Using xpath.evaluate(): xpath exp with multiple namespaces

Hi,
I have to evaluate a xpath expression with parent node and child node having different namespaces. Like : env:parent/mig:child.
I have set the namespacecontext for the child node[i.e., for the prefix 'mig'.]
But am getting this error :
javax.xml.transform.TransformerException: Prefix must resolve to a namespace: env
How can I set the namespace context for both the parent and child nodes? Or is there are any other way of doing it?
I cant use //mig:child as the requirement needs the whole xpath expression [env:parent/mig:child] to be given as input for the xpath.evaluate() method.
Here's the code :
File file = new File("D:\\Backup\\XMLs\\test.xml");
Document xmlDocument = builder.parse(file);
XPath xpathEvaluator = XPathFactory.newInstance().newXPath();
xpathEvaluator.setNamespaceContext(new NamespaceContextProvider("env", "http://xmlns.oracle.com/apps/account/1.0"));
NodeList nodeList =
(NodeList)xpathEvaluator.evaluate("/env:parent/mig:child", xmlDocument,
XPathConstants.NODESET);
xpathEvaluator.setNamespaceContext(new NamespaceContextProvider("mig", "http://xmlns.oracle.com/apps/account/1.0"));
Thanks in advance.

If you want, I can help you tomorrow. Call me at my nieuwegein office or mail me at marco[dot]gralike[at]amis[dot]nl
I have some time tomorrow, so I can help you with this. My next presentation for UKOUG will be on XML indexes strategies anyway...
In the meantime and/or also have a look at:
XML Howto's (http://www.liberidu.com/blog/?page_id=441) specifically:
XML Indexing
* Unstructured XMLIndex (part 1) – The Concepts (http://www.liberidu.com/blog/?p=228)
* Unstructured XMLIndex (Part 2) – XMLIndex Path Subsetting (http://www.liberidu.com/blog/?p=242)
* Unstructured XMLIndex (Part 3) – XMLIndex Syntax Dissected (http://www.liberidu.com/blog/?p=259)
* Unstructured XMLIndex Performance and Fuzzy XPath Searches (http://www.liberidu.com/blog/?p=310)
* Structured XMLIndex (Part 1) – Rules of Numb (http://www.liberidu.com/blog/?p=1791)
* Structured XMLIndex (Part 2) – Howto build a structured XMLIndex (http://www.liberidu.com/blog/?p=1798)
* Structured XMLIndex (Part 3) – Building Multiple XMLIndex Structures (http://www.liberidu.com/blog/?p=1805)
The posts were based on Index for XML with Repeated Elements maybe that is a bit better to read than on my notepad on the internet (aka blog)
Edited by: Marco Gralike on Oct 28, 2010 7:51 PM

Similar Messages

XPath expression with multiple namespaces?

Hello all.
I have been scouring the forums and Google and can't seem to find anything similar to my problem.
I have the following XML with three namespaces. Two are defined in the root element and the third is defined in the the IdSession element. To make things even more confusing, the second and third namespaces have the same prefix, 'f'.
This is the xml:
<?xml version="1.0" encoding="ISO-8859-1" ?>
<NamespaceTestCall xmlns="http://my.default/namespace"
xmlns:f="http://my.second/namespace">
...<f:Level1>
......<f:IdSession xmlns:f="http://my.third/namespace">12345</f:IdSession>
......<Language>ENG</Language>
...</f:Nivel1>
</NamespaceTestCall>
My question is, how do I get at the IdSession element? Don't I need to create an XPath object and assign it more than one NamespaceContext?
This is what I am doing:
Document xmlDocument = loadXML(xmlSource);
XPath xpathEvaluator = XPathFactory.newInstance().newXPath();
xpathEvaluator.setNamespaceContext(new NamespaceContextProvider("a", "http://my.third/namespace"));
... xpathEvaluator.evaluate("//a:IdSession", ...);
This code works but it might not return the 'IdSession' I want, since by searching like this '//a:IdSession' it is looking in the whole document. If there were another 'IdSession' somewhere else in the document with the same URI, it would be returned. I want the 'IdSession' that lives inside of 'Level1'.
So what I would like to do is something like this:
... xpathEvaluator.evaluate("/*/Level1/a:IdSession", ...);
But this does NOT work because 'Level1' has its own namespace. So what it seems like I need to do is the following:
... xpathEvaluator.evaluate("/*/b:Level1/a:IdSession", ...);
Having already added the 'Level1' namespace to the XPath object, with the prefix 'b'. But unlike JDOM, there is no 'add' functionality, only 'set', meaning if you call set twice the second call overwrites the first.
Is there anyway to do this?
Many thanks!
Bob

