Exploring metaprogramming using Ruby language
Download as PPTX, PDF2 likes808 views
The document provides an introduction to metaprogramming using the Ruby programming language, explaining its concepts, including reflection and generative programming. It discusses Ruby's dynamic and reflective nature, highlighting how it allows programs to treat code as data for increased flexibility and productivity. Additionally, it addresses potential challenges of metaprogramming, such as increased complexity and the possibility of generating ‘metabugs’.
Exploring metaprogramming using Ruby language
- 1. Exploring metaprogramming using Ruby language A brief introduction Harshal Ganpatrao Hayatnagarkar PhD Candidate in Dept. of Computer Science 1 Friday, October 3, 2014
- 2. Engineering marvels Ford Model T (approx. 1918) Tesla Model S (2010) 10/24/2014 2
- 3. Assembly Lines: Process marvels Ford Model T Assembly Line Tesla Robotic Assembly Line 10/24/2014 3
- 4. Assembly Lines: Process marvels People building cars A worker on assembly line does same task repetitively, item after item and day after day. People building robots building cars Is this like programming? People building robots build robots building cars What is this analogous to? Courtesy: http://stackoverflow.com/questions/3468246/whats-the-use-of-metaprogramming 10/24/2014 4
- 5. Metaphor … People building robots building robots… building cars What is this analogous to? 10/24/2014 5
- 6. Meta-programming • Meta (‘beyond’ in Greek) + programming Programming Abstraction – Similar to meta-data, meta-model, meta-physics and so on. • “Writing of computer programs that with ability to treat programs as their data, as it meant that a program could be designed to read, generate, analyse and/or transform other programs, and even modify itself while running. In some cases, this allows programmers to minimize the number of lines of code to express a solution (hence reducing development time), or it gives programs greater flexibility to efficiently handle new situations without recompilation.” - Wikipedia 10/24/2014 6
- 7. Revelation Data-Code Duality 10/24/2014 7
- 8. Two aspects Reflection Generative programming 10/24/2014 8
- 9. Reflection • Ability of a computer program to examine and modify the structure and behavior (specifically values, meta-data, properties and functions) of the program at runtime. – Metalanguage in which metaprogram is written. – object language, programs written in which are manipulated. – reflection or reflexivity is ability of a programming language to be its own metalanguage. • Fairly common in programming languages – C++ (limited) – Java (+ other JVM languages), C# (+ other CLR languages) – JavaScript, Python, PHP, Ruby 10/24/2014 9
- 10. Generative Programming • “A style of computer programming that uses automated source code creation through generic frames, classes, prototypes, templates, aspects, and code generators to improve programmer productivity.” - Wikipedia • For example, – Internal to language: C/C++ preprocessor macros, C++ templates, LISP macros, Ruby/Python/PHP evaluations. – External: Code generation templates. 10/24/2014 10
- 11. Ruby Programming Language • Dynamic, reflective, object-oriented, general-purpose programming language. • Influenced by Perl, Smalltalk, Eiffel, Ada, and Lisp. • Supports multiple paradigms functional, object-oriented, and imperative. • With dynamic type system and automatic memory management. Yukihiro "Matz" Matsumoto 10/24/2014 11
- 12. Meta Object Protocol • Everything is an object, including classes. • Method calling is analogous to message passing where arguments are payload. 10/24/2014 12
- 13. 10/24/2014 13
- 14. Open Classes 10/24/2014 14
- 15. Dynamic methods 10/24/2014 15
- 16. Dynamic methods Ruby on Rails – ActiveRecord example 10/24/2014 16
- 17. Generative Programming • Code generation – one or more levels – ‘eval’ expression family – Takes source code in string form and embed in current context. 10/24/2014 17
- 18. Two-level generation • Code generating code generating code Level 1 Level 2 ‘Level 1’ eval ‘Level 2’ eval 10/24/2014 18
- 19. Pitfalls • Metabugs? – Bugs generating bugs generating… – And there is no ‘programs debugging programs debugging programs…’. • Increased code complexity 10/24/2014 19
- 20. Further exploration… • Model-driven development • Domain-specific languages • Ruby on Rails for web development framework – An example of metaprogramming features. 10/24/2014 20
- 21. Questions? And thank you for your interest ! 10/24/2014 21
Editor's Notes
- #6: This metaphor can explain metaphor in two ways: Robots building robots… Self-aware robots (autonomic computing)
- #9: 1. Reflection: Recalling the example of self-aware robots, if a robot can inspect itself for faults and optimization, it is wonderful. Such a capability requires many features in the robotic system, which could be provided by reflection.
- #10: In many languages such as C/C++, once code is generated, the information about types and variables is lost after compilation (only exception is RTTI where name of type is retained at runtime). However, the