[NTG-context] beta

mathew meta at pobox.com
Sat Mar 26 23:00:27 CET 2011

On Sat, Mar 26, 2011 at 05:42, Hans Hagen <pragma at wxs.nl> wrote:
> On 26-3-2011 9:19, Taco Hoekwater wrote:
>> As I wrote in that message, the ghostscript dingbats will not be in
>> the actual unicode locations because the Type1 font does not carry
>> enough information for the reencoding to work.
> Indeed and I was somewhat puzzled by this topic coming up again.

There's a simple explanation: I didn't receive the previous on-list
replies. I just searched my spam folders and ConTeXt list folder, and
I still can't find them. Very mysterious.

> Now, as the gs dingbats seem to have the same names as the official ones
> (some weird numbering scheme) i added a reencoding vector (if the gs files
> had been different I'd not done that).

Taco writes:
>First, run this file:

That gives me an error:

! LuaTeX error <main ctx instance>:35: attempt to index field
'iterators' (a nil value)
stack traceback:
	<main ctx instance>:35: in function 'show_all'
	<main ctx instance>:1: in main chunk.

\ShowCompleteFont ... \ctxlua { fonts.show_all() }
                                                   \stopcolumns \page \egroup
l.146 ...ompleteFont{name:dejavusansmono}{10pt}{1}

I'd really like to know how to get a dump of all the symbols in a font
with the codes to address them by, as in general I don't think we can
expect Hans to build indices for every dingbat font out there :-)

I actually had no idea that the glyphs in font files had font-specific
names. I thought they were just numbered according to some standard
encoding, and that it was a matter of translating from glyph name or
Unicode number to whatever encoding was used in the font file.


More information about the ntg-context mailing list