{"id":193075,"date":"2020-11-16T12:47:19","date_gmt":"2020-11-16T04:47:19","guid":{"rendered":"http:\/\/4563.org\/?p=193075"},"modified":"2020-11-16T12:47:19","modified_gmt":"2020-11-16T04:47:19","slug":"hive-%e8%87%aa%e5%ae%9a%e4%b9%89%e5%87%bd%e6%95%b0%e6%b5%85%e8%b0%88%ef%bc%88udf%e3%80%81udaf%e3%80%81udtf%ef%bc%89","status":"publish","type":"post","link":"http:\/\/4563.org\/?p=193075","title":{"rendered":"hive \u81ea\u5b9a\u4e49\u51fd\u6570\u6d45\u8c08\uff08UDF\u3001UDAF\u3001UDTF\uff09"},"content":{"rendered":"<h2><a href=\"https:\/\/www.studytime.xin\/article\/#hive\u4e3a\u4ec0\u4e48\u9700\u8981\u81ea\u5b9a\u4e49\u51fd\u6570\" title=\"hive\u4e3a\u4ec0\u4e48\u9700\u8981\u81ea\u5b9a\u4e49\u51fd\u6570\"><\/a>hive\u4e3a\u4ec0\u4e48\u9700\u8981\u81ea\u5b9a\u4e49\u51fd\u6570<\/h2>\n<p>hive\u7684\u5185\u7f6e\u51fd\u6570\u6ee1\u8db3\u4e0d\u4e86\u6240\u6709\u7684\u4e1a\u52a1\u9700\u6c42\uff0chive\u63d0\u4f9b\u5f88\u591a\u7684\u6a21\u5757\u53ef\u4ee5\u81ea\u5b9a\u4e49\u529f\u80fd\uff0c\u6bd4\u5982\uff1a\u81ea\u5b9a\u4e49\u51fd\u6570\u3001serde\u3001\u8f93\u5165\u8f93\u51fa\u683c\u5f0f\u7b49\u3002<\/p>\n<h2><a href=\"https:\/\/www.studytime.xin\/article\/#\u5e38\u89c1\u81ea\u5b9a\u4e49\u51fd\u6570\u6709\u54ea\u4e9b\" title=\"\u5e38\u89c1\u81ea\u5b9a\u4e49\u51fd\u6570\u6709\u54ea\u4e9b\"><\/a>\u5e38\u89c1\u81ea\u5b9a\u4e49\u51fd\u6570\u6709\u54ea\u4e9b<\/h2>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#UDF\" title=\"UDF\"><\/a>UDF<\/h3>\n<p>\u7528\u6237\u81ea\u5b9a\u4e49\u51fd\u6570\uff0cuser defined function\u3002\u4e00\u5bf9\u4e00\u7684\u8f93\u5165\u8f93\u51fa\u3002\uff08\u6700\u5e38\u7528\u7684\uff09\u3002<\/p>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#UDTF\" title=\"UDTF\"><\/a>UDTF<\/h3>\n<p>\u7528\u6237\u81ea\u5b9a\u4e49\u8868\u751f\u6210\u51fd\u6570\u3002user defined table-generate function.\u4e00\u5bf9\u591a\u7684\u8f93\u5165\u8f93\u51fa\u3002lateral view explode<\/p>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#UDAF\" title=\"UDAF\"><\/a>UDAF<\/h3>\n<p>\u7528\u6237\u81ea\u5b9a\u4e49\u805a\u5408\u51fd\u6570\u3002user defined aggregate function\u3002\u591a\u5bf9\u4e00\u7684\u8f93\u5165\u8f93\u51fa count sum max\u3002<\/p>\n<h2><a href=\"https:\/\/www.studytime.xin\/article\/#\u81ea\u5b9a\u4e49\u51fd\u6570\u5b9e\u73b0\" title=\"\u81ea\u5b9a\u4e49\u51fd\u6570\u5b9e\u73b0\"><\/a>\u81ea\u5b9a\u4e49\u51fd\u6570\u5b9e\u73b0<\/h2>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#UDF\u683c\u5f0f\" title=\"UDF\u683c\u5f0f\"><\/a>UDF\u683c\u5f0f<\/h3>\n<p>\u5148\u5728\u5de5\u7a0b\u4e0b\u65b0\u5efa\u4e00\u4e2apom.xml,\u52a0\u5165\u4ee5\u4e0bmaven\u7684\u4f9d\u8d56\u5305 \u8bf7\u67e5\u770b<code>code\/pom.xml<\/code><\/p>\n<p>\u5b9a\u4e49UDF\u51fd\u6570\u8981\u6ce8\u610f\u4e0b\u9762\u51e0\u70b9:<\/p>\n<ol>\n<li>\u7ee7\u627f<code>org.apache.hadoop.hive.ql.exec.UDF<\/code><\/li>\n<li>\u91cd\u5199<code>evaluate<\/code>()\uff0c\u8fd9\u4e2a\u65b9\u6cd5\u4e0d\u662f\u7531\u63a5\u53e3\u5b9a\u4e49\u7684,\u56e0\u4e3a\u5b83\u53ef\u63a5\u53d7\u7684\u53c2\u6570\u7684\u4e2a\u6570,\u6570\u636e\u7c7b\u578b\u90fd\u662f\u4e0d\u786e\u5b9a\u7684\u3002Hive\u4f1a\u68c0\u67e5UDF,\u770b\u80fd\u5426\u627e\u5230\u548c\u51fd\u6570\u8c03\u7528\u76f8\u5339\u914d\u7684evaluate()\u65b9\u6cd5<\/li>\n<\/ol>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#\u81ea\u5b9a\u4e49\u51fd\u6570\u6848\u4f8b\" title=\"\u81ea\u5b9a\u4e49\u51fd\u6570\u6848\u4f8b