[XeTeX] fontspec 1.5

[Jonathan Kew]

On 2 Dec 2004, at 5:30 pm, Chris Menzel wrote:

> On Thu, Dec 02, 2004 at 08:42:24AM -0600, Herb Schulz wrote:
>> On 12/1/04 8:55 PM, "Will Robertson" <will at guerilla.net.au> wrote:
>>
>>> Okay.
>>>
>>> This must be the stuff I added to get the Euler math fonts working.
>>> I'll change that to a class option.
>>>
>>> I'll also have a look in the Lucida math packages and see if I can 
>>> get
>>> proper support working there too.
>>>
>>> Thanks for the help,
>>> Will
>>
>> Howdy,
>>
>> That would be wonderful. Thanks for the great job on getting XeLaTeX 
>> ``this
>> user'' friendly. I don't think I'll ever fully understand how La/TeX 
>> does
>> maths but it sure does it nicely.
>
> I *hope* I'm missing something here, but as far as I can tell there are
> still numerous Euler math characters not working with the latest 
> version
> of fontspec; notably:
>
> \{ -- renders as \}
> \} -- renders as |
> \land -- renders as \lor
> \lor -- renders as \{
> \aleph -- renders as \mathcal{A}
> \neg -- renders \Re
>
> And I also know that this is NOT a deficiency in Will's terrific
> fontspec work, but with Euler script, I was just hoping he might have
> found a workaround for the latest version.  I am unfortunately clueless
> as to how to fix Euler script; but I may take the time to learn over 
> the
> break -- I would really like to have Hoefler Text / Euler for the
> semi-mathy papers I write! :-)
>

The deficiency is actually with xetex/xdv2pdf's so-called "support" for 
legacy TeX fonts, which is pretty fragile. It tries several fairly 
messy tricks in an attempt to guess the glyph numbers needed to render 
the character codes coming from TeX, via the Quartz glyph-rendering 
APIs, but it's not as robust as I'd like it to be.

I believe it could be fixed if there were a specific encoding file 
associated with Euler in the psfonts.map file that associates tfm names 
with downloadable fonts; someone with sufficient motivation might want 
to give this a try.

JK