Hello,
Sorry, that was my bad. I should have explained that NamespaceContextProvider is nothing more than my implementation of the NamespaceContext interface. The way I did it, I simply implemented getNamespaceURI() and getPrefix(). And the constructor accepted two parameters -- prefix and URI. So my problem was that when I assigned this NamespaceContext to my XPath object it would only have one prefix and one URI.
But I found an implementation here:
http://www.oreillynet.com/cs/user/view/cs_msg/50304
that instead of only having one prefix and URI uses a map. Thus its method setNamespace() adds the prefix and URi to the map, and getPrefix() and getPrefixes() retrieve them from the map.
Now when I want to use more than one namespace I simply call setNamespace() as many times as necessary, adding a prefix and URI pair each time, and when I am done I assign it to my XPath object.
And it works!
Thanks for the response!
Bob

Using the opmn:ormi protocol with multiple nodes

Hi
If I use the opmn:ormi protocol with multiple nodes, will it load balance between the nodes or will it simply try to connect to the first node and if it can't it will connect to the second node?.
Instead of specifying multiple nodes can I set the host to point to a load balancer?
Which solution is best?

repost

XPath query on multiple namespace XML document

Hi,
I am trying to locate an element in a WordML XML document using XPath.
The structure looks like:
<w:body>
     <wx:sect>
          <aml:annotation aml:id="0" w:type="Word.Bookmark.Start" w:name="DOC0000966"/>
...So I use the following expression "/wordDocument//body/sect" wich gives me a list with the <w:sect> nodes.
Then I search for "annotation[@type='Word.Bookmark.Start']" starting from each <w:sect> node. This yields the <aml:annotation> elem.
So far so good.
Now I have to handle a second case as well.
<w:body>
     <wx:sect>
     <w:p>
          <aml:annotation aml:id="0" w:type="Word.Bookmark.Start" w:name="DOC0000966"/>
...As you can see the <aml:annotation> is nested inside a <w:p> elem.
My intention was to use the expression "//annotation[@type='Word.Bookmark.Start']"to handle both cases, but this expression does not match neither.
I have tried several different alternatives with expressions and NamingContext implementations, none working.
What am I missing?
Thanks in advance,
Mariano.

Hello,
I decided to remove all namespaces and prefixes from the XML document and now the expression works.
I have chaned the following:
DocumentBuilderFactory is now namespace aware (it isn't by default).
          DocumentBuilderFactory builderFactory = DocumentBuilderFactory.newInstance();
          builderFactory.setNamespaceAware(true);I picked this NamespaceContext implementation from JAXP 1.4 samples and mapped all prefixes and namespaces coming inside the XML.
          NamespaceContextImpl ns = new NamespaceContextImpl();
          ns.bindPrefixToNamespaceURI("aml", "http://schemas.microsoft.com/aml/2001/core");
          ns.bindPrefixToNamespaceURI("dt", "uuid:C2F41010-65B3-11d1-A29F-00AA00C1488");
          ns.bindPrefixToNamespaceURI("o", "urn:schemas-microsoft-com:office:office");
          ns.bindPrefixToNamespaceURI("sl", "http://schemas.microsoft.com/schemaLibrary/2003/core");
          ns.bindPrefixToNamespaceURI("v", "urn:schemas-microsoft-com:vml");
          ns.bindPrefixToNamespaceURI("w", "http://schemas.microsoft.com/office/word/2003/wordml");
          ns.bindPrefixToNamespaceURI("w10", "urn:schemas-microsoft-com:office:word");
          ns.bindPrefixToNamespaceURI("wx", "http://schemas.microsoft.com/office/word/2003/auxHint");
          XPathFactory xpathFactory = XPathFactory.newInstance();
          XPath xpath = xpathFactory.newXPath();
          xpath.setNamespaceContext(ns);Finally, I prefixed all elements in xpath expressions.
"//aml:annotation[@w:type='Word.Bookmark.Start']"Happier now,
Mariano.

How to use Xpath effectively with Java

I am using Xpath for parsing a String with following syntax.This is a string suppose String xmlShopstring.
<shopper>
            <upc>0123456789</upc>
            <desc>Planeters Peanutus</desc>
            <regprice>1.99</regprice>
            <errorcode></errorcode>
        </shopper>In this case how can I use xpath parsing ?

