A simple preprocessor with the following rules:
- It processes files line-by-line If it finds an instruction in a line, the whole line is treated as an instruction and its contents outside of it is being discarded. This is to prevent going as deep as creating/using real parsers for real languages. So if it's a real programming language, you'll have to use single-line comment for that to prevent syntax checkers going crazy. But that was the intention anyway.
- It has 4 available instructions:
- #ifeq key value — accept contents in ifeq block only if passed value matches such in ifeq arguments
- #ifneq key value — accept contents in ifeq block only if passed value doesn't match such in ifeq arguments
- #else — inverts ifeq/ifneq block from the next line until the end of it
- #endif — closes the block
- For ifeq/ifneq it uses values passed in command-line arguments like this:
$ cirpp key1=value1 key2=value2
- All keys and values are strings consisting only of latin alphabet characters, numbers and underscore
// #ifeq GNOME_VERSION 45
... do Gnome 45 things
// #else
... do pre-45 Gnome things
// #endif
# preprocess source for Gnome 45
$ cat extension.js | cirpp GNOME_VERSION=45 > build/extension.js
# preprocess source for Gnome 44
$ cat extension.js | cirpp > build/extension.jsMIT