SlideShare a Scribd company logo

Clojure: Simple By Design

•
•
All Things Open
All Things Open

All Things Open 2014 - Day 2 Thursday, October 23rd, 2014 Stuart Halloway Founder/President of Cognitect Back Dev Clojure: Simple By Design

1 of 85
Download to read offline
Simple by Design: Clojure @stuarthalloway stu@cognitect.com Copyright Cognitect, Inc. This presentation is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License. See http://creativecommons.org/licenses/by-nc-sa/3.0/us/
Thoughtworks Radar Oct 2012: Adopt Clojure http://www.thoughtworks.com/radar
Redmonk Top 20 Clojure http://redmonk.com/sogrady/2014/06/13/language-rankings-6-14/
Design ! specification of an artifact using components to meet goals subject to constraints
Simple ! not compound
Valuable But Not Simple convenience beginner-friendliness ease familiarity smaller size lesser count
Simple ! not compound
Agenda examples of simplicity in Clojure benefits of simplicity in Clojure producing Clojure
Examples of Simplicity syntax protocols values and references
Examples of Simplicity syntax protocols values and references
"Hello World"
"Hello World" that is the program
Everything is Data
type examples string "foo" character f integer 42, 42N floating point 3.14, 3.14M boolean true nil nil symbol foo, + keyword :foo, ::foo
type properties examples list sequential (1 2 3) vector sequential and random access [1 2 3] map associative {:a 100! :b 90} set membership #{:a :b}
Function Call semantics: fn call args (+ 2 2) structure: symbol longs list
Function Definition define a fn fn name docstring (defn greet! "Returns a friendly greeting"! [your-name]! (str "Hello, " your-name)) arguments fn body
…Still Just Data symbol symbol string (defn greet! "Returns a friendly greeting"! [your-name]! (str "Hello, " your-name)) vector list
Complexities Avoided lots of syntax to learn ordering dependencies operator precedence code over data bias tedious metaprogramming
Examples of Simplicity syntax protocols values and references
Reversible Jacket Belt String
Reversible interface implementation Jacket Belt String
Reversible interface implementation extension Jacket Belt String
(defprotocol Reversible! (reverse [_]))! ! ! ! (defrecord ReversibleTie [a b])! ! ! ! (extend-protocol Reversible! ReversibleTie! (reverse [tie] (->ReversibleTie (:b tie) (:a tie)))! String! (reverse [s] (-> s! StringBuilder.! .reverse! .toString))) interface implementation extension
Complexities Avoided adapter pattern wrapper pattern translator pattern monkey patching StringUtils note that these are all combinatorial
Examples of Simplicity syntax protocols values and references
Me 182
Me 182 nachos
Me 182 nachos 184
Watch OO Flounder Me 182 nachos 184 instance? field? method? ??
If you have more things than names, your design is broken.
