when first i saw this post(o) i fleshed out a concept that did something similar the 'engine' in this case was a transpiler that took my scripting language and

when first i saw this post(o) i fleshed out a concept that did something similar the 'engine' in this case was a transpiler that took my scripting language and wrote, and compiled if necessary, code for complex algorithms with simple parts simultaneously into whatever languages i implemented it would do a lookup of an on disk copy of this document, ideally normalised but i only implemented two languages so the data was already agreeably normal, to determine how the output code should be implemented i only had it writing in python and c, doing: if statements, for and while loops, unoptimised typing, basic operators; before i realised i was having too much fun on altering the engine to tweak the resultant code than doing the algo work that i set out to do but each time i think 'i should look into such and such language', i muse on digging up that old commit as a fun aside, workaround to languages' different approaches to memory each of the output programs utilised an array as a faux stack that could handle object implements with an accompanying algo that could take the stack and read from or write to valid, though a subset of, json (o) http://rigaux.org/language-study/syntax-across-languages/ http://rigaux.org/language-study/syntax-across-languages/