You can use classes from javax.xml, org.w3c.dom packages for performing XML operations in java.
These APIs have very rich set of classes and methods for performing XML operations effectively.
pravi.pravi wrote: how can I use xpath parsing?You need to parse XML String to org.w3c.dom.Document with use of following classes:
javax.xml.parsers.DocumentBuilderFactory
javax.xml.parsers.DocumentBuilderOnce you parse XML String to org.w3c.dom.Document you can use following classes and others for very effective XPath parsing.
javax.xml.xpath.XPathFactory
javax.xml.xpath.XPath
javax.xml.xpath.XPathExpression
javax.xml.xpath.XPathConstantsI have listed some classes which can help you to perform XPath parsing, you should also explore other classes in the API for more XML operations.
Refer thread: http://forums.sun.com/thread.jspa?threadID=5357836
Thanks,
Tejas Purohit

XML Structured Index with multiple namespaces

Hi,
I'm having some trouble creating an xmlindex with structured component on a clob xmltype column without registered schema, whose data uses multiple namespaces.
The code I'm using atm:
CREATE TABLE "DECLARATIONS"
    "ID" NUMBER(19,0),
    "XML" "SYS"."XMLTYPE"
CREATE INDEX decl_header_ix ON "DECLARATIONS"(xml) INDEXTYPE IS XDB.XMLINDEX
PARAMETERS ('PATHS (INCLUDE (/emcs:emcsDeclaration/emcs:header//*)
                      NAMESPACE MAPPING (xmlns:emcs="http://www.myurl.eu/myapp/schema/emcs/nl"))');
INSERT INTO "DECLARATIONS" VALUES (1,'
<?xml version = ''1.0'' encoding = ''UTF-8'' standalone = ''yes''?>
<emcs:emcsDeclaration xsi:schemaLocation="http://www.myurl.eu/myapp/schema/emcs/nl emcs_domain.xsd"
xmlns:common="http://www.myurl.eu/myapp/schema/common"
xmlns:emcs="http://www.myurl.eu/myapp/schema/emcs/nl"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
   <emcs:header>
      <common:identifier>70</common:identifier>
      <common:declarationSequenceNumber>54566</common:declarationSequenceNumber>
      <common:dateCreated>2010-10-21-01:00</common:dateCreated>
      <common:status>01 Draft e-AAD in preparation</common:status>
   </emcs:header>
</emcs:emcsDeclaration>');A this moment it's not desirable for us to register the schemas used in oracle. According to the documentation I should be able to add a structured component to the index as follows:
BEGIN
      DBMS_XMLINDEX.registerParameter('MY_XSI_GROUP_PARAMETER'
              , 'ADD_GROUP GROUP MY_XSI_GROUP
               XMLTABLE decl_header
               XMLNAMESPACES (''http://www.myurl.eu/myapp/schema/emcs/nl'' AS emcs,
                       ''http://www.myurl.eu/myapp/schema/common'' AS common),
                    COLUMNS
              status VARCHAR2(30) PATH ''/emcs:emcsDeclaration/emcs:header/common:status/text()''
   END;
ALTER INDEX DECL_HEADER_IX PARAMETERS('PARAM MY_XSI_GROUP_PARAMETER');However this results in an ORA-00904: invalid identifier. After some experimenting it seems that oracle tries to parse the namespace URLs as identifiers (even tho http://download.oracle.com/docs/cd/E14072_01/appdev.112/e10492/xdb_indexing.htm#BCGJAAGH & http://download.oracle.com/docs/cd/E14072_01/appdev.112/e10492/xdb_xquery.htm#BABJCHCC specify the former), so I swapped them around:
BEGIN
      DBMS_XMLINDEX.dropParameter('MY_XSI_GROUP_PARAMETER');
      DBMS_XMLINDEX.registerParameter('MY_XSI_GROUP_PARAMETER'
          , 'ADD_GROUP GROUP MY_XSI_GROUP
          XMLTABLE decl_header
          XMLNAMESPACES (emcs ''http://www.myurl.eu/myapp/schema/emcs/nl'',
          common ''http://www.myurl.eu/myapp/schema/common''),
          COLUMNS
          status varchar2(30) PATH ''/emcs:emcsDeclaration/emcs:header/common:status/text()''
   END;
ALTER INDEX DECL_HEADER_IX PARAMETERS('PARAM MY_XSI_GROUP_PARAMETER');Oracle seems to get a bit further with this, resulting in a ORA-19102: XQuery string literal expected. Here I pretty much hit a dead end. Removing the xmlnamespaces declaration altogether leads to a ORA-31013: Invalid XPATH expression. Going through the examples on http://www.liberidu.com/blog/?p=1805 works fine, but as soon as I try to add namespaces to it they stop working as well.
So my question is: how do I get xmlnamespaces (with non-default namespaces) to work in a structured xmlindex component?

