hive 血缘

🧙 Questions

• 依赖导入
• 解析sql血缘
• 解析sql

☄️ Ideas

依赖导入

<dependencies>
    <dependency>
        <groupId>org.apache.hive</groupId>
        <artifactId>hive-exec</artifactId>
        <version>3.1.2</version>
    </dependency>
</dependencies>

解析sql血缘

package com.example.demo;

import lombok.SneakyThrows;
import org.apache.hadoop.hive.ql.tools.LineageInfo;
import org.springframework.stereotype.Service;

import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;

@Service
public class HiveSqlService {

    @SneakyThrows
    public void parseHiveLineageInfo(String sql) {

        LineageInfo lineageInfo = new LineageInfo();
        lineageInfo.getLineageInfo(sql);

        // 输出表
        List<String> outTables = new ArrayList<>();
        lineageInfo.getOutputTableList().forEach(e -> {
            if (!outTables.contains(e)) {
                outTables.add(e);
            }
        });

        System.out.println("outputTables:" + Arrays.toString(outTables.toArray()));

        // 输入表
        List<String> inputTables = new ArrayList<>();
        lineageInfo.getInputTableList().forEach(e -> {
            if (!inputTables.contains(e)) {
                inputTables.add(e);
            }
        });

        System.out.println("inputTables:" + Arrays.toString(inputTables.toArray()));
    }
}

解析sql

package com.example.demo;

import lombok.SneakyThrows;
import org.apache.hadoop.hive.ql.lib.*;
import org.apache.hadoop.hive.ql.parse.*;
import org.springframework.stereotype.Service;

import java.util.ArrayList;
import java.util.LinkedHashMap;
import java.util.Map;
import java.util.Stack;

@Service
public class HiveSqlService implements NodeProcessor {

    @SneakyThrows
    public void parseHiveSql(String sql) {

        ASTNode tree = ParseUtils.parse(sql, null);
        while ((tree.getToken() == null) && (tree.getChildCount() > 0)) {
            tree = (ASTNode) tree.getChild(0);
        }

        Map<Rule, NodeProcessor> rules = new LinkedHashMap<>();
        Dispatcher dispatcher = new DefaultRuleDispatcher(this, rules, null);
        GraphWalker ogw = new DefaultGraphWalker(dispatcher);
        ArrayList<Node> topNodes = new ArrayList<>();
        topNodes.add(tree);
        ogw.startWalking(topNodes, null);
    }

    @Override
    public Object process(Node node, Stack<Node> stack, NodeProcessorCtx nodeProcessorCtx, Object... objects) throws SemanticException {
        ASTNode pt = (ASTNode) node;
        switch (pt.getToken().getType()) {

            // 修改的是哪张表
            case HiveParser.TOK_ALTERTABLE:
                String alterTableName = BaseSemanticAnalyzer.getUnescapedName((ASTNode) pt.getChild(0));
                System.out.println("修改了表:" + alterTableName);
                break;
            // 修改后的表名
            case HiveParser.TOK_ALTERTABLE_RENAME:
                String afterAlterTableName = BaseSemanticAnalyzer.getUnescapedName((ASTNode) pt.getChild(0).getChild(0));
                System.out.println("修改后的表名:" + afterAlterTableName);
                break;
            // 创建表
            case HiveParser.TOK_CREATETABLE:
                String creatTableName = BaseSemanticAnalyzer.getUnescapedName((ASTNode) pt.getChild(0).getChild(0));
                System.out.println("创建新的表名:" + creatTableName);
                break;
            // 删除表
            case HiveParser.TOK_DROPTABLE:
                String deleteTableName = BaseSemanticAnalyzer.getUnescapedName((ASTNode) pt.getChild(0).getChild(0));
                System.out.println("删除表名:" + deleteTableName);
                break;
            // 添加表字段
            case HiveParser.TOK_ALTERTABLE_ADDCOLS:
                String newCol = BaseSemanticAnalyzer.getUnescapedName((ASTNode) pt.getChild(0));
                String colType = BaseSemanticAnalyzer.getUnescapedName((ASTNode) pt.getChild(2));
                System.out.println("添加字段:" + newCol + "字段类型为:" + colType);
                break;
            // 字段重命名
            case HiveParser.TOK_ALTERTABLE_RENAMECOL:
                String prevCol = BaseSemanticAnalyzer.getUnescapedName((ASTNode) pt.getChild(0));
                String nextCol = BaseSemanticAnalyzer.getUnescapedName((ASTNode) pt.getChild(1));
                String type = BaseSemanticAnalyzer.getUnescapedName((ASTNode) pt.getChild(2));
                System.out.println("将" + prevCol + "修改为" + nextCol + "字段类型为" + type);
                break;
        }
        return null;
    }
}

🔗 Links

• hive website