通用XSLT实现无嵌套XML转JSON的可行性及方法

Absolutely! For flat XML structures (no nested elements under the root, just direct key-value pairs), you can create a fully generic XSLT that doesn't require hardcoding any element names—just like the XML-to-CSV approach you referenced. Here's a practical implementation:

Core Approach #

The logic is simple and flexible:

• Target the root element of your XML
• Iterate over all its direct child elements
• Map each element’s name to a JSON key, and its text content to the corresponding value
• Ensure the output follows valid JSON syntax (proper quoting, no trailing commas, etc.)

Generic XSLT Code #

<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="2.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
    <xsl:output method="text" encoding="UTF-8" indent="yes"/>

    <!-- Match the root node to start building the JSON object -->
    <xsl:template match="/">
        <xsl:text>{</xsl:text>
        <!-- Process every direct child element of the root -->
        <xsl:for-each select="*/*">
            <!-- Write the JSON key (element name) -->
            <xsl:text>"</xsl:text>
            <xsl:value-of select="local-name()"/>
            <xsl:text>": "</xsl:text>
            <!-- Write the JSON value, escaping double quotes to avoid invalid JSON -->
            <xsl:value-of select="replace(., '&quot;', '\&quot;')"/>
            <xsl:text>"</xsl:text>
            <!-- Add a comma only if this isn't the last element -->
            <xsl:if test="position() != last()">
                <xsl:text>,</xsl:text>
            </xsl:if>
        </xsl:for-each>
        <xsl:text>}</xsl:text>
    </xsl:template>

    <!-- Ignore empty whitespace text nodes to keep output clean -->
    <xsl:template match="text()[not(normalize-space())]"/>
</xsl:stylesheet>

How It Works #

Let’s break down the key components:

• Root Matching: The top template starts the JSON object with { and targets the root of your XML (works with any root name, like your <POSTrans> example).
• Element Traversal: */* selects all direct children of the root, so you don’t need to specify element names upfront.
• Value Escaping: The replace() function handles double quotes in element text, preventing them from breaking the JSON structure.
• Comma Management: The position() != last() check ensures we don’t add a trailing comma, which would make the JSON invalid.
• Whitespace Cleanup: The final template filters out empty whitespace nodes that might otherwise clutter the output.

Test with Your Sample XML #

Using your example snippet (extended with a few more elements for clarity):

<POSTrans>
    <POSDocNo>INV1200004K89</POSDocNo>
    <Invoice_Exchange>Invoice</Invoice_Exchange>
    <CustomerID>CUST-789</CustomerID>
    <Total>299.99</Total>
</POSTrans>

Applying the XSLT will output valid JSON:

{
    "POSDocNo": "INV1200004K89",
    "Invoice_Exchange": "Invoice",
    "CustomerID": "CUST-789",
    "Total": "299.99"
}

Edge Case Adjustments #

• Empty Elements: If an element has no text, it will output as "ElementName": "" (valid JSON).
• Additional Escaping: To handle backslashes or other special characters, add extra replace() calls (e.g., replace(replace(., '"', '\"'), '\\', '\\\\')).
• XSLT 1.0 Compatibility: If you’re limited to XSLT 1.0, replace the replace() function with a recursive escaping template— the core traversal logic stays the same.

内容的提问来源于stack exchange，提问作者paul