自己动手实现mybatis动态sql的方法

发现要坚持写博客真的是一件很困难的事情，各种原因都会导致顾不上博客。本来打算写自己动手实现orm，看看时间，还是先实现一个动态sql，下次有时间再补上orm完整的实现吧。

用过mybatis的人，估计对动态sql都不陌生，如果没有用过，就当看看热闹吧。我第一次接触mysql是在大四的时候，当时就觉得动态sql这东西很牛，很灵活，一直想搞明白怎么实现的，尽管当时已经能够写ioc，mvc和简单的orm框架(仿mybaits但是没有动态sql部分)，但是仍然找不到mybatis核心的动态sql到底在哪实现的，怎么实现的，可能是那些代码太绕根本没法看懂，直到目前，我都没有勇气去看mybatis的动态sql部分，大概是天生对算法有莫名其妙的敬畏吧。

几年前因为想做一个配置平台，想用解析型语言替代java的实现，可以让配置人员在页面上方便的编写少量代码实现复杂的业务逻辑（包括数据库操作）。当时java已经有js解析引擎，但是大多数人都说效率太低，不知道我发什么疯就想到自己实现一个解析语言。不过实现自己的语言也是我一直的梦想，解析语言相对编译型语言入手简单，于是我就果断动手了，写完才知道，其实自己的实现估计还没有当时的js引擎效率高，那时的我真的是很年轻很简单。今天谈到的动态sql实现其实就是受到那时候解析语言的启发。

废话不多说直接开始聊动态sql，请看下面例子，先声明这里的例子并不是一个正确的sql的写法，只是想写一个尽量复杂的嵌套结构，如果把这种复杂的情况实现了，那么简单一点的就更加不在话下了。

delete from pl_pagewidget
<if test="widgetcodes != null">
 where pagewidgetcode in
 <foreach collection="widgetcodes" item="item" index="index" open="(" separator="," close=")">
 <if test="index == 0">
 #{item}
 </if>
 <foreach collection="bs" item="b" index="index1" open="(" separator="," close=")">
  #{b}
 </foreach>
 </foreach>
</if>
<if test="a != null">
 and a = #{a}
</if>

要实现解析出上面例子的sql，首先一个难点类似是test属性里的条件怎么判断真假，不过这个难点在struts2中学到的ognl表达式面前就比较小儿科了。不知道有么有朋友遇到过一个比较奇葩的现象，就是有时候明明在mybatis动态sql中写如下表达式，但是当n=0的时候居然是满足条件的也就是test里的值是false，0居然不能满足这个表达式的条件，这里就是ognl库的原因了。没办法它就是这么玩的，当成特殊情况记住就可以了

test="n != null and n !=''"

ognl表达式使用很方便如下

import java.util.HashMap;
import java.util.Map;
import ognl.Ognl;
public class OgnlTest {
 //输出结果:false
 public static void main(String[] args) throws Exception {
 String con1 = "n != null and n != ''";
 Map<String,Object> root = new HashMap<>();
 root.put("n", 0);
 System.out.println(Ognl.getValue(con1,root));
 }
}

要实现解析上面例子的sql，第二个难点就是虽然这个sql披上一层xml的皮就是一个标准的sql，如下

<sql>
 delete from pl_pagewidget
 <if test="widgetcodes != null">
 where pagewidgetcode in
 <foreach collection="widgetcodes" item="item" index="index" open="(" separator="," close=")">
 <if test="index == 0">
 #{item}
 </if>
 <foreach collection="bs" item="b" index="index1" open="(" separator="," close=")">
  #{b}
 </foreach>
 </foreach>
 </if>
 <if test="a != null">
 and a = #{a}
 </if>
</sql>

但是要解析上面的xml和我们平时不一样，这个xml是标签和文本混合的，正常我们开发中应该很少会用到解析这种xml。不过我们常用的解析xml的工具dom4j其实可以很好的解析这种sql，只不过很少可能用到。Element类的content()方法就可以返回一个Node的集合，再通过遍历这个集合，判断每个Node的类型就可以了。解决了这两个重点，只需要加上一点技巧就可以解析这种动态sql了。

