that if the glyphs are needed to make the font subset usable,

then this means adding the coding for glyphs which are called by other characters.

So the names of these (sub-glyphs, if you like) should be able to be added to the /CharSet list.

(I’m saying this without having examined the actual coding.)

Is there really a problem with coding this?

Hans

-----------------------------------------------------------------
                                         Hans Hagen | PRAGMA ADE
             Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
     tel: 038 477 53 69 | www.pragma-ade.com | www.pragma-pod.nl
-----------------------------------------------------------------

Cheers,