unicode maps README
		===================

Each unicode_XXX.map file contains a list of unicode code points that might
plausibly be used from inside gnuplot.  These are only relevant to the
PostScript terminal drivers, and only if you want to use UTF-8 encoding.

If you have a TTF unicode font containing the relevant characters, you can
use the unicode map file to instruct the external program ttf2pt1 to prepare
an Adobe Type 1 font.

For example:

  ttf2pt1 -p ft -e -L unicode_big.map arialuni.ttf

This will extract character glyphs for these particular code points
from arialuni.ttf and create a new file arialuni.ps.  You can use this new
*.pfa font file from gnuplot by saying

  set encoding utf8
  set term post fontfile 'arialuni.ps' font 'ArialUnicodeMS'

Note that gnuplot can run ttf2pt1 for you on the fly if you
first define the appropriate command in the environmental variable
GNUPLOT_TTFTOPFA.   For example:

  setenv GNUPLOT_TTFTOPFA "ttf2pt1 -e -p ft -L unimap.map %s -"

In this case you could say

  set encoding utf8
  set term post fontfile 'arialuni.ttf' font 'ArialUnicodeMS'

In either case, the converted glyphs will be embedded in your
gnuplot output file.  There are several caveats:

1) This can make the file quite large.

2) If a glyph isn't in the starting *.ttf font to begin with, nothing
   you do here will fix that.

3) If gnuplot runs the conversion for you, it will be triggered each
   time you plot.

Several unicode map files are provided.  They differ in how many
characters they request, and hence in the size of the resulting font file.

  unicode_small.map  (462 characters)
  ===================================

- the Latin1 character set
- the basic Greek alphabet
- basic math and physical symbols
- 133 less obscure math symbols from Unicode pages 34 and 35


  unicode_big.map  (863 characters)
  =================================

- the Latin1 character set 
- extended Latin character sets needed to support other European languages
- the Cyrillic alphabet, with extensions for various slavic languages
  (may not be complete - please report omissions)
- the basic Greek and Hebrew alphabets
- basic math symbols
- 176 mathematical symbols, some obscure, from Unicode pages 34 and 35
- physical and astronomical symbols

The unicode_big map makes a total of about 860 glyphs available to use
from gnuplot, but only if they exist in the original TTF font file.
The maximum number that ttf2pt1 can process is 1024.
This is only a tiny fraction of the total characters indexed by Unicode.
But if you need more, probably you would do better to use a different
gnuplot terminal that supports direct use of UTF-8 TTF fonts (pdf, png, svg).