我正在搜索可以在Oracle数据库中查找行的正则表达式。行包含一个如下所示的字符串:
<?xml version="1.0" encoding="ISO-8859-1" standalone="no"?>
<characteristics>
<characteristic id="1001" mandatory="true" seq="1">
<wildcard/>
</characteristic>
<characteristic id="10001" mandatory="false" seq="1">
<value negation="false" seq="1">63</value>
</characteristic>
<characteristic id="10002" mandatory="false" seq="1">
<value negation="false" seq="1">64</value>
</characteristic>
<characteristic id="10051" mandatory="false" seq="1">
<value negation="false" seq="1">65</value>
</characteristic>
<characteristic id="10052" mandatory="false" seq="1">
<value negation="false" seq="1">66</value>
</characteristic>
<characteristic id="1010" mandatory="false" seq="100">
<wildcard/>
<value negation="false" seq="1">314</value>
</characteristic>
</characteristics>
正则表达式必须查找包含具有包含通配符的特殊ID的特征的所有行:
<wildcard/>
例如:我想获取所有行,其中包含带有通配符的ID为1000和1001的特征。这意味着必须有两个字符串
<characteristic id="1001" ...
<wildcard/>
...
</characteristic>
和
<characteristic id="1000" ...
<wildcard/>
...
</characteristic>
在里面。因此,这两个字符串可以有不同的顺序。
答案 0 :(得分:2)
假设数据存储为XMLtype
select *
from t
where EXISTSNODE(doc,'/characteristics/characteristic[@id="1000"]/wildcard') = 1
and EXISTSNODE(doc,'/characteristics/characteristic[@id="1001"]/wildcard') = 1
;
替代语法(由@Tomalak建议)
select *
from t
where EXISTSNODE(doc,'/characteristics[characteristic[@id="1001"]/wildcard and characteristic[@id="1010"]/wildcard]') = 1
出于学习目的,如果数据存储为字符串
select *
from (select xmltype(doc) as doc from t)
where EXISTSNODE(doc,'/characteristics/characteristic[@id="1000"]/wildcard') = 1
and EXISTSNODE(doc,'/characteristics/characteristic[@id="1001"]/wildcard') = 1
;