我用到的技巧是根据java语法格式得到的启发。比如java中有局部变量和全局变量，不考虑引用传递这种情况，如果全局变量int i = 1；方法里面传入这个全局变量，然后在方法里面修改，在方法里面看到的是改变后的值，但是在方法外面看到的仍然是1。这个现象其实学过java应该都知道。还有就是当方法调用的时候，方法里面可以看到全局变量，也可以看到局部变量，方法调用结束后局部变量会被清空释放(看垃圾搜集器高兴)。介绍了这些直接上代码了

import java.io.StringReader;
import java.text.SimpleDateFormat;
import java.util.Arrays;
import java.util.Date;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
import org.apache.commons.collections.MapUtils;
import org.apache.commons.lang.StringUtils;
import org.dom4j.Document;
import org.dom4j.Element;
import org.dom4j.Node;
import org.dom4j.Text;
import org.dom4j.io.SAXReader;
import com.rd.sql.Attrs;
import com.rd.sql.BaseNode;
import com.rd.sql.NodeFactory;
public class SqlParser {
 private Map<String,Object> currParams = new HashMap<String,Object>();

 /**
 delete from pl_pagewidget
 <if test="widgetcodes != null">
  where pagewidgetcode in
  <foreach collection="widgetcodes" item="item" index="index" open="(" separator="," close=")">
  <if test="index == 0">
  #{item}
  </if>
  <foreach collection="bs" item="b" index="index1" open="(" separator="," close=")">
  #{b}
  </foreach>
  </foreach>
 </if>
 <if test="a != null">
  and a = #{a}
 </if>
 */
 public static void main(String[] args) throws Exception {
 Map<String, Object> map = new HashMap<String, Object>();
 map.put("widgetcodes", Arrays.asList("1", "2"));
 map.put("bs", Arrays.asList("3", "4"));
 map.put("a", 1);
 SqlParser parser = new SqlParser();
 System.out
  .println(parser.parser("delete from pl_pagewidget\n"
    + "\t<if test=\"widgetcodes != null\">\n"
    + "\t\twhere pagewidgetcode in\n"
    + "\t\t<foreach collection=\"widgetcodes\" item=\"item\" index=\"index\" open=\"(\" separator=\",\" close=\")\">\n"
    + "\t\t <if test=\"index == 0\">\n"
    + "\t\t #{item}\n"
    + "\t\t </if>\n"
    + "\t\t <foreach collection=\"bs\" item=\"b\" index=\"index1\" open=\"(\" separator=\",\" close=\")\">\n"
    + "\t\t\t#{b}\n" + "\t\t </foreach>\n"
    + "\t\t</foreach>\n" + "\t</if>\n"
    + "\t<if test=\"a != null\">\n"
    + "\t\tand a = #{a}\n" + "\t</if>\n", map));
 System.out.println(parser.getParams());
 }
 public String parser(String xml, Map<String, Object> params)
  throws Exception {
 // xml = "<?xml version=\"1.0\" encoding=\"UTF-8\"?>"+xml;
 //给输入的动态sql套一层xml标签
 xml = "<sql>"+xml+"</sql>";
 SAXReader reader = new SAXReader(false);
 Document document = reader.read(new StringReader(xml));
 Element element = document.getRootElement();
 Map<String, Object> currParams = new HashMap<String, Object>();
 StringBuilder sb = new StringBuilder();
 //开始解析
 parserElement(element, currParams, params, sb);
 return sb.toString();
 }
 /**
 * 使用递归解析动态sql
 * @param ele1 待解析的xml标签
 * @param currParams
 * @param globalParams
 * @param sb
 * @throws Exception
 */
 private void parserElement(Element ele1, Map<String, Object> currParams,
  Map<String, Object> globalParams, StringBuilder sb)
  throws Exception {
 // 解析一个节点，比如解析到了一个if节点，假如test判断为true这里就返回true
 TempVal val = parserOneElement(currParams, globalParams, ele1, sb);
 //得到解析的这个节点的抽象节点对象
 BaseNode node = val.getNode();
 /**
  * 实际上这句之上的语句只是解析了xml的标签，并没有解析标签里的内容，这里
  * 表示要解析内容之前，如果有前置操作做一点前置操作
  */
 node.pre(currParams, globalParams, ele1, sb);