If you want, I can help you tomorrow. Call me at my nieuwegein office or mail me at marco[dot]gralike[at]amis[dot]nl
I have some time tomorrow, so I can help you with this. My next presentation for UKOUG will be on XML indexes strategies anyway...
In the meantime and/or also have a look at:
XML Howto's (http://www.liberidu.com/blog/?page_id=441) specifically:
XML Indexing
* Unstructured XMLIndex (part 1) – The Concepts (http://www.liberidu.com/blog/?p=228)
* Unstructured XMLIndex (Part 2) – XMLIndex Path Subsetting (http://www.liberidu.com/blog/?p=242)
* Unstructured XMLIndex (Part 3) – XMLIndex Syntax Dissected (http://www.liberidu.com/blog/?p=259)
* Unstructured XMLIndex Performance and Fuzzy XPath Searches (http://www.liberidu.com/blog/?p=310)
* Structured XMLIndex (Part 1) – Rules of Numb (http://www.liberidu.com/blog/?p=1791)
* Structured XMLIndex (Part 2) – Howto build a structured XMLIndex (http://www.liberidu.com/blog/?p=1798)
* Structured XMLIndex (Part 3) – Building Multiple XMLIndex Structures (http://www.liberidu.com/blog/?p=1805)
The posts were based on Index for XML with Repeated Elements maybe that is a bit better to read than on my notepad on the internet (aka blog)
Edited by: Marco Gralike on Oct 28, 2010 7:51 PM

Using a seagate external drive with multiple macs

i got a 1TB seagate external hard drive for backup and transfer purposes... it had a little program to help set it up and i set it up for use with mac and pc ( only onther option was stricty this computer.) it works on my computer and a pc but when i tried transfering files from another older mac i have it showed that i could only read on that computer...
how can i get it to work with multiple macs...?

I have the oldest macbook (1.83. 2 gigs ram, 60 gig internal hard drive)
I have all my logic content on my external maxtor firewire 400 drive (usb 2.0 is actually faster than firewire 400) and everything is fine. the good thing is if your external drive has 7200rpm, but the ones with 5600rpm should do as well.
Once you drag your apple loops to logic from your external drive, logic will remember for the next visit. And when you are opening a project which is on the external drive , just find it with logic finder and all should be fine.

XSD validation with multiple namespaces

Hi All,
I'm trying to validate some XML using an XSD that contains multiple namespace schema descriptions, as such, the main XSD file must import an XSD for each namespace.
The difficulty is that I cannot seem to find a way (in Oracle) to run a XSD validation using this (multi-XSD file) method.
Has anyone out there tackled a similar problem?
Cheers,
Ben

check out the class
CL_XML_SCHEMA
Regards
Raja

Any possiblity to use single Inbound Service Interface with multiple Receiver Agreements

Hi All,
Scenario: PROXY to FILE.
Requirement is to receive to two locations.
Is it possible to use single Inbound interface with multiple Receiver Agreements. I tried using two operation in the Inbound Service Interface, and Operation Specific in Determination. But couldnt proceed further. Do I need to use seperate Service Interface for two receiver locations?
Please suggest some solution.

Hi Naveen,
There is a possibility but in the case in the same server location with dynamic configuration: Placing file in two different directories using single receiver communication Channel
With different location you would need to develop a new adapter module as Amit Srivastava explains and develops here Send File to Two Different Locations using Adapter Module
Regards.

Can you use the same Apple ID with multiple iMac's?

I am a teacher and I have multiple iMac's in my room and I need to be able to purchase apps one time and have them on all my computers. Is this possible? I know that I can do it with multiple iPad's but can I do with iMac's?

Yes logon to each computers itunes and away you go.
To redownload on each itunes got to the itunes store, select Purhased from the quick links section and navigate to what you want to redownload.
If you want to do it for apps that run on the Macs not on the iPad you need to login to the Mac AppStore and login with the same apple id and download apps that work on macs

XPath query with out namespace is possible ?

