This project was a programming challenge issued to me by a buddy of mine, James Brock.
The overall goal was to create a sed-like string matching function that would use a parser function to scan through a string and select all parts of the string that match the parser as a 'Right' value, while returning non-matching strings as 'Left' values.
One of the goals was to use a parser function from a parser combinator library in place of a regular expression. He did not specify a parser combinator library, so I chose Attoparsec.
I use cabal new-build to build this project, and once built thusly,
you can test the sed-like function in a GHCi session by running
cabal new-repl. Here is an example session:
λ pureSed double "look at number 1.414"
[Left "look at number ",Right ("1.414",1.414)]
λ pureSed double "look at number 1.414 it is quite nice"
[Left "look at number ",Right ("1.414",1.414),Left " it is quite nice"]
λ pureSed double "look at number 1.414 it is quite nice 4e0"
[Left "look at number ",Right ("1.414",1.414),Left " it is quite nice ",Right ("4e0",4.0)]
λ pureSed double ".3e3 look at number 1.414 it is quite nice 4e0"
[Left ".",Right ("3e3",3000.0),Left " look at number ",Right ("1.414",1.414),Left " it is quite nice ",Right ("4e0",4.0)]
λ pureSed double "0.3e3 look at number 1.414 it is quite nice 4e0"
[Right ("0.3e3",300.0),Left " look at number ",Right ("1.414",1.414),Left " it is quite nice ",Right ("4e0 ",4.0)]
λ pureSed double "0.3e3 1.414 4e0"
[Right ("0.3e3",300.0),Left " ",Right ("1.414",1.414),Left " ",Right ("4e0",4.0)]
BE WARNED: Efficient string matching was NOT a goal of this challenge.