keentype

> modern languages like Markdown Markdown was created in 2004. From the creator: > ... the single biggest source of inspiration for Markdown’s syntax is the format of plain text email. Email goes back to 1965, though I suspect Markdown's influence stems from the more widely adopted email usage of the 1990s. I'm not sure if I'd call Markdown "modern". Quite popular though! > Part of LaTeX's success was the absolutely beautiful documents it can make with nothing but a personal computer. I'd say that was TeX's success, with LaTeX bolted on later to greatly improve TeX's extensibility. Keep in mind, there are a number of TeX-centric implementations beyond LaTeX. For example, my fork of NTS, called KeenType, is a pure Java version of TeX that can typeset beautifully and has at its core Knuth's original TeX files. I specifically avoided integra

My port of the New Typesetting System (NTS), called KeenType, whittles down the Java-based implementation to a few of Knuth's files. Namely, "plain.tex" and "hyphen.tex": https://github.com/DaveJarvis/KeenType/tree/main/tex/src/mai... Getting familiar with the fonts required understanding the difference between font metrics (TFM files) and the fonts themselves. To make matters a little less straightforward, Knuth created a special character mapping for indexes into the fonts. It was not easy to find a font that mapped those glyphs exactly. The closest font was BaKoMa: https://github.com/DaveJarvis/KeenType/tree/main/tex/src/mai... This required hard-coding a mapping between Knuth's code points and the actual code points in the target font: https://github.com/DaveJarvis/KeenType/blob/989dbe26f68eda75... Plain TeX has a lot of moving parts,

A while ago I optimized KeenType, which is a Java-based fork of NTS that can convert TeX commands into SVG images: https://github.com/DaveJarvis/keentype The following tutorial shows the real-time rendering speed of KeenType within my text editor, KeenWrite: https://youtu.be/vgyYXwwF_lc?list=PLB-WIt1cZYLm1MMx2FBG9KWzP... KeenWrite can produce web pages from Markdown with the TeX commands included as either as plain TeX macros (usable by JavaScript TeX renderers) or as pre-rendered SVG elements (no JavaScript required).

When trying to make my editor, KeenWrite[0], render TeX in real time using JMathTeX[1], I used initially used JFreeSVG[2] to generate SVG images (not rasterize them). After running the code through async-profiler[3], it turned out that JFreeSVG was using Java's NumberFormat to convert double values to strings. Comparative benchmarks when generating 1,000,000 SVGs from TeX expressions, 1 run each, no warm-up: * NumberFormat: 778060 milliseconds (~13 minutes) * RyuDouble: 421160 milliseconds (~7 minutes) Eventually I replaced JMathTeX with my NTS fork[4] and replaced the JFreeSVG library with a custom class to create SVG document strings[5]. Going from non-real-time TeX to real-time TeX rendering was worth it. [0]: https://github.com/DaveJarvis/keenwrite [1]: https://jmathtex.sourceforge.net/ [2]: https://github.com/jfree/jfreesvg/ [3]: htt

In Java, many number-to-string implementations use NumberFormat. This is abysmally slow if the problem domain doesn't require internationalization, which is the case for machine-readable file formats, such as SVG. When I performance tested JMathTeX for rendering TeX, one major bottleneck for converting TeX into SVG elements was JFreeSVG's use of NumberFormat. Replacing NumberFormat with RyuDouble doubled the throughput[0]. (Reusing a StringBuilder to concatenate strings yielded another doubling.) For KeenType[1], a fork of the New Typesetting System (and more complete plain TeX implementation than JMathTeX), I added an SVG generator that converts floating point numbers to strings using a StackOverflow answer[2], not the Ryu algorithm[3]. Better performance, simpler algorithm. Knuth's advice holds: measure then optimize. [0]: https://githu