Font Conversion

During the conversion of RTF documents to TeX, we translate a non-Latin text to encodings that are legible for TeX. Our conversion programs are designed in such a way that they can standardize the submitted non-Latin fonts before converting the file to TeX. The result is a new RTF file in which the original fonts have been replaced by a new one, most likely a Unicode font. Apart from conversion of fonts in RTF documents, we also offer the conversion of fonts themselves. We can convert fonts in conventional formats (like PostScript and OpenType) to TeX fonts that can completely be adapted to your wishes.

Problems with non-Latin fonts

One of the complications working with fonts of non-Latin alphabets is their lack of standardized font tables. A conventional font can be seen as a table with 256 boxes, each containing a symbol, a so-called glyph. For Latin fonts, the division of the boxes is registered in the international standard, like ASCII and ISO Latin 1; the glyph ‘A’ is always positioned in box 65 en the ‘a’ in box 97. Due to a long lasting lack of standards, certain fonts have varying encodings. For example, the glyph for the character ‘chi’ has a different position in the Greek Athenian font (box 120) than in Greek Helena (box 99). The consequence is that a Word-user canĂ•t simply convert Greek Athenian to Greek Helena; the character ‘chi’ will change into the character that occupies box 120 in Helena, namely the character ‘ksi’. Such a conversion can only be executed by a program that knows the different positions of the glyphs of both fonts.


Unicode is developed to solve the aforementioned uncontrolled growth of non-Latin encodings. Unicode is the international initiative to standardize all characters of the entire language in one gigantic table (with 65.536 boxes), that doesn’t just include letters, but also punctuation marks and numerous technical symbols. The Unicode standard is implemented in many platforms and applications. Together with the development of OpenType, a font format with Unicode-functionality was introduced.

Conversion to TeX

The conversion programs that we developed for the conversion of documents containing non-Latin fonts to TeX, use an extensive library of conversion tables. These tables contain information about the position of glyphs in many common non-Latin fonts. When we encounter a font that is not yet represented in our library, we compose a table for this font and incorporated it. As a result, our library has expanded into a versatile tool for font conversion during the past few years. We treat submitted texts in Unicode font the same way as texts in non-Unicode fonts, since all the relevant information (like Greek, Hebrew and Arabic Code Charts) is already present in our library.

Conversion to RTF with Unicode

We can also convert fonts to Unicode and afterwards save the result in an RTF file. Imagine you are editing a volume with articles in several different types of Greek. If your publisher desires a uniform submission, we could convert all the Greek fonts to Unicode. The result of this conversion is an RTF file in which all the non-Latin fonts will be reproduced in OpenType font.

TeX fonts

Flexibility is the most important feature of the TeX system. This especially holds for the font management of this system. TeX doesn't work with conventional fonts but with special TeX fonts which have to be created by the user from conventional fonts. This process enables the user to apply certain special features to the font, for instance in the area of ligatures. The possibilities are so broad that font management is the hardest element of the already steep learning curve of TeX. Because of our extensive experience with complex scientific typesetting we were able to master TeX-font management. We can help TeX users by making PostScript- and OpenType fonts available for their TeX system. We also have extensive experience with generating TeX font metrics (TFM files) and the construction of virtual fonts. Virtual fonts are fonts that have been composed from other existing fonts; this technique can be used to add old-style numbers or ligatures from an expert font to a regular font. We are also experienced in the encodings for non-Latin fonts, for left-to-right scripts as well as right-to-left scripts.