\"><\/a>\u81ea\u5b9a\u4e49\u51fd\u6570\u6848\u4f8b<\/h3>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><span>2<\/span><br><span>3<\/span><br><span>4<\/span><br><span>5<\/span><br><span>6<\/span><br><span>7<\/span><br><span>8<\/span><br><span>9<\/span><br><span>10<\/span><br><span>11<\/span><br><span>12<\/span><br><span>13<\/span><br><span>14<\/span><br><span>15<\/span><br><span>16<\/span><br><span>17<\/span><br><span>18<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span><span>public<\/span> <span><span>class<\/span> <span>FirstUDF<\/span> <span>extends<\/span> <span>UDF<\/span> <\/span>&#123;<\/span><br><span>    <span><span>public<\/span> String <span>evaluate<\/span><span>(String str)<\/span><\/span>&#123;<\/span><br><span>        String upper = <span>null<\/span>;<\/span><br><span>        <span>\/\/1\u3001\u68c0\u67e5\u8f93\u5165\u53c2\u6570<\/span><\/span><br><span>        <span>if<\/span> (StringUtils.isEmpty(str))&#123;<\/span><br><span>\u200b<\/span><br><span>        &#125; <span>else<\/span> &#123;<\/span><br><span>            upper = str.toUpperCase();<\/span><br><span>        &#125;<\/span><br><span>\u200b<\/span><br><span>        <span>return<\/span> upper;<\/span><br><span>    &#125;<\/span><br><span>\u200b<\/span><br><span>    <span>\/\/\u8c03\u8bd5\u81ea\u5b9a\u4e49\u51fd\u6570<\/span><\/span><br><span>    <span><span>public<\/span> <span>static<\/span> <span>void<\/span> <span>main<\/span><span>(String[] args)<\/span><\/span>&#123;<\/span><br><span>        System.out.println(<span>new<\/span> firstUDF().evaluate(<span>\"studytime\"<\/span>));<\/span><br><span>    &#125;<\/span><br><span>&#125;<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h2><a href=\"https:\/\/www.studytime.xin\/article\/#\u51fd\u6570\u52a0\u8f7d\u65b9\u5f0f\" title=\"\u51fd\u6570\u52a0\u8f7d\u65b9\u5f0f\"><\/a>\u51fd\u6570\u52a0\u8f7d\u65b9\u5f0f<\/h2>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#\u547d\u4ee4\u52a0\u8f7d\uff0c\u8fd9\u79cd\u52a0\u8f7d\u53ea\u5bf9\u672csession\u6709\u6548\" title=\"\u547d\u4ee4\u52a0\u8f7d\uff0c\u8fd9\u79cd\u52a0\u8f7d\u53ea\u5bf9\u672csession\u6709\u6548\"><\/a>\u547d\u4ee4\u52a0\u8f7d\uff0c\u8fd9\u79cd\u52a0\u8f7d\u53ea\u5bf9\u672csession\u6709\u6548<\/h3>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u5c06\u7f16\u5199\u7684udf\u7684jar\u5305\u4e0a\u4f20\u5230\u670d\u52a1\u5668\u4e0a\uff0c\u5e76\u4e14\u5c06jar\u5305\u6dfb\u52a0\u5230hive\u7684class-path\u4e2d\" title=\"\u5c06\u7f16\u5199\u7684udf\u7684jar\u5305\u4e0a\u4f20\u5230\u670d\u52a1\u5668\u4e0a\uff0c\u5e76\u4e14\u5c06jar\u5305\u6dfb\u52a0\u5230hive\u7684class path\u4e2d\"><\/a>\u5c06\u7f16\u5199\u7684udf\u7684jar\u5305\u4e0a\u4f20\u5230\u670d\u52a1\u5668\u4e0a\uff0c\u5e76\u4e14\u5c06jar\u5305\u6dfb\u52a0\u5230hive\u7684class path\u4e2d<\/h4>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u8fdb\u5165\u5230hive\u5ba2\u6237\u7aef-\u6267\u884c\u4e0b\u9762\u547d\u4ee4\" title=\"\u8fdb\u5165\u5230hive\u5ba2\u6237\u7aef,\u6267\u884c\u4e0b\u9762\u547d\u4ee4\"><\/a>\u8fdb\u5165\u5230hive\u5ba2\u6237\u7aef,\u6267\u884c\u4e0b\u9762\u547d\u4ee4<\/h4>\n<p><code>add jar \/hivedata\/udf.jar<\/code><\/p>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u521b\u5efa\u4e00\u4e2a\u4e34\u65f6\u51fd\u6570\u540d-\u8981\u8ddf\u4e0a\u9762hive\u5728\u540c\u4e00\u4e2asession\u91cc\u9762\uff1a\" title=\"\u521b\u5efa\u4e00\u4e2a\u4e34\u65f6\u51fd\u6570\u540d,\u8981\u8ddf\u4e0a\u9762hive\u5728\u540c\u4e00\u4e2asession\u91cc\u9762\uff1a\"><\/a>\u521b\u5efa\u4e00\u4e2a\u4e34\u65f6\u51fd\u6570\u540d,\u8981\u8ddf\u4e0a\u9762hive\u5728\u540c\u4e00\u4e2asession\u91cc\u9762\uff1a<\/h4>\n<p><code>create