Clojure’s Simplicity Me 182 nachos 184 reference value pure function succession function new value
Values and References (defprotocol Nachos! (yum [_] "eat some nachos"))! ! (defrecord Person [name lbs]! Nachos! (yum [person]! (update-in person [:lbs] + 2)))! ! (def me (atom (->Person "Stu" 182)))! ! (def me-before @me)! ! (swap! me yum)! ! (def me-after @me)
Values and References (defprotocol Nachos! (yum [_] "eat some nachos"))! ! (defrecord Person [name lbs]! Nachos! (yum [person]! functional (update-in person [:lbs] + 2)))! ! (def me (atom (->Person "Stu" 182)))! ! (def me-before @me)! ! (swap! me yum)! ! (def me-after @me)
Values and References (defprotocol Nachos! (yum [_] "eat some nachos"))! ! (defrecord Person [name lbs]! Nachos! (yum [person]! (update-in person [:lbs] + 2)))! ! (def me (atom (->Person "Stu" 182)))! ! (def me-before @me)! ! (swap! me yum)! ! (def me-after @me) update semantics
Values and References (defprotocol Nachos! (yum [_] "eat some nachos"))! ! (defrecord Person [name lbs]! Nachos! (yum [person]! (update-in person [:lbs] + 2)))! ! (def me (atom (->Person "Stu" 182)))! ! (def me-before @me)! ! (swap! me yum)! ! (def me-after @me) multiple point-in-time values
Complexities Avoided incidental complexity temporal reasoning single-threading locking defensive copying setter methods String vs. StringBuilder vs. StringBuffer note (again!) that these are all combinatorial
Benefits of Clojure concision generality robustness agility
Benefits of Clojure concision generality robustness agility
StringUtils indexOfAny https://commons.apache.org/proper/commons-lang/javadocs/api-2.6/org/apache/commons/lang/StringUtils.html
indexOfAny Spec StringUtils.indexOfAny(null, *) = -1! StringUtils.indexOfAny("", *) = -1! StringUtils.indexOfAny(*, null) = -1! StringUtils.indexOfAny(*, []) = -1! StringUtils.indexOfAny("zzabyycdxx",['z','a']) = 0! StringUtils.indexOfAny("zzabyycdxx",['b','y']) = 3! StringUtils.indexOfAny("aba", ['z']) = -1
// From Apache Commons Lang, http://commons.apache.org/lang/! public static int indexOfAny(String str, char[] searchChars) {! if (isEmpty(str) || ArrayUtils.isEmpty(searchChars)) {! return -1;! }! for (int i = 0; i < str.length(); i++) {! char ch = str.charAt(i);! for (int j = 0; j < searchChars.length; j++) {! if (searchChars[j] == ch) {! return i;! }! }! }! return -1;! } indexOfAny Impl
public static int indexOfAny(String str, char[] searchChars) {! when (searchChars)! for (int i = 0; i < str.length(); i++) {! char ch = str.charAt(i);! for (int j = 0; j < searchChars.length; j++) {! if (searchChars[j] == ch) {! return i;! }! }! }! }! } - Corner Cases
indexOfAny(str, searchChars) {! when (searchChars)! for (i = 0; i < str.length(); i++) {! ch = str.charAt(i);! for (j = 0; j < searchChars.length; j++) {! if (searchChars[j] == ch) {! return i;! }! }! }! }! } - Type Decls
indexOfAny(str, searchChars) {! when (searchChars)! for (i = 0; i < str.length(); i++) {! ch = str.charAt(i); ! when searchChars(ch) i;! }! }! } + When Clause
indexOfAny(str, searchChars) {! when (searchChars)! for ([i, ch] in indexed(str)) {! when searchChars(ch) i;! }! }! } + Comprehension
Lispify (defn index-filter [pred coll]! (when pred ! (for [[idx elt] (indexed coll) :when (pred elt)] idx)))
Benefits of Clojure concision generality robustness agility
imperative functional searches strings searches any sequence matches characters matches any predicate returns first match returns lazy seq of all matches
; idxs of heads in stream of coin flips! (index-filter #{:h} [:t :t :h :t :h :t :t :t :h :h]) ! -> (2 4 8 9)! ! ; Fibonaccis pass 1000 at n=17! (first ! (index-filter #(> % 1000) (fibo)))! -> 17 + Generality
Clojure programs can have fewer lines of code than OO programs have files
Benefits of Clojure concision generality robustness agility
