XPath表达式有条件地获取相邻节点

Question

I have an xml file with some content like the following:

<p>
    <r>
        <t xml:space="preserve">Reading is easier, </t>
    </r>
    <r>
        <fldChar fldCharType="begin"/>
    </r>
    <r>
        <instrText xml:space="preserve"> REF _Ref516568558 \r \p \h </instrText>
    </r>
    <r>
        <fldChar fldCharType="separate"/>
    </r>
    <r>
        <t>This is all the text I want to capture</t>
    </r>
    <r>
        <fldChar fldCharType="end"/>
    </r>
    <r>
        <t xml:space="preserve">, in the new Reading view </t>
    </r>
    <r>
        <fldChar fldCharType="begin"/>
    </r>
    <r>
        <instrText xml:space="preserve"> REF _Not516755367 \r \h </instrText>
    </r>
    <r>
        <fldChar fldCharType="separate"/>
    </r>
    <r>
        <t>But not this...</t>
    </r>
    <r>
        <fldChar fldCharType="end"/>
    </r>
    <r>
        <t xml:space="preserve"> Some other text... </t>
    </r>
</p>

I know that I can use the XPath expression //instrText[contains(text(), '_Ref')] to get <instrText xml:space="preserve"> REF _Ref516568558 \\r \\p \\h </instrText> .

Now what I want to get is the text within t nodes between <fldChar fldCharType="begin"/> and <fldChar fldCharType="end"/> if between these two tags there is a instrText with text that contains '_Ref' ie instrText[contains(text(), '_Ref'] .

Based on this, from the example xml, I would expect only: <t>This is all the text I want to capture</t> to be returned.

Can this be done with a single XPath 1.0 expression?

Answer 1

试试这个： p/r[preceding-sibling::r[fldChar/@fldCharType='begin'] and following-sibling::r[fldChar/@fldCharType='end']]/t[contains(., '_Ref')]

Answer 2

这就是我最终使用的内容： //p/r[preceding-sibling::r[fldChar/@fldCharType='begin'] and following-sibling::r[fldChar/@fldCharType='end']][instrText[contains(text(), '_Ref')]]/following-sibling::r[t][1]