[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

MathTime version of new encodings

While Matthias has taken care of the Lucida math layout (I hope), 
I have recently been working on the MathTime math layout of the 
new encodings.  Here's a quick summary of the status:

What can be implemented?

* MathTime only:

   MC:  Latin, italics Greek, and upright Greek capitals from MTMI,
        lowercase upright Greek missing, digits from Times,
        remaining symbols and delimiters from Times, MTMI and MTSY

   MSP: symbols from MTSY and MTMI, Script from ZapfChancery,
        several symbols unavailable or faked (rather badly)

   MXP: everything from MTEX, for the scriptsize and scriptscriptsize 
	radicals I need to hardwire the 10/7.6/6 size scaling

   no bold versions avaialbe

   tentative font names: xmta, xmtb, xmtc

* MathTime + MathTime Plus:

   MC:  as above combined with lowercase upright Greek from MTGU,
        very near complete except for inverted and reflected symbols.
        MTSY replaced by MTSYN which provides a few extra symbols 
        which are otherwise taken from Times.

   MSP: as above combined with Script alphabet from MTMS (Math Script)

   MXP: exactly as above

   everything available in normal and bold, some also in heavy

   tentative font names: xmpa, xmpb, xmpc (normal),
   xmpab, xmpbb, xmpcb (bold), xmpah, xmpbh, xmpch (heavy)


What's missing or faked?

  MC:  missing: 
	- lgroup/rgroup and lmoustache/rmoustache (I'm unconvinced 
	  if it was a good idea to include slots for them anyhow)

	- Hebrew beth, gimel, daleth  (aleph is in MTSY)

	- hslash, lambdaslash  (hbar, lambdabar can be faked)

	- inverted Omega, reflected epsilon, reflected ampersand,
 	  reflected prime, digamma (all unfakable) 

	 - upright partial  (fill the slot with italic partial?)

       faked: 

	- triple vertical bar
	- hbar and lambdabar (will need some fine tuning)

  MSP: missing: 

	- square subset/supserset, left/right triangle/triangleequal,
	- squiggly right arrow, square box and big diamond
	  (all needed for LASY compatibility)

	- leq/geqslant, precedes/followsorcurly (I've seen journals
	  using the AMS leq/geqslant combined with MathTime)

	- triple muchless/greater (<<<, >>>), reflected sim and simeq,

       faked: 

	- double and triple smallint
	- all the negated relations (will need some fine tuning)

	- doteq, bowtie (both defined as macros in plain.tex)
	- divideontimes (faked as the name suggests)
	- simeq, congruent, approxequal, lesssim, greatersim

  MXP: faked:
	- double and triple integral, triple extensible bar
	- middle and extension pieces for horizontal braces

       missing: 
	- double contour integral, extra sizes of wide accents

While MC and MXP are almost complete with only few exceptions, MSP
remains incomplete as far as the LASY and AMS symbols are concerned
and there is little one can do about it short of borrowing symbols
from other font sets or faking some of them.

About the implementation:

I'll send my patches privately to Matthias so that he can prepare
a new release for distribution.  As for the fontinst installation,
I've chosen to install from PL + ETX rather than AFM since that
allow greater flexiblity to resolve glyph naming conflicts.
I've also chosen to install from MTMI rather than RMTMI since 
that's the only way to preserve the kerning information hidden
in the MTMI virtual font.

So long,
Cheers, Ulrik.

P.S. Matthias, what's the status of the Lucida math layout?