Functional Programming

Transcript

1. Functional Programming Ionut G. Stan - OpenAgile 2010
2. Functional Programming what why how
3. Functional Programming what why how
4. What is FP a programming style
5. What is FP a programming style conceptually derived from lambda calculus (1930s)
6. What is FP a programming style conceptually derived from lambda calculus (1930s) not procedural programming
7. What is FP a programming style conceptually derived from lambda calculus (1930s) not procedural programming functions as in mathematical functions
8. What is FP a programming style conceptually derived from lambda calculus (1930s) not procedural programming functions as in mathematical functions math function: input completely determines the output
9. Imperative Programming programming: telling a computer what to do
10. Imperative Programming programming: telling a computer what to do imperative languages are leaky abstractions
11. Imperative Programming programming: telling a computer what to do imperative languages are leaky abstractions also called von Neumann languages
12. Imperative Programming programming: telling a computer what to do imperative languages are leaky abstractions also called von Neumann languages von Neumann architecture is about modifying the state of the computer
13. von Neumann architecture
14. Imperative Programming programming: telling a computer what to do imperative languages are leaky abstractions also called von Neumann languages von Neumann architecture is about modifying the state of the computer computation model in imperative languages reflects von Neumann architecture
15. Imperative Programming programming: telling a computer what to do imperative languages are leaky abstractions also called von Neumann languages von Neumann architecture is about modifying the state of the computer computation model in imperative languages reflects von Neumann architecture imperative programming: what and how to do
16. What is FP a programming style conceptually derived from lambda calculus (1930s) not procedural programming functions as in mathematical functions trying to plug the abstraction leak
17. What is FP a programming style conceptually derived from lambda calculus (1930s) not procedural programming functions as in mathematical functions trying to plug the abstraction leak tell the computer what to do, not how
18. What is FP a programming style conceptually derived from lambda calculus (1930s) not procedural programming functions as in mathematical functions trying to plug the abstraction leak tell the computer what to do, not how mutations not allowed (no variables, just identifiers)
19. What is FP a programming style conceptually derived from lambda calculus (1930s) not procedural programming functions as in mathematical functions trying to plug the abstraction leak tell the computer what to do, not how mutations not allowed (no variables, just identifiers) no statements, just expressions (if/then/else is expression)
20. What is FP a programming style conceptually derived from lambda calculus (1930s) not procedural programming functions as in mathematical functions trying to plug the abstraction leak tell the computer what to do, not how mutations not allowed (no variables, just identifiers) no statements, just expressions (if/then/else is expression) functions are deterministic and side-effect free
21. What is FP a programming style conceptually derived from lambda calculus (1930s) not procedural programming functions as in mathematical functions trying to plug the abstraction leak tell the computer what to do, not how mutations not allowed (no variables, just identifiers) no statements, just expressions (if/then/else is expression) functions are deterministic and side-effect free functions are all we need to model computation
22. What is FP a programming style conceptually derived from lambda calculus (1930s) not procedural programming functions as in mathematical functions trying to plug the abstraction leak tell the computer what to do, not how mutations not allowed (no variables, just identifiers) no statements, just expressions (if/then/else is expression) functions are deterministic and side-effect free functions are all we need to model computation execution order is not guaranteed
23. Functional Programming what why how
24. Why FP easier to reason about programs
25. Why FP easier to reason about programs heisenbugs
26. Why FP easier to reason about programs heisenbugs race conditions
27. Why FP easier to reason about programs heisenbugs race conditions off by one errors
28. Why FP easier to reason about programs heisenbugs race conditions off by one errors objects trashing another object's internal state
29. Why FP easier to reason about programs
30. Why FP easier to reason about programs easier to parallelize
31. Why FP easier to reason about programs easier to parallelize program correctness proving
32. Why FP easier to reason about programs easier to parallelize program correctness proving composability results in greater and easier reuse
33. Just for fun - reuse in OO languages "You wanted a banana but what you got was a gorilla holding the banana and the entire jungle." Joe Armstrong, creator of Erlang
34. Why FP easier to reason about programs easier to parallelize program correctness proving composability results in greater and easier reuse try out new perspectives
35. Functional Programming what why how
36. Functional Programming what why how (even in imperative languages)
37. How to FP avoid side-effects/mutation as much as possible
38. How to FP avoid side-effects/mutation as much as possible at least keep them as private as possible in OO
39. How to FP avoid side-effects/mutation as much as possible at least keep them as private as possible in OO treat variables as immutable (constants/final)
40. How to FP avoid side-effects/mutation as much as possible at least keep them as private as possible in OO treat variables as immutable (constants/final) return values (output) based on params (input) only
41. How to FP avoid side-effects/mutation as much as possible at least keep them as private as possible in OO treat variables as immutable (constants/final) return values (output) based on params (input) only play with a functional language
42. Functional Programming what why how example
43. Example
44. Example
45. Example
46. "Conventional programming languages are growing ever more enormous, but not stronger. Inherent defects at the most basic level cause them to be both fat and weak: their primitive word- at-a-time style of programming inherited from their common ancestor -- the von Neumann computer, their close coupling of semantics to state transitions, their division of programming into a world of expressions and a world of statements, their inability to effectively use powerful combining forms for building new programs from existing ones, and their lack of useful mathematical properties for reasoning about programs." John Backus, known for Fortran, Algol and BNF
47. Thank You igstan.ro | ionut.g.stan@gmail.com | @igstan
48. Questions? igstan.ro | ionut.g.stan@gmail.com | @igstan