| 37 | | |
| 38 | | == 搭配 PHP Script == |
| 39 | | |
| 40 | | * 編輯 mapper 的 php 程式 |
| 41 | | {{{ |
| 42 | | ~$ cat > mapper.php << EOF |
| 43 | | #!/usr/bin/php |
| 44 | | <?php |
| 45 | | |
| 46 | | \$word2count = array(); |
| 47 | | |
| 48 | | // 標準輸入為 STDIN (standard input) |
| 49 | | while ((\$line = fgets(STDIN)) !== false) { |
| 50 | | // 移除小寫與空白 |
| 51 | | \$line = strtolower(trim(\$line)); |
| 52 | | // 將行拆解成各個字於 words 陣列中 |
| 53 | | \$words = preg_split('/\W/', \$line, 0, PREG_SPLIT_NO_EMPTY); |
| 54 | | // 將字 +1 |
| 55 | | foreach (\$words as \$word) { |
| 56 | | \$word2count[\$word] += 1; |
| 57 | | } |
| 58 | | } |
| 59 | | |
| 60 | | // 將結果寫到 STDOUT (standard output) |
| 61 | | foreach (\$word2count as \$word => \$count) { |
| 62 | | // 印出 [字 , "tab符號" , "數字" , "結束字元"] |
| 63 | | echo \$word, chr(9), \$count, PHP_EOL; |
| 64 | | } |
| 65 | | ?> |
| 66 | | EOF |
| 67 | | }}} |
| 68 | | * 編輯 reduce 的 php 程式 |
| 69 | | {{{ |
| 70 | | ~$ cat > reducer.php << EOF |
| 71 | | #!/usr/bin/php |
| 72 | | <?php |
| 73 | | |
| 74 | | \$word2count = array(); |
| 75 | | |
| 76 | | // 輸入為 STDIN |
| 77 | | while ((\$line = fgets(STDIN)) !== false) { |
| 78 | | // 移除多餘的空白 |
| 79 | | \$line = trim(\$line); |
| 80 | | // 每一行的格式為 (單字 "tab" 數字) ，紀錄到(\$word, \$count) |
| 81 | | list(\$word, \$count) = explode(chr(9), \$line); |
| 82 | | // 轉換格式string -> int |
| 83 | | \$count = intval(\$count); |
| 84 | | // 加總 |
| 85 | | if (\$count > 0) \$word2count[\$word] += \$count; |
| 86 | | } |
| 87 | | |
| 88 | | // 此行不必要，但可讓output排列更完整 |
| 89 | | ksort(\$word2count); |
| 90 | | |
| 91 | | // 將結果寫到 STDOUT (standard output) |
| 92 | | foreach (\$word2count as \$word => \$count) { |
| 93 | | echo \$word, chr(9), \$count, PHP_EOL; |
| 94 | | } |
| 95 | | ?> |
| 96 | | EOF |
| 97 | | }}} |
| 98 | | * 修改執行權限 |
| 99 | | {{{ |
| 100 | | ~$ chmod a+x *.php |
| 101 | | }}} |
| 102 | | * 測試是否能運作 |
| 103 | | {{{ |
| 104 | | ~$ echo "i love hadoop, hadoop love u" | ./mapper.php | ./reducer.php |
| 105 | | }}} |
| 106 | | * 開始執行 |
| 107 | | {{{ |
| 108 | | ~$ hadoop jar hadoop-streaming.jar -mapper mapper.php -reducer reducer.php -input lab9_input -output lab9_out3 -file mapper.php -file reducer.php |
| 109 | | }}} |
| 110 | | * 檢查結果 |
| 111 | | {{{ |
| 112 | | ~$ hadoop fs -cat lab9_out3/part-00000 |
| 113 | | }}} |
