解析带有Javascript的“非标准” XML？

Question

Good afternoon,

I am trying to parse through an XML file with strictly Javascript (I would prefer to not use an add on such as jQuery because of memory constraints in the module this codes resides in). The XML file is populated by a module within a control system I have in an assembly line controller so I think the format is a little different that typical XML files. I can load the file correctly (I used an alert to see the raw text output and it is correct), but I can't load the value of the tag. A portion of the XML file loaded is below:

<tag name = "Parts_Built_Buffer[490].DateTime" valueType="cip:dt_STRINGI" path="1,0" access="read" display="String">
    <value xsi:type="cip:dt_STRINGI">20140804110319</value>
</tag>

And it continues for 500 such tags... Every tag has a unique name due to the nature of the module requirements, so I want to loop through all the tags to get the values and build a table. For this my code (minus the loop part as I need to crawl (load 1 value) before I run (load all values)) is:

var builtDataView_url = "/user/system/dataviews/Parts_Built_Buffer.xml";

    var getDataView = new XMLHttpRequest();
    getDataView.open("GET", builtDataView_url, false);
    getDataView.send();
    var builtXML = getDataView.responseXML;
    alert("response is " + getDataView.responseText)

    var job0 = builtXML.getElementsByTagName("/tag[Name='Parts_Built_Buffer[0].DateTime']/value")[0];
    alert("job0 is " + job0)

In the first alert I get a text view of the XML and it appears as expected. However, for job0 I get that it is undefined (if I alert job0.length I get 0). I need this to work in both Chrome and IE. Any advice will be greatly appreciated.

EDIT: I got to play around in the console some today (I ran out of time yesterday and this morning) with the data returned. I am fairly new to javascript and am self (ie Google) taught. The console dialog is below, the XML returned is not grouping the value tag with the parts_built... tag. Since I am new to this, no making fun of my attempts to read the data if the function calls don't make sense, please...

data.getElementsByTagName("*")
HTMLCollection[101]
data.getElementsByTagName("*")[0]
<view xmlns:xsi=​"http:​/​/​www.w3.org/​2001/​XMLSchema-instance" xmlns=​"http:​/​/​    www.rockwellautomation.com/​technologies/​data_access/​data_views/​1.0/​" xmlns:cip=​"http:​/​/​www.rockwellautomation.com/​technologies/​data_access/​data_types/​1.0/​" xsi:schemaLocation=​"http:​/​/​www.rockwellautomation.com/​technologies/​data_access/​data_views/​1.0/​ /​schema/​  dataview.xsd" name=​"EWEB_Data" description>​…​</view>​
data.getElementsByTagName("*")[1]
<tag name=​"Parts_Built_Buffer[490]​.DateTime" valueType=​"cip:​dt_STRINGI" path=​"1,0"    access=​"read" display=​"String">​
<value xsi:type=​"cip:​dt_STRINGI">​20140805111641​</value>​
</tag>​
data.getElementsByTagName("*")[1].value
undefined
data.getElementsByTagName("*")[2]
<value xsi:type=​"cip:​dt_STRINGI">​20140805111641​</value>​
data.getElementsByTagName("*")[3]
<tag name=​"Parts_Built_Buffer[490]​.VIN" valueType=​"cip:​dt_STRINGI" path=​"1,0" access=​    "read" display=​"String">​…​</tag>​
data.getElementsByTagName("*")[4]
<value xsi:type=​"cip:​dt_STRINGI">​</value>​
data.getElementsByTagName("*")[5]
<tag name=​"Parts_Built_Buffer[490]​.SerNum" valueType=​"cip:​dt_STRINGI" path=​"1,0"     access=​"read" display=​"String">​
<value xsi:type=​"cip:​dt_STRINGI">​643Y037S14217​</value>​
</tag>​
data.getElementsByTagName("*")[6]
<value xsi:type=​"cip:​dt_STRINGI">​643Y037S14217​</value>​
data.getElementsByTagName("*")[5].getElementsByTagName("*")
[
<value xsi:type=​"cip:​dt_STRINGI">​643Y037S14217​</value>​
]
data.getElementsByTagName("tag name*")[5].getElementsByTagName("*")
TypeError: Cannot read property 'getElementsByTagName' of undefined
data.getElementsByTagName("Parts_Built_Buffer*")[5].getElementsByTagName("*")
TypeError: Cannot read property 'getElementsByTagName' of undefined

This code will produce the value (keep in mind I only care about the even tag indices as they hold the values... data.getElementsByTagName("*")[6].textContent

This seems very clunky though so any advice will still be appreciated... At least it looks like I'm not going to have to start with the text output and parse through a long text string...

Answer 1

You can use the standard DOM properties: childNodes , firstChild , nextSibling .

So, I would navigate to the node that contains all tags, and then I would use a for loop on the childNodes . Something similar to this:

var main_node = something;
var table = {};
for (var i = 0; i < main_node.childNodes.length; i++) {
    var node = main_node.childNodes[i];
    if (node.nodeType == 1) {
        // Insert your code here.
        var name = node.getAttribute('name');
        var value = writeYourFunctionToExtractValueFromNode(node);
        table[name] = value;
    }
}

// After the pre-processing step above, you can just use:
alert(table['Parts_Built_Foobar']);

The check for nodeType is to skip comments, whitespace, and other non-relevant nodes.

Everything I mentioned here is just standard DOM, and is available essentially everywhere (including in languages other than JavaScript).