redis_pali_word_statistic.php 5.9 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168
  1. <?php
  2. require_once "../config.php";
  3. require_once "../install/filelist.php";
  4. require_once "../redis/function.php";
  5. $redis = redis_connect();
  6. $strKey='pali://wordstatisitic.hash';
  7. if (PHP_SAPI == "cli") {
  8. if ($argc >= 2) {
  9. $command = $argv[1];
  10. } else {
  11. exit;
  12. }
  13. {
  14. if ($redis == false) {
  15. echo "no redis connect\n";
  16. exit;
  17. }
  18. switch ($command) {
  19. case 'init':
  20. # code...
  21. $dirXmlBase = _DIR_PALI_CSV_ . "/";
  22. $book = array(1, 2, 3, 4, 5, 6, 7, 8, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 153, 152, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217);
  23. $redis->delete($strKey);
  24. foreach ($book as $key => $value) {
  25. # code...
  26. echo "runing:{$value}\n";
  27. $outputFileNameHead = $_filelist[$value];
  28. $dirXml = $outputFileNameHead . "/";
  29. // 打开文件并读取数据
  30. $irow = 0;
  31. if (($fp = fopen($dirXmlBase . $dirXml . $outputFileNameHead . ".csv", "r")) !== false) {
  32. while (($str = fgets($fp)) !== false) {
  33. $data = explode(",", $str);
  34. $irow++;
  35. if ($irow > 1) {
  36. if ($data[6] != ".ctl." && $data[5] != "") {
  37. $word = $redis->hGet($strKey, $data[5]);
  38. if($word){
  39. $eWord = json_decode($word);
  40. $eWord->count++;
  41. }else{
  42. $eWord = [];
  43. $eWord["word"]=$data[5];
  44. $eWord["count"]=1;
  45. $eWord["ref"]=0;
  46. $eWord["user"]=0;
  47. $eWord["len"]=mb_strlen($data[5],"UTF-8");
  48. }
  49. $redis->hSet($strKey, $data[5],json_encode($eWord,JSON_UNESCAPED_UNICODE));
  50. }
  51. }
  52. }
  53. fclose($fp);
  54. } else {
  55. echo "can not open csv file. filename=" . $dirXmlBase . $dirXml . $outputFileNameHead . ".csv";
  56. }
  57. }
  58. echo "hash done".$redis->hLen($strKey).PHP_EOL;
  59. break;
  60. case "ref":
  61. $dbh = new PDO(_DICT_DB_REGULAR_, "", "", array(PDO::ATTR_PERSISTENT => true,PDO::SQLITE_ATTR_OPEN_FLAGS => PDO::SQLITE_OPEN_READONLY));
  62. $dbh->setAttribute(PDO::ATTR_ERRMODE, PDO::ERRMODE_WARNING);
  63. $query = "SELECT pali from "._TABLE_DICT_REGULAR_." where 1 group by pali";
  64. $stmt = $dbh->prepare($query);
  65. $stmt->execute();
  66. $count = 0;
  67. while ($row = $stmt->fetch(PDO::FETCH_ASSOC)) {
  68. # code...
  69. $word = $redis->hGet($strKey, $row["pali"]);
  70. if($word){
  71. $eWord = json_decode($word,true);
  72. $eWord["ref"]=1;
  73. $redis->hSet($strKey, $row["pali"],json_encode($eWord,JSON_UNESCAPED_UNICODE));
  74. $count++;
  75. }
  76. }
  77. echo "regular count:".$count.PHP_EOL;
  78. $dbh = new PDO(_DICT_DB_IRREGULAR_, "", "", array(PDO::ATTR_PERSISTENT => true,PDO::SQLITE_ATTR_OPEN_FLAGS => PDO::SQLITE_OPEN_READONLY));
  79. $dbh->setAttribute(PDO::ATTR_ERRMODE, PDO::ERRMODE_WARNING);
  80. $query = "SELECT pali from "._TABLE_DICT_IRREGULAR_." where 1 group by pali";
  81. $stmt = $dbh->prepare($query);
  82. $stmt->execute();
  83. $count = 0;
  84. while ($row = $stmt->fetch(PDO::FETCH_ASSOC)) {
  85. # code...
  86. $word = $redis->hGet($strKey, $row["pali"]);
  87. if($word){
  88. $eWord = json_decode($word,true);
  89. $eWord["ref"]=1;
  90. $redis->hSet($strKey, $row["pali"],json_encode($eWord,JSON_UNESCAPED_UNICODE));
  91. $count++;
  92. }
  93. }
  94. echo "irregular count:".$count.PHP_EOL;
  95. break;
  96. case 'update':
  97. $dbh = new PDO(_FILE_DB_WBW_, "", "", array(PDO::ATTR_PERSISTENT => true,PDO::SQLITE_ATTR_OPEN_FLAGS => PDO::SQLITE_OPEN_READONLY));
  98. $dbh->setAttribute(PDO::ATTR_ERRMODE, PDO::ERRMODE_WARNING);
  99. $query = "SELECT pali from dict where 1 group by pali";
  100. $stmt = $dbh->prepare($query);
  101. $stmt->execute();
  102. $count = 0;
  103. while ($row = $stmt->fetch(PDO::FETCH_ASSOC)) {
  104. # code...
  105. $word = $redis->hGet($strKey, $row["pali"]);
  106. if($word){
  107. $eWord = json_decode($word,true);
  108. $eWord["user"]=1;
  109. $redis->hSet($strKey, $row["pali"],json_encode($eWord,JSON_UNESCAPED_UNICODE));
  110. $count++;
  111. }
  112. }
  113. echo "user wbw count:".$count.PHP_EOL;
  114. $start=1;
  115. $all=0;
  116. $ref=0;
  117. $user=0;
  118. $cover=0;
  119. $all1=0;
  120. $ref1=0;
  121. $user1=0;
  122. $cover1=0;
  123. while($word = $redis->hGet("pali://wordindex.hash",$start)){
  124. $eWord = $redis->hGet($strKey, $word);
  125. $eWord = json_decode($eWord,true);
  126. $all += $eWord["count"]*$eWord["len"];
  127. $all1++;
  128. $ref += $eWord["ref"]*$eWord["len"];
  129. $ref1 += $eWord["ref"];
  130. $user += $eWord["user"]*$eWord["len"];
  131. $user1 += $eWord["user"];
  132. if($eWord["ref"]==1 || $eWord["user"]==1){
  133. $cover += $eWord["len"];
  134. $cover1 += 1;
  135. }
  136. $start++;
  137. }
  138. $file = fopen(_DIR_LOG_."/dict_cover.log","a");
  139. if($file){
  140. fputs($file,date("Y-m-d h:i:sa").sprintf("字符数 全部:{$all},参考:{$ref},用户字典:{$user},总和:{$cover},单词数 全部:{$all1},参考:{$ref1},用户字典:{$user1},总和:{$cover1}",$all,$ref,$user,$cover,$all1,$ref1,$user1,$cover1));
  141. fclose($file);
  142. }
  143. echo "字符数 全部:{$all},参考:{$ref},用户字典:{$user},总和:{$cover}".PHP_EOL;
  144. echo "单词数 全部:{$all1},参考:{$ref1},用户字典:{$user1},总和:{$cover1}".PHP_EOL;
  145. break;
  146. default:
  147. # code...
  148. break;
  149. }
  150. }
  151. } else {
  152. echo " null ".PHP_EOL;
  153. }
  154. echo "<h2>齐活!功德无量!all done!</h2>";