temporary function toUP as &#39;com.qf.hive.FirstUDF&#39;;<\/code><br \/>\u200b<\/p>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u68c0\u67e5\u51fd\u6570\u662f\u5426\u521b\u5efa\u6210\u529f\" title=\"\u68c0\u67e5\u51fd\u6570\u662f\u5426\u521b\u5efa\u6210\u529f\"><\/a>\u68c0\u67e5\u51fd\u6570\u662f\u5426\u521b\u5efa\u6210\u529f<\/h4>\n<p><code>show functions;<\/code><br \/>\u200b<\/p>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u6d4b\u8bd5\u529f\u80fd\" title=\"\u6d4b\u8bd5\u529f\u80fd\"><\/a>\u6d4b\u8bd5\u529f\u80fd<\/h4>\n<p><code>select toUp(&#39;abcdef&#39;);<\/code><br \/>\u200b<\/p>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u5220\u9664\u51fd\u6570\" title=\"\u5220\u9664\u51fd\u6570\"><\/a>\u5220\u9664\u51fd\u6570<\/h4>\n<p><code>drop temporary function if exists tolow;<\/code><\/p>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#\u542f\u52a8\u53c2\u6570\u52a0\u8f7d-\u4e5f\u662f\u5728\u672csession\u6709\u6548\uff0c\u4e34\u65f6\u51fd\u6570\" title=\"\u542f\u52a8\u53c2\u6570\u52a0\u8f7d,\u4e5f\u662f\u5728\u672csession\u6709\u6548\uff0c\u4e34\u65f6\u51fd\u6570\"><\/a>\u542f\u52a8\u53c2\u6570\u52a0\u8f7d,\u4e5f\u662f\u5728\u672csession\u6709\u6548\uff0c\u4e34\u65f6\u51fd\u6570<\/h3>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u5c06\u7f16\u5199\u7684udf\u7684jar\u5305\u4e0a\u4f20\u5230\u670d\u52a1\u5668\u4e0a\" title=\"\u5c06\u7f16\u5199\u7684udf\u7684jar\u5305\u4e0a\u4f20\u5230\u670d\u52a1\u5668\u4e0a\"><\/a>\u5c06\u7f16\u5199\u7684udf\u7684jar\u5305\u4e0a\u4f20\u5230\u670d\u52a1\u5668\u4e0a<\/h4>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u521b\u5efa\u914d\u7f6e\u6587\u4ef6\" title=\"\u521b\u5efa\u914d\u7f6e\u6587\u4ef6\"><\/a>\u521b\u5efa\u914d\u7f6e\u6587\u4ef6<\/h4>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><span>2<\/span><br><span>3<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span>vi .&#x2F;hive-init<\/span><br><span>add jar &#x2F;hivedata&#x2F;udf.jar;<\/span><br><span>create temporary function toup as &#39;com.qf.hive.FirstUDF&#39;;<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u542f\u52a8hive\u7684\u65f6\u5019\u5e26\u4e0a\u521d\u59cb\u5316\u6587\u4ef6\" title=\"\u542f\u52a8hive\u7684\u65f6\u5019\u5e26\u4e0a\u521d\u59cb\u5316\u6587\u4ef6\"><\/a>\u542f\u52a8hive\u7684\u65f6\u5019\u5e26\u4e0a\u521d\u59cb\u5316\u6587\u4ef6<\/h4>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><span>2<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span>hive -i .&#x2F;hive-init<\/span><br><span>select toup(&#39;abcdef&#39;)<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#\u914d\u7f6e\u6587\u4ef6\u52a0\u8f7d\uff0c\u901a\u8fc7\u914d\u7f6e\u6587\u4ef6\u65b9\u5f0f\u8fd9\u79cd\u53ea\u8981\u7528hive\u547d\u4ee4\u884c\u542f\u52a8\u90fd\u4f1a\u52a0\u8f7d\u51fd\u6570\" title=\"\u914d\u7f6e\u6587\u4ef6\u52a0\u8f7d\uff0c\u901a\u8fc7\u914d\u7f6e\u6587\u4ef6\u65b9\u5f0f\u8fd9\u79cd\u53ea\u8981\u7528hive\u547d\u4ee4\u884c\u542f\u52a8\u90fd\u4f1a\u52a0\u8f7d\u51fd\u6570\"><\/a>\u914d\u7f6e\u6587\u4ef6\u52a0\u8f7d\uff0c\u901a\u8fc7\u914d\u7f6e\u6587\u4ef6\u65b9\u5f0f\u8fd9\u79cd\u53ea\u8981\u7528hive\u547d\u4ee4\u884c\u542f\u52a8\u90fd\