imperative functional functions 1 1 classes 1 0 internal exit points 2 0 variables 3 0 branches 4 0 boolean ops 1 0 function calls* 6 3 total 18 4
Benefits of Clojure concision generality robustness agility
Plain Immutable Collection Objects (PICOs)
PICOS Everywhere collections directories files XML JSON result sets web requests web responses sessions configuration metrics logs
Consuming JSON What actors are in more than one movie currently topping the box office charts? http://developer.rottentomatoes.com/docs/ read/json/v10/Box_Office_Movies
Consuming JSON find the JSON input! download it! parse json! walk the movies! accumulating cast! extract actor name! get frequencies! sort by highest frequency http://developer.rottentomatoes.com/docs/ read/json/v10/Box_Office_Movies
Consuming JSON (->> box-office-uri! slurp! json/read-json! :movies! (mapcat :abridged_cast)! (map :name)! frequencies! (sort-by (comp - second))) http://developer.rottentomatoes.com/docs/ read/json/v10/Box_Office_Movies
Consuming JSON ["Shiloh Fernandez" 2] ! ["Ray Liotta" 2] ! ["Isla Fisher" 2] ! ["Bradley Cooper" 2] ! ["Dwayne "The Rock" Johnson" 2] ! ["Morgan Freeman" 2] ! ["Michael Shannon" 2] ! ["Joel Edgerton" 2] ! ["Susan Sarandon" 2] ! ["Leonardo DiCaprio" 2] http://developer.rottentomatoes.com/docs/ read/json/v10/Box_Office_Movies
PICOs for Big Data (defn my-data-2 [] (->> (pig/load-tsv "input.tsv") (pig/map (fn [[a b c]] {:sum (+ (Integer/valueOf a) (Integer/valueOf b)) :name c})) (pig/filter (fn [{:keys [sum]}] (< sum 5))))) ! => (pig/dump (my-data-2)) [{:sum 3, :name "foo"}] https://github.com/Netflix/PigPen
Me 182 nachos 184 reference value pure function new value succession function
Me t-1 transact t-2 connection db value pure function new db value ACID
ACID data of record persistent data structures: “scm for business data” distributed, componentized, read scalable & elastic information and logic as PICOs in any peer process
Connect and Query Connection conn = ! connect("datomic:ddb://us-east-1/mb/mbrainz");! ! ! Database db = conn.db();! ! ! Set results = q(..., db);! ! ! Set crossDbResults = q(..., db1, db2);! ! ! Entity e = db.entity(42);
Connect and Query Connection conn = ! connect("datomic:ddb://us-east-1/mb/mbrainz");! ! ! Database db = conn.db();! ! ! Set results = q(..., db);! ! ! Set crossDbResults = q(..., db1, db2);! ! ! Entity e = db.entity(42); database is a lazily realized value, available to all peers equally
Producing Clojure
Design ! specification of an artifact using components to meet goals subject to constraints
Goal ! give skilled devs superpowers to build business software systems
Goal ! give skilled devs superpowers to build business software systems
Goal ! give skilled devs superpowers to build business software systems
Constraints for wide adoption open source target established platforms for viability performance stability
Constraints for wide adoption open source target established platforms for viability performance stability
Open Source licensed under EPL contributor agreement artifacts in Maven Central not just language: bunch of libs too
Might Surprise You we take patches, not pull requests we prefer designs over patches we prefer problem statements over designs
Constraints for wide adoption open source target established platforms for viability performance stability
Server Performance Clojure note: log scale! http://benchmarksgame.alioth.debian.org/u64q/which-programs-are-fastest.php
Constraints for wide adoption open source target established platforms for viability performance stability
2009
Maintaining Programming Clojure release date breakage* 1.0 05/2009 - 1.1 12/2009 None 1.2 08/2010 None 1.3 09/2011 Small 1.4 04/2012 None 1.5 03/2013 None 1.6 03/2014 None 1.7 TBD None
One size does not fit all
Examples of Simplicity syntax protocols values and references PICOs!