 //判断是否还需要解析节点里的内容，例如if节点test结果为true
 boolean flag = val.isContinue();
 // 得到该节点下的所有子节点的集合，包含普通文本
 List<Node> nodes = ele1.content();
 if (flag && !nodes.isEmpty()) {
  /**
  * 这里表示要进一步解析节点里的内容了，可以把节点类比成一个方法的外壳
  * 里面的内容类比成方法里的具体语句，开始解析节点的内容之前
  * 先创建本节点下的局部参数的容器，最方便当然是map
  */
  Map<String, Object> params = new HashMap<String, Object>();
  /**
  * 把外面传进来的局部参数，直接放入容器，由于本例中参数都是常用数据类型
  * 不会存在引用类型所以，这里相当于是一个copy，为了不影响外面传入的对象
  * 可以类比方法调用传入参数的情况
  */
  params.putAll(currParams);
  //循环所有子节点
  for (int i = 0; i < nodes.size();) {
  Node n = nodes.get(i);
  //如果节点是普通文本
  if (n instanceof Text) {
   String text = ((Text) n).getStringValue();
   if (StringUtils.isNotEmpty(text.trim())) {
   //处理一下文本，如处理#{xx},直接替换${yy}为真实传入的值
   sb.append(handText(text, params,globalParams));
   }
   i++;
  } else if (n instanceof Element) {
   Element e1 = (Element) n;
   // 递归解析xml子元素
   parserElement(e1, params, globalParams, sb);
   // 如果循环标志不为true则解析下一个标签
   // 这里表示需要重复解析这个循环标签，则i不变,反之继续处理下一个元素
   boolean while_flag = MapUtils.getBoolean(params,
    Attrs.WHILE_FLAG, false);
   if (!while_flag
    || !NodeFactory.isWhile(n.getName())
    || e1.attributeValue(Attrs.INDEX) == null
    || !e1.attributeValue(Attrs.INDEX).equals(
     params.get(Attrs.WHILE_INDEX))) {
   i++;
   }
  }
  }
  //节点处理之后做一些啥事
  node.after(currParams, globalParams, ele1, sb);
  // 回收当前作用域参数
  params.clear();
  params = null;
 }
 }
 /**
 * 处理文本替换掉#{item}这种参数
 * @param str
 * @param params
 * @return
 * @throws Exception
 */
 private String handText(String str, Map<String, Object> params,Map<String, Object> globalParams)
  throws Exception {
 //获取foreach这种标签中用于记录循环的变量
 String indexStr = MapUtils.getString(params, Attrs.WHILE_INDEX);
 Integer index = null;
 if(StringUtils.isNotEmpty(indexStr)) {
  index = MapUtils.getInteger(params, indexStr);
 }
 //匹配#{a}这种参数
 String reg1 = "(#\\{)(\\w+)(\\})";
 //匹配${a}这种参数
 String reg2 = "(\\$\\{)(\\w+)(\\})";
 Pattern p1 = Pattern.compile(reg1);
 Matcher m1 = p1.matcher(str);
 Pattern p2 = Pattern.compile(reg2);
 Matcher m2 = p2.matcher(str);
 String whileList = MapUtils.getString(params, Attrs.WHILE_LIST);
 Map<String,Object> allParams = getAllParams(params, globalParams);
 while(m1.find()) {
  String tmpKey = m1.group(2);
  String key = whileList == null?tmpKey:(whileList+"_"+tmpKey);
  key = index == null?key:(key+index);
  String reKey = "#{"+key+"}";
  //如果在foreach类似的循环里，可能需要将参数#{xx}替换成#{xx_0},#{xx_1}
  str = str.replace(m1.group(0), reKey);
  currParams.put(key, allParams.get(tmpKey));
 }
 while(m2.find()) {
  String tmpKey = m2.group(2);
  Object value = allParams.get(tmpKey);
  if(value != null) {
  str = str.replace(m2.group(0), getValue(value));
  }
 }
 return str;
 }
 private String getValue(Object value) {
 String result = "";
 if(value instanceof Date) {
  SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");
  result = sdf.format((Date)value);
 } else {
  result = String.valueOf(value);
 }
 return result;
 }
 private Map<String, Object> getAllParams(Map<String, Object> currParams,
  Map<String, Object> globalParams) {
 Map<String,Object> allParams = new HashMap<String,Object>();
 allParams.putAll(globalParams);
 allParams.putAll(currParams);
 return allParams;
 }