u4f1a\u52a0\u8f7d\u51fd\u6570<\/h3>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u5c06\u7f16\u5199\u7684udf\u7684jar\u5305\u4e0a\u4f20\u5230\u670d\u52a1\u5668\u4e0a-1\" title=\"\u5c06\u7f16\u5199\u7684udf\u7684jar\u5305\u4e0a\u4f20\u5230\u670d\u52a1\u5668\u4e0a\"><\/a>\u5c06\u7f16\u5199\u7684udf\u7684jar\u5305\u4e0a\u4f20\u5230\u670d\u52a1\u5668\u4e0a<\/h4>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u5728hive\u7684\u5b89\u88c5\u76ee\u5f55\u7684bin\u76ee\u5f55\u4e0b\u521b\u5efa\u4e00\u4e2a\u914d\u7f6e\u6587\u4ef6\uff0c\u6587\u4ef6\u540d\uff1a-hiverc\" title=\"\u5728hive\u7684\u5b89\u88c5\u76ee\u5f55\u7684bin\u76ee\u5f55\u4e0b\u521b\u5efa\u4e00\u4e2a\u914d\u7f6e\u6587\u4ef6\uff0c\u6587\u4ef6\u540d\uff1a.hiverc\"><\/a>\u5728hive\u7684\u5b89\u88c5\u76ee\u5f55\u7684bin\u76ee\u5f55\u4e0b\u521b\u5efa\u4e00\u4e2a\u914d\u7f6e\u6587\u4ef6\uff0c\u6587\u4ef6\u540d\uff1a.hiverc<\/h4>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><span>2<\/span><br><span>3<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span>vi .&#x2F;bin&#x2F;.hiverc<\/span><br><span>add jar &#x2F;hivedata&#x2F;udf.jar;<\/span><br><span>create temporary function toup as &#39;com.qf.hive.FirstUDF&#39;;<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u542f\u52a8hive\" title=\"\u542f\u52a8hive\"><\/a>\u542f\u52a8hive<\/h4>\n<p><code>show functions ;<\/code><\/p>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#UDTF\u683c\u5f0f\uff0cUDTF\u662f\u4e00\u5bf9\u591a\u7684\u8f93\u5165\u8f93\u51fa-\u5b9e\u73b0UDTF\u9700\u8981\u5b8c\u6210\u4e0b\u9762\u6b65\u9aa4\" title=\"UDTF\u683c\u5f0f\uff0cUDTF\u662f\u4e00\u5bf9\u591a\u7684\u8f93\u5165\u8f93\u51fa,\u5b9e\u73b0UDTF\u9700\u8981\u5b8c\u6210\u4e0b\u9762\u6b65\u9aa4\"><\/a>UDTF\u683c\u5f0f\uff0cUDTF\u662f\u4e00\u5bf9\u591a\u7684\u8f93\u5165\u8f93\u51fa,\u5b9e\u73b0UDTF\u9700\u8981\u5b8c\u6210\u4e0b\u9762\u6b65\u9aa4<\/h3>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u7ee7\u627forg-apache-hadoop-hive-ql-udf-generic-GenericUDF\uff0c\" title=\"\u7ee7\u627forg.apache.hadoop.hive.ql.udf.generic.GenericUDF\uff0c\"><\/a>\u7ee7\u627f<code>org.apache.hadoop.hive.ql.udf.generic.GenericUDF<\/code>\uff0c<\/h4>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u91cd\u5199initlizer\uff08\uff09\u3001getdisplay\uff08\uff09\u3001evaluate-\u3002\" title=\"\u91cd\u5199initlizer\uff08\uff09\u3001getdisplay\uff08\uff09\u3001evaluate()\u3002\"><\/a>\u91cd\u5199initlizer\uff08\uff09\u3001getdisplay\uff08\uff09\u3001evaluate()\u3002<\/h4>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u6267\u884c\u6d41\u7a0b\u5982\u4e0b\" title=\"\u6267\u884c\u6d41\u7a0b\u5982\u4e0b\"><\/a>\u6267\u884c\u6d41\u7a0b\u5982\u4e0b<\/h4>\n<ol>\n<li>UDTF\u9996\u5148\u4f1a\u8c03\u7528initialize\u65b9\u6cd5\uff0c\u6b64\u65b9\u6cd5\u8fd4\u56deUDTF\u7684\u8fd4\u56de\u884c\u7684\u4fe1\u606f\uff08\u8fd4\u56de\u4e2a\u6570\uff0c\u7c7b\u578b\uff09\u3002<\/li>\n<li>\u521d\u59cb\u5316\u5b8c\u6210\u540e\uff0c\u4f1a\u8c03\u7528process\u65b9\u6cd5,\u771f\u6b63\u7684\u5904\u7406\u8fc7\u7a0b\u5728process\u51fd\u6570\u4e2d\uff0c\u5728process\u4e2d\uff0c\u6bcf\u4e00\u6b21forward()\u8c03\u7528\u4ea7\u751f\u4e00\u884c\uff1b\u5982\u679c\u4ea7\u751f\u591a\u5217\u53ef\u4ee5\u5c06\u591a\u4e2a\u5217\u7684\u503c\u653e\u5728\u4e00\u4e2a\u6570\u7ec4\u4e2d\uff0c\u7136\u540e\u5c06\u8be5\u6570\u7ec4\u4f20\u5165\u5230forward()\u51fd\u6570\u3002<\/li>\n<li>\u6700\u540eclose()\u