Some years ago I set out to build a lexer and parser for AsciiDoc to use it in IntelliJ for syntax highlighting and AST manipulations.
How hard can it be? Without knowing much about parsing, I extended an existing minimal JFlex parser, and gave it a try. This talk will revisit several problems I faced, and show when and how not to use JFlex.
After several iterations there is now a parser that works and can parse AsciiDoc including several quirks of the language. Still, when looking at the future, the AsciiDoc WG is looking for new ways to build a better suited parser in line with a new AsciiDoc language specification.
Those who'll accept the challenge are invited to the AsciiDoc WG: To help with guidance, to pick the right tools, and as mentors.