The syntax of \latinfamily

[Rebecca and Rowland]

>[snip]
>>> From the discussions in tex-fonts the last six months, I
>>>have learnt that the main source of a euro symbol (which people are likely
>>>to want) is the Adobe eurofonts, which are not connected to specific font
>>>families, but which still exist in a few different designs, so that one
>>>cannot simply use one eurofont with all font families (and expect the euro
>>>symbol to look right).
>>
>>Erm...  actually, you can (sort of); just to prove I'm not only a whinging
>>Pom, you might like to look at this package on CTAN:
>>
>>macros/latex/contrib/supported/eurofont/
>>
>
>I yield to your superior knowledge of what might or might not be
>aestetically acceptable in this matter.

The aesthetics apply more to the source code than the output...  It's like
this: the usual reason one might want to use a single euro symbol for all
founts is for ease of use.  You can define a single command, \euro, that
you can use in all documents to get a euro symbol.  In the normal way of
things, this command would select a euro symbol from a single fount family.
The eurofont package uses a big switch statement to do things a bit
differently: you use the \euro command, and get an appropriate euro symbol
for the particular fount family you're using at that point.  The way it
works is this: if the current fount family is in *that* list, you get
*this* euro symbol, and so on for several different lists.  There are
various tweaks and bobbles which means it's really very flexible, but it's
not really a sensible way of selecting glyphs in general.

I mentioned this package because there's a good chance other people on this
list might well want to use euro symbols, and it's probably useful to know
that you don't have to play around with fontinst to get a decent output

> My intention was only to give an
>example of a case where a user would want to supply \latinfamily with more
>information than what is currently possible.

Indeed, and it's a pretty good example.  It's just that there's always
another way...

>Actually, this is only a special case of a more general situation: The user
>specifies that glyphs (almost always symbols) that could not found in any
>of the fonts in the font family should be taken from font #1, if they can
>be found there. #1 could for example be psyr8r.

Isn't this what the NFSS is for?

Rowland.