If your font had a CMAP table for your proprietary encoding, the 'Unicode' fields for the glyphs will contain wrong values.
Make sure that each Unicode code point from the Devanagari range is assigned the correct glyph from your font. Assign the right Unicode values to glyphs (in VOLT) Now, you can load your font into VOLT and proceed with conversion. It is also a good time to revisit your choice of conjuncts and alternate letterforms. The gain is simplicity of your VOLT tables and the help you get from Uniscribe that controls application of those features.
This step will, again, most probably require you to produce quite a few of new composite glyphs.
If, for example, your encoding assumed that DEVANAGARI LETTER O is coded as DEVANAGARI LETTER A plus DEVANAGARI VOWEL SIGN O, you would probably need to create a new glyph for DEVANAGARI LETTER O.Īs a result of this step, you may be creating several new glyphs that are, in fact, composites of already existing shapes (so you don't really need to design new shapes for them). There should be at least one glyph for each code point in the Devanagari range. Glyph setĮxamine your font and make sure it covers the Unicode range for Devanagari. They only reflect one person's limited experience. Different approaches are of course possible so please take my writings with a grain of salt. For more details one would refer to the Unicode Standard, the OpenType specification and the VOLT release notes. What does it take to convert it to Unicode? Where do I start? So, suppose you have a font that was used to print Devanagari in some proprietary encoding. We would like to see many more Unicode Devanagari fonts appear, and hope that appropriate standards will emerge along the way. Whichever way you look at it, having only Mangal is hardly sufficient for all text processing purposes and for all the languages that use the script.