Hi Gerrit, Am 05.10.2017 um 13:11 schrieb Gerrit Imsieke:
If the TeX code within the terms contains curly braces, it’s probably time to use a real parser instead of regexes, maybe as part of a conversion tool such as latexml (http://dlmf.nist.gov/LaTeXML/) plus custom XSLT.
I have tried latexml and I have objections on two points: 1. the resulting xml is to complex to be easily read 2. not all packages are supported when I tried it. Since I am a science author and not a freal with computer languages the task to provide a package to be used with latexml by far is over my head. What I like about oxygen that I do not have to worry about computer languages, parsing and transforming, but I can work on my book with a minimal knowledge of xml. Within a month, the testing period, I converted some 300 pages of LaTeX to docbook 5 xml and can see at least the result as an epub which is quite nice. For the polishing and preparing the book for print and electronic publishing, I will turn to the professionals to do the xslt stuff. Maybe later when the book is almost ready I might become interested to try xslt myself. My 5cts Bernhard -- spitzhalde9 D-79853 lenzkirch bernhard.kleine@gmx.net www.b-kleine.com, www.urseetal.net - thunderbird mit enigmail GPG schlüssel: D5257409 fingerprint: 08 B7 F8 70 22 7A FC C1 15 49 CA A6 C7 6F A0 2E D5 25 74 09