A Parser inspired by llhttp for handling speedy reading and parsing of json datatypes this parser borrows the same functionallity as llhttp so that callbacks can be used to filter and walk through very specific data rather than loads of unwanted data.
-
Developers tend to filter loads of json responses for just one object or array.
-
Loading Large sums of data into memory is a waste of time. When we get into gigabyte or terribyte territories this can become a serious problem with average tools.
-
Data Conversions should be done immediatley and by yourself and not loaded into memory first thing.
-
You can scrape unlimited supplies of Json
- without loosing memory
- requiring too much memory
-
You can set a recursion limit to control how many containers the parser remebers.
- The Parser only requires one Malloc Call and can be easily done with others like Python's CAPI (PyMem_Malloc) or mimalloc for instance...
-
You can skip or ignore datatypes you do not wish to use...
-
You can pause the parser on any callback allowing languages like Python to create Yeild Generators for certain types of data being returned.
-
This library should be compatiable with llhttp meaning there should be no compiling errors between these libraries if they are being intermixed.
-
RFC 8259 should be strictly obyed.
-
You can build off of lljson to create a more advanced library for other languages or other algorythms in general.
-
Modern Json Libraries can be slow because they have to allocate unwanted memory before you can move through all of it.
Currently I am finishing up the first version and need to add some native C Api stuff. You can compile the current C Code using typescript which I programmed with in the hopes that contribution will come around at somepoint.