 // 解析一个xml元素
 private TempVal parserOneElement(Map<String, Object> currParams,
  Map<String, Object> globalParams, Element ele, StringBuilder sb)
  throws Exception {
 //获取xml标签名
 String eleName = ele.getName();
 //解析一个节点后是否继续，如遇到if这种节点，就需要判断test里是否为空
 boolean isContinue = false;
 //声明一个抽象节点
 BaseNode node = null;
 if (StringUtils.isNotEmpty(eleName)) {
  //使用节点工厂根据节点名得到一个节点对象比如是if节点还是foreach节点
  node = NodeFactory.create(eleName);
  //解析一下这个节点，返回是否还需要解析节点里的内容
  isContinue = node.parse(currParams, globalParams, ele, sb);
 }
 return new TempVal(isContinue, ele, node);
 }
 public Map<String, Object> getParams() {
 return currParams;
 }
 /**
 * 封装一个xml元素被解析后的结果
 * @author rongdi
 */
 final static class TempVal {
 private boolean isContinue;
 private Element ele;
 private BaseNode node;
 public TempVal(boolean isContinue, Element ele, BaseNode node) {
  this.isContinue = isContinue;
  this.ele = ele;
  this.node = node;
 }
 public boolean isContinue() {
  return isContinue;
 }
 public void setContinue(boolean isContinue) {
  this.isContinue = isContinue;
 }
 public Element getEle() {
  return ele;
 }
 public void setEle(Element ele) {
  this.ele = ele;
 }
 public BaseNode getNode() {
  return node;
 }
 public void setNode(BaseNode node) {
  this.node = node;
 }
 }
}

import org.dom4j.Element;
import java.util.HashMap;
import java.util.Map;
/**
 * 抽象节点
 * @author rongdi
 */
public abstract class BaseNode {
 public abstract boolean parse(Map<String, Object> currParams, Map<String, Object> globalParams, Element ele,StringBuilder sb) throws Exception;
 public void pre(Map<String, Object> currParams,Map<String, Object> globalParams,Element ele,StringBuilder sb) throws Exception {
 }
 public void after(Map<String, Object> currParams,Map<String, Object> globalParams,Element ele,StringBuilder sb) throws Exception {
 }
 protected Map<String, Object> getAllParams(Map<String, Object> currParams,
      Map<String, Object> globalParams) {
 Map<String,Object> allParams = new HashMap<String,Object>();
 allParams.putAll(globalParams);
 allParams.putAll(currParams);
 return allParams;
 }
}

import java.util.Map;
import ognl.Ognl;
import org.apache.commons.lang.StringUtils;
import org.dom4j.Element;
/**
 * if节点
 * @author rongdi
 */
public class IfNode extends BaseNode{
 @Override
 public boolean parse(Map<String, Object> currParams, Map<String, Object> globalParams, Element ele,StringBuilder sb) throws Exception {
 //得到if节点的test属性
 String testStr = ele.attributeValue("test");
 boolean test = false;
 try {
  if(StringUtils.isNotEmpty(testStr)) {
  //合并全局变量和局部变量
  Map<String, Object> allParams = getAllParams(currParams,globalParams);
  //使用ognl判断true或者false
  test = (Boolean) Ognl.getValue(testStr,allParams);
  }
 } catch (Exception e) {
  e.printStackTrace();
  throw new Exception("判断操作参数"+testStr+"不合法");
 }
 if(ele.content() != null && ele.content().size()==0) {
  test = true;
 }
 return test;
 }
}