65b9\u6cd5\u8c03\u7528\uff0c\u5bf9\u9700\u8981\u6e05\u7406\u7684\u65b9\u6cd5\u8fdb\u884c\u6e05\u7406\u3002<\/li>\n<\/ol>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#\u81ea\u5b9a\u4e49\u51fd\u6570\u6848\u4f8b-1\" title=\"\u81ea\u5b9a\u4e49\u51fd\u6570\u6848\u4f8b\"><\/a>\u81ea\u5b9a\u4e49\u51fd\u6570\u6848\u4f8b<\/h3>\n<p>\u628a\u201dk1:v1;k2:v2;k3:v3\u201d\u7c7b\u4f3c\u7684\u7684\u5b57\u7b26\u4e32\u89e3\u6790\u6210\u6bcf\u4e00\u884c\u591a\u884c,\u6bcf\u4e00\u884c\u6309\u7167key:value\u683c\u5f0f\u8f93\u51fa<\/p>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u4ee3\u7801\" title=\"\u4ee3\u7801\"><\/a>\u4ee3\u7801<\/h4>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><span>2<\/span><br><span>3<\/span><br><span>4<\/span><br><span>5<\/span><br><span>6<\/span><br><span>7<\/span><br><span>8<\/span><br><span>9<\/span><br><span>10<\/span><br><span>11<\/span><br><span>12<\/span><br><span>13<\/span><br><span>14<\/span><br><span>15<\/span><br><span>16<\/span><br><span>17<\/span><br><span>18<\/span><br><span>19<\/span><br><span>20<\/span><br><span>21<\/span><br><span>22<\/span><br><span>23<\/span><br><span>24<\/span><br><span>25<\/span><br><span>26<\/span><br><span>27<\/span><br><span>28<\/span><br><span>29<\/span><br><span>30<\/span><br><span>31<\/span><br><span>32<\/span><br><span>33<\/span><br><span>34<\/span><br><span>35<\/span><br><span>36<\/span><br><span>37<\/span><br><span>38<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span><span>package<\/span> com.qf.hive;<\/span><br><span>\u200b<\/span><br><span> <span>public<\/span> <span><span>class<\/span> <span>ParseMapUDTF<\/span> <span>extends<\/span> <span>GenericUDTF<\/span><\/span>&#123;<\/span><br><span>     <span>@Override<\/span><\/span><br><span>     <span><span>public<\/span> <span>void<\/span> <span>close<\/span><span>()<\/span> <span>throws<\/span> HiveException <\/span>&#123;<\/span><br><span>     &#125;<\/span><br><span>\u200b<\/span><br><span>     <span>@Override<\/span><\/span><br><span>     <span><span>public<\/span> StructObjectInspector <span>initialize<\/span><span>(ObjectInspector[] args)<\/span><\/span><\/span><br><span><span>             <span>throws<\/span> UDFArgumentException <\/span>&#123;<\/span><br><span>         <span>if<\/span> (args.length != <span>1<\/span>) &#123;<\/span><br><span>             <span>throw<\/span> <span>new<\/span> UDFArgumentLengthException(<span>\" \u53ea\u80fd\u4f20\u5165\u4e00\u4e2a\u53c2\u6570\"<\/span>);<\/span><br><span>         &#125;<\/span><br><span>\u200b<\/span><br><span>         ArrayList&lt;String&gt; fieldNameList = <span>new<\/span> ArrayList&lt;String&gt;();<\/span><br><span>         ArrayList&lt;ObjectInspector&gt; fieldOIs = <span>new<\/span> ArrayList&lt;ObjectInspector&gt;();<\/span><br><span>         fieldNameList.add(<span>\"map\"<\/span>);<\/span><br><span>         fieldOIs.add(PrimitiveObjectInspectorFactory.javaStringObjectInspector);<\/span><br><span>         fieldNameList.add(<span>\"key\"<\/span>);<\/span><br><span>         fieldOIs.add(PrimitiveObjectInspectorFactory.javaStringObjectInspector);<\/span><br><span>\u200b<\/span><br><span>         <span>return<\/span> ObjectInspectorFactory.getStandardStructObjectInspector(fieldNameList,fieldOIs);<\/span><br><span>     &#125;<\/span><br><span>\u200b<\/span><br><span>     <span>@Override<\/span><\/span><br><span>     <span><span>public<\/span> <span>void<\/span> <span>process<\/span><span>(Object[] args)<\/span> <span>throws<\/span> HiveException <\/span>&#123;<\/span><br><span>         String input = args[<span>0<\/span>].toString();<\/span><br><span>         String[] paramString = input.split(<span>\";\"<\/span>);<\/span><br><span>         <span>for<\/span>(<span>int<\/span> i=<span>0<\/span>; i&lt;paramString.length; i++) &#123;<\/span><br><span>             <span>try<\/span> &#123;<\/span><br><span>                 String[] result = paramString[i].split(<span>\":\"<\/span>);<\/span><br><span>                 forward(result);<\/span><br><span>             &#125; <span>catch<\/span> (Exception e) &#123;<\/span><br><span>                 <span>continue<\/span>;<\/span><br><span>             &#125;<\/span><br><span>         &#125;<\/span><br><span>     &#125;<\/span><br><span> &#125;<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u6253\u5305\u52a0\u8f7d\" title=\"\u6253\u5305\u52a0\u8f7d\"><\/a>\u6253\u5305\u52a0\u8f7d<\/h4>\n<p>\u5bf9\u4e0a\u8ff0\u547d\u4ee4\u6e90\u6587\u4ef6\u6253\u5305\u4e3audf.jar,\u62f7\u8d1d\u5230\u670d\u52a1\u5668\u7684\/hivedata\/\u76ee\u5f55<br \/>\u5728Hive\u5ba2\u6237\u7aef\u628audf.jar\u52a0\u5165\u5230hive\u4e2d,\u5982\u4e0b:<\/p>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span>add jar &#x2F;hivedata&#x2F;udf.jar;<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u521b\u5efa\u4e34\u65f6\u51fd\u6570\" title=\"\u521b\u5efa\u4e34\u65f6\u51fd\u6570\"><\/a>\u521b\u5efa\u4e34\u65f6\u51fd\u6570<\/h4>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><span>2<\/span><br><span>3<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span>create temporary function parseMap as &#39;com.qf.hive.ParseMapUDTF&#39;; # \u521b\u5efa\u4e00\u4e2a\u4e34\u65f6\u51fd\u6570parseMap<\/span><br><span><\/span><br><span>show functions ;<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u6d4b\u8bd5\u4e34\u65f6\u51fd\u6570\" title=\"\u6d4b\u8bd5\u4e34\u65f6\u51fd\u6570\"><\/a>\u6d4b\u8bd5\u4e34\u65f6\u51fd\u6570<\/h4>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><span>2<\/span><br><span>3<\/span><br><span>4<\/span><br><span>5<\/span><br><span>6<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span>select parseMap(&quot;name:zhang;age:30;address:shenzhen&quot;)<\/span><br><span><\/span><br><span>#map  key  <\/span><br><span>name    zhang<\/span><br><span>age 30<\/span><br><span>address shenzhen<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#UDAF\u683c\u5f0f\" title=\"UDAF\u683c\u5f0f\"><\/a>UDAF\u683c\u5f0f<\/h3>\n<p>\u7528\u6237\u81ea\u5b9a\u4e49\u805a\u5408\u51fd\u6570\u3002user defined aggregate function\u3002\u591a\u5bf9\u4e00\u7684\u8f93\u5165\u8f93\u51fa count sum max\u3002\u5b9a\u4e49\u4e00\u4e2aUDAF\u9700\u8981\u5982\u4e0b\u6b65\u9aa4:<\/p>\n<ol>\n<li>UDF\u81ea\u5b9a\u4e49\u51fd\u6570\u5fc5\u987b\u662forg.apache.hadoop.hive.ql.exec.UDAF\u7684\u5b50\u7c7b,\u5e76\u4e14\u5305\u542b\u4e00\u4e2a\u6216\u591a\u4e2a\u4e2a\u5d4c\u5957\u7684\u7684\u5b9e\u73b0\u4e86org.apache.hadoop.hive.ql.exec.UDAFEvaluator\u7684\u9759\u6001\u7c7b\u3002<\/li>\n<li>\u51fd\u6570\u7c7b\u9700\u8981\u7ee7\u627fUDAF\u7c7b\uff0c\u5185\u90e8\u7c7bEvaluator\u5b9eUDAFEvaluator\u63a5\u53e3\u3002<\/li>\n<li>Evaluator\u9700\u8981\u5b9e\u73b0 