Benefits of Clojure concision generality robustness agility
@stuarthalloway
Clojure http://clojure.com. The Clojure language. http://cognitect.com/. The company behind Clojure, ClojureScript, & Datomic. http://blog.cognitect.com/cognicast/. The Cognicast. http://bit.ly/clojure-bookshelf. 40 recommendations from Rich. http://clojure.in/. Planet Clojure. ! @stuarthalloway https://github.com/stuarthalloway/presentations/wiki. Presentations. https://github.com/stuarthalloway/exploring-clojure. Sample Code. http://pragprog.com/book/shcloj2/programming-clojure. Programming Clojure. mailto:stu@cognitect.com

More Related Content

Clojure: Simple By Design

  • 1. Simple by Design: Clojure @stuarthalloway stu@cognitect.com Copyright Cognitect, Inc. This presentation is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License. See http://creativecommons.org/licenses/by-nc-sa/3.0/us/
  • 2. Thoughtworks Radar Oct 2012: Adopt Clojure http://www.thoughtworks.com/radar
  • 3. Redmonk Top 20 Clojure http://redmonk.com/sogrady/2014/06/13/language-rankings-6-14/
  • 4. Design ! specification of an artifact using components to meet goals subject to constraints
  • 5. Simple ! not compound
  • 6. Valuable But Not Simple convenience beginner-friendliness ease familiarity smaller size lesser count
  • 7. Simple ! not compound
  • 8. Agenda examples of simplicity in Clojure benefits of simplicity in Clojure producing Clojure
  • 9. Examples of Simplicity syntax protocols values and references
  • 10. Examples of Simplicity syntax protocols values and references
  • 12. "Hello World" that is the program
  • 14. type examples string "foo" character f integer 42, 42N floating point 3.14, 3.14M boolean true nil nil symbol foo, + keyword :foo, ::foo
  • 15. type properties examples list sequential (1 2 3) vector sequential and random access [1 2 3] map associative {:a 100! :b 90} set membership #{:a :b}
  • 16. Function Call semantics: fn call args (+ 2 2) structure: symbol longs list
  • 17. Function Definition define a fn fn name docstring (defn greet! "Returns a friendly greeting"! [your-name]! (str "Hello, " your-name)) arguments fn body
  • 18. …Still Just Data symbol symbol string (defn greet! "Returns a friendly greeting"! [your-name]! (str "Hello, " your-name)) vector list
  • 19. Complexities Avoided lots of syntax to learn ordering dependencies operator precedence code over data bias tedious metaprogramming
  • 20. Examples of Simplicity syntax protocols values and references
  • 23. Reversible interface implementation extension Jacket Belt String
  • 24. (defprotocol Reversible! (reverse [_]))! ! ! ! (defrecord ReversibleTie [a b])! ! ! ! (extend-protocol Reversible! ReversibleTie! (reverse [tie] (->ReversibleTie (:b tie) (:a tie)))! String! (reverse [s] (-> s! StringBuilder.! .reverse! .toString))) interface implementation extension
  • 25. Complexities Avoided adapter pattern wrapper pattern translator pattern monkey patching StringUtils note that these are all combinatorial
  • 26. Examples of Simplicity syntax protocols values and references
  • 30. Watch OO Flounder Me 182 nachos 184 instance? field? method? ??
  • 31. If you have more things than names, your design is broken.
  • 32. Clojure’s Simplicity Me 182 nachos 184 reference value pure function succession function new value
  • 33. Values and References (defprotocol Nachos! (yum [_] "eat some nachos"))! ! (defrecord Person [name lbs]! Nachos! (yum [person]! (update-in person [:lbs] + 2)))! ! (def me (atom (->Person "Stu" 182)))! ! (def me-before @me)! ! (swap! me yum)! ! (def me-after @me)
  • 34. Values and References (defprotocol Nachos! (yum [_] "eat some nachos"))! ! (defrecord Person [name lbs]! Nachos! (yum [person]! functional (update-in person [:lbs] + 2)))! ! (def me (atom (->Person "Stu" 182)))! ! (def me-before @me)! ! (swap! me yum)! ! (def me-after @me)
