sql server 上的 XML 解析

2021-10-02 00:00:00 xml sql xml-parsing sql-server

如何以格式返回数据:

Column Name t01     t02     t03     t04
Data        c01     c02     c03     c04

<orders xmlns="www address">
    <order>
        <order-date>2019-09-05</order-date>
        <created-by>storefront</created-by>
        <original-order-no>000001</original-order-no>
        <currency>USD</currency>
        <taxation>gross</taxation>
        <invoice-no>0099999</invoice-no>
        <custom-attributes>
            <custom-attribute attribute-id="t01">c01</custom-attribute>
            <custom-attribute attribute-id="t02">c02</custom-attribute>
            <custom-attribute attribute-id="t03">c03</custom-attribute>
            <custom-attribute attribute-id="t04">c04</custom-attribute>
        </custom-attributes>    
    </order>
</orders>

推荐答案

从你的问题来看，有一点不清楚:输出列的命名.

From your question there's one thing not clear: The naming of the output columns.

在您预期的输出中，它们的名称与它们的 attribute-id 一样.但在您的评论中，听起来像是您选择了前 4 个属性，而您想忽略其余属性.

In your expected output they are named like their attribute-id. But in your comments it sounds, like you are picking the first 4 attributes and you want to omit the rest.

我想展示两种方法，选择你更喜欢的一种:

I want to show two approaches, pick the one you like more:

DECLARE @mockupTable TABLE(ID INT IDENTITY, YourXml XML); INSERT INTO @mockupTable VALUES (N'<orders xmlns="www address"> <order> <order-date>2019-09-05</order-date> <created-by>storefront</created-by> <original-order-no>000001</original-order-no> <currency>USD</currency> <taxation>gross</taxation> <invoice-no>0099999</invoice-no> <custom-attributes> <custom-attribute attribute-id="t01">c01</custom-attribute> <custom-attribute attribute-id="t02">c02</custom-attribute> <custom-attribute attribute-id="t03">c03</custom-attribute> <custom-attribute attribute-id="t04">c04</custom-attribute> </custom-attributes> </order> </orders>');

--此查询将使用 attribute-id 选择相应的属性.
--我们可以使用相同的名称安全地返回它
--如果你的XML没有对应的属性，就会出现NULL值

--This query will use the attribute-id to pick the corresponding attribute.
--We can savely return this with the same name
--If your XML does not have the corresponding attribute, there will be a NULL value

WITH XMLNAMESPACES(DEFAULT 'www address') SELECT o.value('(order-date/text())[1]','date') OrderDate --As in your other questions ,o.value('(custom-attributes/custom-attribute[@attribute-id="t01"]/text())[1]','varchar(100)') AS t01 ,o.value('(custom-attributes/custom-attribute[@attribute-id="t02"]/text())[1]','varchar(100)') AS t02 ,o.value('(custom-attributes/custom-attribute[@attribute-id="t03"]/text())[1]','varchar(100)') AS t03 ,o.value('(custom-attributes/custom-attribute[@attribute-id="t04"]/text())[1]','varchar(100)') AS t04 FROM @mockupTable t CROSS APPLY t.YourXml.nodes('/orders/order') A(o);

--这个比较容易.它只会选择前四个属性，无论它们有什么 id.

--This one is easier. It will pick just the first four attributes, no matter what id they have.

WITH XMLNAMESPACES(DEFAULT 'www address') SELECT o.value('(order-date/text())[1]','date') OrderDate --As in your other questions ,o.value('(custom-attributes/custom-attribute[1]/text())[1]','varchar(100)') AS ca1 ,o.value('(custom-attributes/custom-attribute[2]/text())[1]','varchar(100)') AS ca2 ,o.value('(custom-attributes/custom-attribute[3]/text())[1]','varchar(100)') AS ca3 ,o.value('(custom-attributes/custom-attribute[4]/text())[1]','varchar(100)') AS ca4 FROM @mockupTable t CROSS APPLY t.YourXml.nodes('/orders/order') A(o);

相关文章