import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.Set;
import ognl.Ognl;
import org.apache.commons.collections.MapUtils;
import org.apache.commons.lang.StringUtils;
import org.dom4j.Element;
/**
 foreach节点属性如下
 collection 需要遍历的集合
 item 遍历集合后每个元素存放的变量
 index 遍历集合的索引数如0,1,2...
 separator 遍历后以指定分隔符拼接
 open 遍历后拼接开始的符号如 (
 close 遍历后拼接结束的符号如 )
 */
public class ForeachNode extends BaseNode {
 @Override
 public boolean parse(Map<String, Object> currParams, Map<String, Object> globalParams, Element ele, StringBuilder sb) throws Exception {
 String conditionStr = null;
 String collectionStr = ele.attributeValue("collection");
 String itemStr = ele.attributeValue("item");
 String index = ele.attributeValue("index");
 String separatorStr = ele.attributeValue("separator");
 String openStr = ele.attributeValue("open");
 String closeStr = ele.attributeValue("close");
 if(StringUtils.isEmpty(index)) {
  index = "index";
 }
 if(StringUtils.isEmpty(separatorStr)) {
  separatorStr = ",";
 }
 if(StringUtils.isNotEmpty(openStr)) {
  currParams.put(Attrs.WHILE_OPEN,openStr);
 }
 if(StringUtils.isNotEmpty(closeStr)) {
  currParams.put(Attrs.WHILE_CLOSE,closeStr);
 }
 if(StringUtils.isNotEmpty(collectionStr)) {
  currParams.put(Attrs.WHILE_LIST,collectionStr);
 }
 currParams.put(Attrs.WHILE_SEPARATOR,separatorStr);
 if(index != null) {
  /**
  * 如果局部变量中存在当前循环变量的值，就表示已经不是第一次进入循环标签了，移除掉开始标记
  * 并将局部变量值加1
  */
  if(currParams.get(index) != null) {
  currParams.remove(Attrs.WHILE_START);
  currParams.put(index+"_", (Integer)currParams.get(index+"_") + 1);
  } else { //第一次进入循环标签内
  currParams.put(Attrs.WHILE_START,true);
  currParams.put(index+"_", 0);
  }
  currParams.put(index, (Integer)currParams.get(index+"_"));
 }
 boolean condition = true;
 Map<String, Object> allParams = getAllParams(currParams,globalParams);
 Object collection = null;
 if(StringUtils.isNotEmpty(collectionStr)) {
  //得到待循环的集合
  collection = Ognl.getValue(collectionStr,allParams);
  //如果集合属性不为空，但是条件为null则默认加上一个边界条件
  if(StringUtils.isEmpty(conditionStr)) {
  //这里只是用集合演示一下，也可以再加上数组，只不过改成.length而已
  if(collection instanceof List) {
   conditionStr = index+"_<"+collectionStr+".size()";
  } else if(collection instanceof Map){
   Map map = (Map)collection;
   Set set = map.entrySet();
   List list = new ArrayList(set);
   allParams.put("_list_", list);
   conditionStr = index+"_<_list_"+".size()";
  }
  }
 }
 currParams.remove(Attrs.WHILE_END);
 if(StringUtils.isNotEmpty(conditionStr)) {
  //计算条件的值
  condition = (Boolean)Ognl.getValue(conditionStr,allParams);
