Use a lexer/parser lib like ANTLR

[Khaos66]

Instead of spliting all strings by yourself you should use a lexer like ANTLR.
This has many benefits:

• There might already be a markdown language definition for ANTLR on the internet.
• You will get a stream of tokens, that can be parsed way more easy
• Parsing might be even faster (ANTLR4 is highly optimized)
• For each token you know the exact index of the original string, that it originates of. Helpfull for New Plugin: CheckBox #50

[Khaos66]

https://github.com/mar9000/antmark is an ANTLR markdown grammar from 8 years ago...
There is a request for a standard grammar on the ANTLR repo antlr/grammars-v4#472
Also they mention this article http://roopc.net/posts/2014/markdown-cfg/ which nullifies all of this =/

[Rast1234]

if anyone is interested, there is a more or less standartized spec now and and it comes with a test suite https://commonmark.org/