  • 35. Values and References (defprotocol Nachos! (yum [_] "eat some nachos"))! ! (defrecord Person [name lbs]! Nachos! (yum [person]! (update-in person [:lbs] + 2)))! ! (def me (atom (->Person "Stu" 182)))! ! (def me-before @me)! ! (swap! me yum)! ! (def me-after @me) update semantics
  • 36. Values and References (defprotocol Nachos! (yum [_] "eat some nachos"))! ! (defrecord Person [name lbs]! Nachos! (yum [person]! (update-in person [:lbs] + 2)))! ! (def me (atom (->Person "Stu" 182)))! ! (def me-before @me)! ! (swap! me yum)! ! (def me-after @me) multiple point-in-time values
  • 37. Complexities Avoided incidental complexity temporal reasoning single-threading locking defensive copying setter methods String vs. StringBuilder vs. StringBuffer note (again!) that these are all combinatorial
  • 38. Benefits of Clojure concision generality robustness agility
  • 39. Benefits of Clojure concision generality robustness agility
  • 41. indexOfAny Spec StringUtils.indexOfAny(null, *) = -1! StringUtils.indexOfAny("", *) = -1! StringUtils.indexOfAny(*, null) = -1! StringUtils.indexOfAny(*, []) = -1! StringUtils.indexOfAny("zzabyycdxx",['z','a']) = 0! StringUtils.indexOfAny("zzabyycdxx",['b','y']) = 3! StringUtils.indexOfAny("aba", ['z']) = -1
  • 42. // From Apache Commons Lang, http://commons.apache.org/lang/! public static int indexOfAny(String str, char[] searchChars) {! if (isEmpty(str) || ArrayUtils.isEmpty(searchChars)) {! return -1;! }! for (int i = 0; i < str.length(); i++) {! char ch = str.charAt(i);! for (int j = 0; j < searchChars.length; j++) {! if (searchChars[j] == ch) {! return i;! }! }! }! return -1;! } indexOfAny Impl
  • 43. public static int indexOfAny(String str, char[] searchChars) {! when (searchChars)! for (int i = 0; i < str.length(); i++) {! char ch = str.charAt(i);! for (int j = 0; j < searchChars.length; j++) {! if (searchChars[j] == ch) {! return i;! }! }! }! }! } - Corner Cases
  • 44. indexOfAny(str, searchChars) {! when (searchChars)! for (i = 0; i < str.length(); i++) {! ch = str.charAt(i);! for (j = 0; j < searchChars.length; j++) {! if (searchChars[j] == ch) {! return i;! }! }! }! }! } - Type Decls
  • 45. indexOfAny(str, searchChars) {! when (searchChars)! for (i = 0; i < str.length(); i++) {! ch = str.charAt(i); ! when searchChars(ch) i;! }! }! } + When Clause
  • 46. indexOfAny(str, searchChars) {! when (searchChars)! for ([i, ch] in indexed(str)) {! when searchChars(ch) i;! }! }! } + Comprehension
  • 47. Lispify (defn index-filter [pred coll]! (when pred ! (for [[idx elt] (indexed coll) :when (pred elt)] idx)))
  • 48. Benefits of Clojure concision generality robustness agility
  • 49. imperative functional searches strings searches any sequence matches characters matches any predicate returns first match returns lazy seq of all matches
  • 50. ; idxs of heads in stream of coin flips! (index-filter #{:h} [:t :t :h :t :h :t :t :t :h :h]) ! -> (2 4 8 9)! ! ; Fibonaccis pass 1000 at n=17! (first ! (index-filter #(> % 1000) (fibo)))! -> 17 + Generality
  • 51. Clojure programs can have fewer lines of code than OO programs have files
  • 52. Benefits of Clojure concision generality robustness agility
  • 53. imperative functional functions 1 1 classes 1 0 internal exit points 2 0 variables 3 0 branches 4 0 boolean ops 1 0 function calls* 6 3 total 18 4
  • 54. Benefits of Clojure concision generality robustness agility
  • 55. Plain Immutable Collection Objects (PICOs)
  • 56. PICOS Everywhere collections directories files XML JSON result sets web requests web responses sessions configuration metrics logs
  • 57. Consuming JSON What actors are in more than one movie currently topping the box office charts? http://developer.rottentomatoes.com/docs/ read/json/v10/Box_Office_Movies
  • 58. Consuming JSON find the JSON input! download it! parse json! walk the movies! accumulating cast! extract actor name! get frequencies! sort by highest frequency http://developer.rottentomatoes.com/docs/ read/json/v10/Box_Office_Movies