  Map<String,Object> tempMap = new HashMap<>();
  tempMap.putAll(allParams);
  tempMap.put(index+"_",(Integer)currParams.get(index+"_") + 1);
  currParams.put(Attrs.WHILE_END,!(Boolean)Ognl.getValue(conditionStr,tempMap));
 }
 boolean flag = true;
 currParams.put(Attrs.WHILE_INDEX, index);
 currParams.put(Attrs.WHILE_FLAG, true);
 if(condition) {
  try {
  if(StringUtils.isNotEmpty(itemStr) && StringUtils.isNotEmpty(collectionStr)) {
   Object value = null;
   int idx = Integer.parseInt(currParams.get(index+"_").toString());
   if(collection instanceof List) {
   value = ((List)collection).get(idx);
   currParams.put(itemStr, value);
   } else if(collection instanceof Map){
   Map map = (Map)collection;
   Set<Map.Entry<String,Object>> set = map.entrySet();
   List<Map.Entry<String,Object>> list = new ArrayList(set);
   currParams.put(itemStr, list.get(idx).getValue());
   currParams.put(index, list.get(idx).getKey());
   }
  }
  } catch (Exception e) {
  throw new Exception("从集合或者映射取值"+currParams.get(index)+"错误"+e.getMessage());
  }
 } else {
  flag = false;
  destroyVars(currParams, index, itemStr);
 }
 return flag;
 }
 /**
 * 如果是第一次进入循环标签，则拼上open的内容
 */
 @Override
 public void pre(Map<String, Object> currParams, Map<String, Object> globalParams, Element ele, StringBuilder sb) throws Exception {
 super.pre(currParams, globalParams, ele, sb);
 boolean start = MapUtils.getBoolean(currParams,Attrs.WHILE_START,false);
 if(start) {
  String open = MapUtils.getString(currParams,Attrs.WHILE_OPEN);
  sb.append(open);
 }
 }
 /**
 * 如果是最后进入循环标签，则最后拼上close的内容
 */
 @Override
 public void after(Map<String, Object> currParams, Map<String, Object> globalParams, Element ele, StringBuilder sb) throws Exception {
 super.after(currParams, globalParams, ele, sb);
 boolean end = MapUtils.getBoolean(currParams,Attrs.WHILE_END,false);
 String separator = MapUtils.getString(currParams,Attrs.WHILE_SEPARATOR);
 if(!end && StringUtils.isNotEmpty(separator)) {
  sb.append(separator);
 }
 if(end) {
  String close = MapUtils.getString(currParams,Attrs.WHILE_CLOSE);
  if(sb.toString().endsWith(separator)) {
  sb.deleteCharAt(sb.length() - 1);
  }
  sb.append(close);
 }
 }
 //释放临时变量
 private void destroyVars(Map<String, Object> currParams, String index,String varStr) {
 currParams.remove(Attrs.WHILE_INDEX);
 currParams.remove(Attrs.WHILE_FLAG);
 currParams.remove(Attrs.WHILE_SEPARATOR);
 currParams.remove(Attrs.WHILE_START);
 currParams.remove(Attrs.WHILE_END);
 currParams.remove(Attrs.WHILE_LIST);
　　　　}
}

