Implement a tool to facilitate the use of impure code in Haskell
The purpose of this tool is to transform all the functions of a program to
monadic form of the type: m(a -> m b), where a, b can also be functions of the
same type, creating nested functions like the following: (a -> m (b -> m1 c)).
We would like to have the ability to compose the effects of nested/different
monads, so we chose to use the extensible-effects library.
The main monad has been used is Eff from Control.Eff.
Eff r a is parameterized by the effect-list r and the monadic-result type
a similar to other monads.
It is the intention that all other monadic computations can be replaced by the
use of Eff.
The programs have been run and tested with the stack tool (version 1.9.3), GHC version 8.6.4
We implement an interface to check the compatibility of the original and transformated programs. There is a pure and impure version of the platform, that is related to the ability of using or not monads as arguments of a function. You can start by accessing the appropriate path (/puretransf or /tranf), and run the following commands:
stack build
stack run
You can load the input file, by running:
:l name
then the tool will automatically generate 3 files. The first,named nameOutput.hs
corresponds to the Original Haskell code, the second is an auxiliary file,
named nameMConvert.hs and the third is the transformated file, named
nameTransfOutput.hs. Afterwards, it will compile and run them, showing
the results in the screen.
*Notes:
name, has no file extension- The input file has the follow format:
a) Every function, has it's own type annotation that ends with a semicolon
;b) Every definition of a specific function ends with a semicolon, except of the last, which ends up with a newline character\nc) It contains aresultfunction (with a not function type) that will be passed to themain, ofnameOutput.hsandnameTransfOutput.hs
extensible-effects is based on the work of
Extensible Effects: An Alternative to Monad Transformers.