  • 59. Consuming JSON (->> box-office-uri! slurp! json/read-json! :movies! (mapcat :abridged_cast)! (map :name)! frequencies! (sort-by (comp - second))) http://developer.rottentomatoes.com/docs/ read/json/v10/Box_Office_Movies
  • 60. Consuming JSON ["Shiloh Fernandez" 2] ! ["Ray Liotta" 2] ! ["Isla Fisher" 2] ! ["Bradley Cooper" 2] ! ["Dwayne "The Rock" Johnson" 2] ! ["Morgan Freeman" 2] ! ["Michael Shannon" 2] ! ["Joel Edgerton" 2] ! ["Susan Sarandon" 2] ! ["Leonardo DiCaprio" 2] http://developer.rottentomatoes.com/docs/ read/json/v10/Box_Office_Movies
  • 61. PICOs for Big Data (defn my-data-2 [] (->> (pig/load-tsv "input.tsv") (pig/map (fn [[a b c]] {:sum (+ (Integer/valueOf a) (Integer/valueOf b)) :name c})) (pig/filter (fn [{:keys [sum]}] (< sum 5))))) ! => (pig/dump (my-data-2)) [{:sum 3, :name "foo"}] https://github.com/Netflix/PigPen
  • 62. Me 182 nachos 184 reference value pure function new value succession function
  • 63. Me t-1 transact t-2 connection db value pure function new db value ACID
  • 64. ACID data of record persistent data structures: “scm for business data” distributed, componentized, read scalable & elastic information and logic as PICOs in any peer process
  • 65. Connect and Query Connection conn = ! connect("datomic:ddb://us-east-1/mb/mbrainz");! ! ! Database db = conn.db();! ! ! Set results = q(..., db);! ! ! Set crossDbResults = q(..., db1, db2);! ! ! Entity e = db.entity(42);
  • 66. Connect and Query Connection conn = ! connect("datomic:ddb://us-east-1/mb/mbrainz");! ! ! Database db = conn.db();! ! ! Set results = q(..., db);! ! ! Set crossDbResults = q(..., db1, db2);! ! ! Entity e = db.entity(42); database is a lazily realized value, available to all peers equally
  • 68. Design ! specification of an artifact using components to meet goals subject to constraints
  • 69. Goal ! give skilled devs superpowers to build business software systems
  • 70. Goal ! give skilled devs superpowers to build business software systems
  • 71. Goal ! give skilled devs superpowers to build business software systems
  • 72. Constraints for wide adoption open source target established platforms for viability performance stability
  • 73. Constraints for wide adoption open source target established platforms for viability performance stability
  • 74. Open Source licensed under EPL contributor agreement artifacts in Maven Central not just language: bunch of libs too
  • 75. Might Surprise You we take patches, not pull requests we prefer designs over patches we prefer problem statements over designs
  • 76. Constraints for wide adoption open source target established platforms for viability performance stability
  • 77. Server Performance Clojure note: log scale! http://benchmarksgame.alioth.debian.org/u64q/which-programs-are-fastest.php
  • 78. Constraints for wide adoption open source target established platforms for viability performance stability
  • 79. 2009
  • 80. Maintaining Programming Clojure release date breakage* 1.0 05/2009 - 1.1 12/2009 None 1.2 08/2010 None 1.3 09/2011 Small 1.4 04/2012 None 1.5 03/2013 None 1.6 03/2014 None 1.7 TBD None
  • 81. One size does not fit all
  • 82. Examples of Simplicity syntax protocols values and references PICOs!
  • 83. Benefits of Clojure concision generality robustness agility
  • 85. Clojure http://clojure.com. The Clojure language. http://cognitect.com/. The company behind Clojure, ClojureScript, & Datomic. http://blog.cognitect.com/cognicast/. The Cognicast. http://bit.ly/clojure-bookshelf. 40 recommendations from Rich. http://clojure.in/. Planet Clojure. ! @stuarthalloway https://github.com/stuarthalloway/presentations/wiki. Presentations. https://github.com/stuarthalloway/exploring-clojure. Sample Code. http://pragprog.com/book/shcloj2/programming-clojure. Programming Clojure. mailto:stu@cognitect.com