import org.dom4j.Element;
import java.util.Map;
public class SqlNode extends BaseNode{
 @Override
 public boolean parse(Map<String, Object> currParams, Map<String, Object> globalParams, Element ele,StringBuilder sb) throws Exception {
 return true;
 }
}

import java.util.Arrays;
import java.util.List;
import java.util.Map;
import java.util.concurrent.ConcurrentHashMap;
/**
 * 节点工厂
 */
public class NodeFactory {

 private static Map<String,BaseNode> nodeMap = new ConcurrentHashMap<String,BaseNode>();
 private final static List<String> whileList = Arrays.asList("foreach");
 static {
 nodeMap.put("if", new IfNode());
 nodeMap.put("sql", new SqlNode());
 nodeMap.put("foreach", new ForeachNode());
 }
 public static boolean isWhile(String elementName) {
 return whileList.contains(elementName);
 }
 public static void addNode(String nodeName,BaseNode node) {

 nodeMap.put(nodeName, node);

 }

 public static BaseNode create(String nodeName) {

 return nodeMap.get(nodeName);

 }

}

/**
 * 各种标记
 * @author rongdi
 */
public class Attrs {

 public final static String TRANSACTIONAL = "transactional";
 public final static String WHILE_START = "while-start";
 public final static String WHILE_END = "while-end";
 public final static String WHILE_OPEN = "while-open";
 public final static String WHILE_CLOSE = "while-close";
 public final static String WHILE_SEPARATOR = "while-separator";
 public final static String WHILE_INDEX = "while-index";

 public final static String WHILE_FLAG = "while-flag";

 public final static String WHILE_LIST = "while-list";

 public final static String WHEN_FLAG = "when-flag";

 public static final String PROCESS_VAR = "process-var";

 public final static String RESULT_FLAG = "result-flag";

 public final static String RETURN_FLAG = "return-flag";

 public final static String CONSOLE_VAR= "console-var";

 public final static String DO = "do";

 public final static String INDEX = "index";

 public final static String CONDITION = "condition";

 public final static String NAME= "name";

 public final static String VALUE= "value";
 public static final String TYPE = "type";
 public static final String FORMAT = "format";

 public static final String IF = "if";

 public static final String ELSE = "else";

 public final static String FILE= "file";

 public static final String DATE = "date";

 public static final String NOW = "now";

 public static final String DECIMAL = "decimal";

 public static final String ID = "id";

 public static final String PARAMS = "params";
 public static final String TARGET = "target";
 public static final String SINGLE = "single";

 public static final String PAGING = "paging";
 public static final String DESC = "desc";
 public static final String BREAK = "break";

 public static final String CONTINUE = "continue";

 public static final String COLLECTION = "collection";
 public static final String VAR = "var";

 public static final String EXECUTOR = "executor-1";

 public static final String ROLLBACK_FLAG = "rollback-flag";

 public static final String SERVICE = "service";

 public static final String REF = "ref";
 public static final String BIZS = "bizs";
 public static final String TITLES = "titles";

 public static final String COLUMNS = "columns";

 public static final String CURRUSER = "currUser";

 public static final String CURRPERM = "currPerm";

 public static final String TASK_EXECUTOR = "taskExecutor";
 public static final String DELIMITER = "delimiter";
 public static final String OPERNAME = "operName";

}

 currParams.remove(varStr);
 currParams.remove(index);
 currParams.remove(index+"_");
 }
}

附上pom文件

<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
 xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
 <modelVersion>4.0.0</modelVersion>
 <groupId>com.rd</groupId>
 <artifactId>parser</artifactId>
 <packaging>jar</packaging>
 <version>1.0-SNAPSHOT</version>
 <name>myparser</name>
 <url>http://maven.apache.org</url>
 <dependencies>
 <dependency>
  <groupId>dom4j</groupId>
  <artifactId>dom4j</artifactId>
  <version>1.6.1</version>
 </dependency>
 <dependency>
  <groupId>opensymphony</groupId>
  <artifactId>ognl</artifactId>
  <version>2.6.11</version>
 </dependency>
 <dependency>
  <groupId>commons-collections</groupId>
  <artifactId>commons-collections</artifactId>
  <version>3.2.1</version>
 </dependency>
 <dependency>
  <groupId>commons-lang</groupId>
  <artifactId>commons-lang</artifactId>
  <version>2.6</version>
 </dependency>
 <dependency>
  <groupId>junit</groupId>
  <artifactId>junit</artifactId>
  <version>3.8.1</version>
  <scope>test</scope>
 </dependency>
 </dependencies>
 <build>
 <resources>
  <resource>
  <directory>src/main/java</directory>
  <includes>
   <include>**/*.xml</include>
  </includes>
  </resource>
  <resource>
  <directory>src/main/resources</directory>
  <includes>
   <include>**/*</include>
  </includes>
  </resource>
 </resources>
 <testResources>
  <testResource>
  <directory>${project.basedir}/src/test/java</directory>
  </testResource>
  <testResource>
  <directory>${project.basedir}/src/test/resources</directory>
  </testResource>
 </testResources>
 <plugins>
  <plugin>
  <groupId>org.apache.maven.plugins</groupId>
  <artifactId>maven-compiler-plugin</artifactId>
  <version>3.1</version>
  <configuration>
   <source>1.8</source>
   <target>1.8</target>
   <encoding>UTF-8</encoding>
  </configuration>
  </plugin>
 </plugins>
 </build>
</project>

以上这篇自己动手实现mybatis动态sql的方法就是小编分享给大家的全部内容了，希望能给大家一个参考，也希望大家多多支持我们。