在VBScript中遍历和过滤xml节点

Question

how can I get company, firstname, lastname, mobile, email, country, city and zipcode values for each customer from the below xml? I tried the following code but when some value is missing (eg email in the first record) the code assigns value from the next record so "ABCars" gets "gwood@gmail.com".

Also, do you know how can I filter the loaded records so that only records with mobile number not existing in another recordset left (recordset is not included in the below code)?

Set xmlDoc = CreateObject("Msxml2.DOMDocument")
xmlDoc.async="false"
xmlDoc.setProperty "SelectionLanguage", "XPath"
xmlDoc.load("test.xml")

set  nodes=xmlDoc.SelectNodes("//customer")
for i=0 to nodes.length-1
set company=xmlDoc.SelectNodes("//customer/company")
set firstname=xmlDoc.SelectNodes("//customer/firstname")
set lastname=xmlDoc.SelectNodes("//customer/lastname")
set mobile=xmlDoc.SelectNodes("//customer/mobile")
set email=xmlDoc.SelectNodes("//customer/email")
set country=xmlDoc.SelectNodes("//customer/address/country")
set city=xmlDoc.SelectNodes("//customer/address/city")
set zipcode=xmlDoc.SelectNodes("//customer/address/zipcode")

XML:

<?xml version="1.0" encoding="UTF-8"?>
<list>
    <category>Cars</category>
    <customers>
        <customer>
            <company>ABCars</company>
            <firstname>Peter</firstname>
            <lastname>Heinrich</lastname>
            <mobile>9141453027</mobile>
            <address>
                <country>Germany</country>
                <city>Berlin</city>
                <zipcode>12345</zipcode>
            </address>
        </customer>
        <customer>
            <company>Best Cars</company>
            <firstname>George</firstname>
            <lastname>Wood</lastname>
            <mobile>123456789</mobile>
            <email>gwood@gmail.com</email>
            <address>
                <country>Great Britain</country>
                <city>Leicaster</city>
                <zipcode>67890</zipcode>
            </address>
        </customer>
    </customers>
</list>

Regards, Przemek

Answer 1

Your loop goes through nodes collection but the loop body queries the whole document for each iteration, so you are doing the same query every time - use the returned nodes as context nodes to select nodes relative to that node:

set  nodes=xmlDoc.SelectNodes("//customer")
for i=0 to nodes.length-1
    ' get the current node
    set node = nodes(i)
    ' run xpath relative to the current node
    set company = node.selectSingleNode("company")
    set firstname = node.selectSingleNode("firstname")
    set lastname = node.selectSingleNode("lastname")
    set mobile = node.selectSingleNode("mobile")
    set email = node.selectSingleNode("email")
    set country = node.selectSingleNode("address/country")
    set city = node.selectSingleNode("address/city")
    set zipcode = node.selectSingleNode("address/zipcode")

next

If there is no node found by the XPath query, Nothing will be returned which you can check and route appropriately:

set company = node.selectSingleNode("company")
if company is Nothing then
    ' do something if no company, eg break loop
    exit for
end if

For the filtering, you could build your XPath query to select only those nodes which aren't in that list, eg

dim xpath: xpath = Empty

do until recordset.EOF
    if xpath <> "" then
       xpath = xpath & " and "
    end if 
    xpath = xpath & "mobile != '" & recordset("mobile") & "'"

    recordset.MoveNext
loop

if xpath <> "" then
    xpath = "[" & xpath & "]"
end if

' eg you end up with something like "//customer[mobile != 123456789 and mobile != 987654321]" 
set nodes = xmldoc.selectNodes("//customer" & xpath)