init\u3001iterate\u3001terminatePartial\u3001merge\u3001terminate\u8fd9\u51e0\u4e2a\u51fd<\/li>\n<\/ol>\n<p>\u8fd9\u51e0\u4e2a\u51fd\u6570\u4f5c\u7528\u5982\u4e0b:<\/p>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u51fd\u6570\u8bf4\u660e\" title=\"\u51fd\u6570\u8bf4\u660e\"><\/a>\u51fd\u6570\u8bf4\u660e<\/h4>\n<p>init\u5b9e\u73b0\u63a5\u53e3UDAFEvaluator\u7684init\u51fd\u6570<br \/>iterate\u6bcf\u6b21\u5bf9\u4e00\u4e2a\u65b0\u503c\u8fdb\u884c\u805a\u96c6\u8ba1\u7b97\u90fd\u4f1a\u8c03\u7528,\u8ba1\u7b97\u51fd\u6570\u8981\u6839\u636e\u8ba1\u7b97\u7684\u7ed3\u679c\u66f4\u65b0\u5176\u5185\u90e8\u72b6\u6001terminatePartial\u65e0\u53c2\u6570\uff0c\u5176\u4e3aiterate\u51fd\u6570\u8f6e\u8f6c\u7ed3\u675f\u540e\uff0c\u8fd4\u56de\u8f6e\u8f6c\u6570\u636emerge\u63a5\u6536terminatePartial\u7684\u8fd4\u56de\u7ed3\u679c\uff0c\u8fdb\u884c\u6570\u636emerge\u64cd\u4f5c\uff0c\u5176\u8fd4\u56de\u7c7b\u578b\u4e3aboolean\u3002terminate\u8fd4\u56de\u6700\u7ec8\u7684\u805a\u96c6\u51fd\u6570\u7ed3\u679c\u3002<\/p>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u81ea\u5b9a\u4e49\u51fd\u6570\u6848\u4f8b-2\" title=\"\u81ea\u5b9a\u4e49\u51fd\u6570\u6848\u4f8b\"><\/a>\u81ea\u5b9a\u4e49\u51fd\u6570\u6848\u4f8b<\/h4>\n<p>\u8ba1\u7b97\u4e00\u7ec4\u6574\u6570\u7684\u6700\u5927\u503c<\/p>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u4ee3\u7801-1\" title=\"\u4ee3\u7801*\"><\/a>\u4ee3\u7801*<\/h4>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><span>2<\/span><br><span>3<\/span><br><span>4<\/span><br><span>5<\/span><br><span>6<\/span><br><span>7<\/span><br><span>8<\/span><br><span>9<\/span><br><span>10<\/span><br><span>11<\/span><br><span>12<\/span><br><span>13<\/span><br><span>14<\/span><br><span>15<\/span><br><span>16<\/span><br><span>17<\/span><br><span>18<\/span><br><span>19<\/span><br><span>20<\/span><br><span>21<\/span><br><span>22<\/span><br><span>23<\/span><br><span>24<\/span><br><span>25<\/span><br><span>26<\/span><br><span>27<\/span><br><span>28<\/span><br><span>29<\/span><br><span>30<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span><span>package<\/span> com.qf.hive;<\/span><br><span>\u200b<\/span><br><span><span>public<\/span> <span><span>class<\/span> <span>MaxValueUDAF<\/span> <span>extends<\/span> <span>UDAF<\/span> <\/span>&#123;<\/span><br><span>    <span>public<\/span> <span>static<\/span> <span><span>class<\/span> <span>MaximumIntUDAFEvaluator<\/span> <span>implements<\/span> <span>UDAFEvaluator<\/span> <\/span>&#123;<\/span><br><span>        <span>private<\/span> IntWritable result;<\/span><br><span>        <span><span>public<\/span> <span>void<\/span> <span>init<\/span><span>()<\/span> <\/span>&#123;<\/span><br><span>            result = <span>null<\/span>;<\/span><br><span>        &#125;<\/span><br><span>        <span><span>public<\/span> <span>boolean<\/span> <span>iterate<\/span><span>(IntWritable value)<\/span> <\/span>&#123;<\/span><br><span>            <span>if<\/span> (value == <span>null<\/span>) &#123;<\/span><br><span>                <span>return<\/span> <span>true<\/span>;<\/span><br><span>            &#125;<\/span><br><span>            <span>if<\/span> (result == <span>null<\/span>) &#123;<\/span><br><span>                result = <span>new<\/span> IntWritable( value.get() );<\/span><br><span>            &#125; <span>else<\/span> &#123;<\/span><br><span>                result.set( Math.max( result.get(), value.get() ) );<\/span><br><span>            &#125;<\/span><br><span>            <span>return<\/span> <span>true<\/span>;<\/span><br><span>        &#125;<\/span><br><span>        <span><span>public<\/span> IntWritable <span>terminatePartial<\/span><span>()<\/span> <\/span>&#123;<\/span><br><span>            <span>return<\/span> result;<\/span><br><span>        &#125;<\/span><br><span>        <span><span>public<\/span> <span>boolean<\/span> <span>merge<\/span><span>(IntWritable other)<\/span> <\/span>&#123;<\/span><br><span>            <span>return<\/span> iterate( other );<\/span><br><span>        &#125;<\/span><br><span>        <span><span>public<\/span> IntWritable <span>terminate<\/span><span>()<\/span> <\/span>&#123;<\/span><br><span>            <span>return<\/span> result;<\/span><br><span>        &#125;<\/span><br><span>    &#125;<\/span><br><span>&#125;<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u6253\u5305\u52a0\u8f7d-1\" title=\"\u6253\u5305\u52a0\u8f7d\"><\/a>\u6253\u5305\u52a0\u8f7d<\/h4>\n<p>\u5bf9\u4e0a\u8ff0\u547d\u4ee4\u6e90\u6587\u4ef6\u6253\u5305\u4e3audf.jar,\u62f7\u8d1d\u5230\u670d\u52a1\u5668\u7684\/hivedata\/\u76ee\u5f55<br \/>\u5728Hive\u5ba2\u6237\u7aef\u628audf.jar\u52a0\u5165\u5230hive\u4e2d,\u5982\u4e0b:<\/p>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span>add jar &#x2F;hivedata&#x2F;udf.jar;<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h4><a href=\"https:\/\/www.studytime.xin\/article\/#\u521b\u5efa\u4e34\u65f6\u51fd\u6570-1\" title=\"\u521b\u5efa\u4e34\u65f6\u51fd\u6570\"><\/a>\u521b\u5efa\u4e34\u65f6\u51fd\u6570<\/h4>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><span>2<\/span><br><span>3<\/span><br><span>4<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span>create temporary function maxInt as &#39;com.qf.hive.MaxValueUDAF&#39;;<\/span><br><span>\u200b<\/span><br><span># \u67e5\u770b\u51fd\u6570\u662f\u5426\u52a0\u5165<\/span><br><span>show functions ;<\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n<h3><a href=\"https:\/\/www.studytime.xin\/article\/#\u6d4b\u8bd5\u4e34\u65f6\u51fd\u6570-1\" title=\"\u6d4b\u8bd5\u4e34\u65f6\u51fd\u6570\"><\/a>\u6d4b\u8bd5\u4e34\u65f6\u51fd\u6570<\/h3>\n<figure>\n<table>\n<tr>\n<td>\n<pre><span>1<\/span><br><span>2<\/span><br><span>3<\/span><br><\/pre>\n<\/td>\n<td>\n<pre><span>select maxInt(mgr) from emp<\/span><br><span><\/span><br><span><span>7902<\/span><\/span><br><\/pre>\n<\/td>\n<\/tr>\n<\/table>\n<\/figure>\n","protected":false},"excerpt":{"rendered":"<p>hive\u4e3a\u4ec0\u4e48\u9700\u8981\u81ea\u5b9a\u4e49\u51fd\u6570 hi&hellip;<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[],"tags":[],"_links":{"self":[{"href":"http:\/\/4563.org\/index.php?rest_route=\/wp\/v2\/posts\/193075"}],"collection":[{"href":"http:\/\/4563.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/4563.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/4563.org\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/4563.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=193075"}],"version-history":[{"count":0,"href":"http:\/\/4563.org\/index.php?rest_route=\/wp\/v2\/posts\/193075\/revisions"}],"wp:attachment":[{"href":"http:\/\/4563.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=193075"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/4563.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=193075"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/4563.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=193075"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}