Red's Parse dialect [1] (basically PEG on steroids, think OMeta [2]) has this kind of functionality build-in, but it's not limited to strings and can work with

Red's Parse dialect [1] (basically PEG on steroids, think OMeta [2]) has this kind of functionality build-in, but it's not limited to strings and can work with any series-like datatypes, including code itself. Consider: ; multi-line string source: { func main() { fmt.Println("hello world") } } parse source [thru "fmt.Println(" copy arguments to ")"] foo: does [ print rejoin [{fmt.Println(fmt.Sprintf("Red says %s", } arguments "))"] ] pattern: [set var skip '<= set rest skip] ; code (which is represented by heterogeneous data structure called "block") example: [ if (width <= 1280 && height <= 800) [ return 1 ] ] parse example rule: [ some [ change pattern (random rejoin [var rest]) | ahead paren! into rule | skip ] ] bar: does [probe example] Running foo and bar yields: >> foo bar fmt.Println(fmt.Sprintf("Red says %s", "hello world")) [ if ("21t8w0dhi" && "tig8e0h0h") [ return 1 ] ] Of course, it's not as specialized and dovetailed to string rewriting as Comby (i.e. it's more general), but the trick is that you can build Comby-like template engine with Parse itself; this idea of building tools with linguistic abstraction is tied into Red's (and Rebol's) design: data-rich homoiconicity and idea of dialects (aka embeded DSLs) in general. Curious to see if Comby can offer similar functionality across multitude of languages and data formats. [1]: https://www.red-lang.org/2013/11/041-introducing-parse.html https://www.red-lang.org/2013/11/041-introducing-parse.html [2]: https://en.wikipedia.org/wiki/OMeta https://en.wikipedia.org/wiki/OMeta