Hi,
our system reciving xml data using the datatype xsd[b] <any/> . am sure the system will receive the data in same format[node hierarchical] but the namespace of the root element may be differ..
i want write xpath which should not be parse based on the namespace of element.
like the method setNameSpaceAware(false) in java;
will it be possible in bpel. if so how ?
i would be appreciate any help.
Thanks,
Bogi

Ahhhh - thank you for the clarification.
Regarding the expansion of variable, is there anything further needed? Reason I ask is, when I use the variable which contains the same string as when I hard code the string in the XPATH query, I do not receive the desired results - I only receive the
'VENDOR_ADVISORY' results.

Spry XmlDataSet: columns using XPath, subPaths with axis

A couple of questions:
I'm trying to add additional columns to a dataset whose value is not based a node or attribute but is derived using an XPath expression containing a function.
Tried to define this using "subPaths" but it results in an error message
var ds = new Spry.Data.XMLDataSet("my.xml", "//node]", {subPaths: ['position()', 'parent::local-name()']});
The idea was to generate a flat table from a the xml tree in order to speed up filtering (filtering by ds.setXPath rebuilds the whole dataset)
Is it possible to specify the columns and types when instantiating a dataset?
var ds = new Spry.Data.XMLDataSet("my.xml", "//node]", {columns: [{'node': 'html'},{'@attr': 'text'}]);
Spry seems to always decode entities in attributes, how do I prevent that?
source XML:
<outline text="attribute with entities <a href="#">link</a>" />
setting column type to text:
ds.setColumnType('@text', 'text');
has no effect, entities are still decoded
Using setXPath('//node') results in an error message however using "//node" when calling the constructor does not.
Any work-around?

Hi Gramps, thank your for responding.
From my post is might not have been obvious that i had studied that page in detail, i have also studied the Spry JavaScript files and the google AJAXSLT library used by Spry for parsing XPaths in JavaScript.
However i can't find an answer to my questions in the documents you reference, could you be more specific in pointing out the secion you believe answers my question(s)?
What i did find was a way to stop spry from throwing "Regular Expression is too complex" from inside the function
Spry.Utils.setInnerHTML
by wrapping the regex evaluation in a try/catch; this temporarily avoids the problems in MSIE but at the cost of not escaping scripts, so i would like to understand why sometimes that exception is thrown. I suspect it has to do with the string being evaluated instead of the regex-pattern.
And i found a way to add static DataSet references, however in order to do that i had to replace the function
Spry.Data.DataSet.prototype.getValue
References i've added so far:
Data reference
Content provider
{ds_XPath}
Spry.Data.SpryXMLDataSet.getXPath()
{ds_URL}
Spry.Data.HTTPSourceDataSet.getURL()
Feature request:
I think it would be great to be able to register additional data references to a spry dataset
ds.addDataReference('ds_XPath', function(dataset, rowContext, valueName){return dataset.getXPath()});
Link to SpryDataMod.js
http://feedfunnel.noagendareport.com/assets/spry/SpryDataMod.js

How to use one I-tunes library with multiple windows users (family) on 1 pc

On our family pc we log in with different accounts. We want to use I-Tunes but all have the same library. I tried to create a library not in the users directory, but in the All Users folder that is accessible for everyone. However, when I import the music in the library for one user and set the same location for another user, I-tunes does not recognize the library. So everyone has to create it's own library.
What are the best practices to overcome this?

Keep in mind that what you have done now is you have multiple iTunes libraries accessing the same media folder. Meaning, that each account can have their own ratings, playlists, etc. but more importantly, each account does NOT know what the other account does, so if your daughter adds a CD to her account, it will go into the shared folder, but it will NOT automatically go into your wife's iTunes library. She would have to add it manually. Likewise, if one account deletes music from their library and tells iTunes to remove it to the trash, the other accounts will not know and iTunes will still try to access the file from the shared location even though the file is now trashed.
Each account has their own database files (.itl files in the iTunes folder) which stores this stuff.
If you want to have everyone use the exact same library which means the same ratings, playlists, etc. then you need to move the database files into that same location and set each iTunes account to read the same data.
How to open an alternate iTunes Library file or create a new one
http://docs.info.apple.com/article.html?artnum=304447
Patrick

How do I go from using a single Apple ID with multiple libraries to using home sharing?

Until now my wife and I have opted to share our iTunes/Apple ID and use multiple libraries to manage our iPhones. This has become increasingly difficult...with issues such as having apps I've downloaded that she doesn't want automatically downloading to her library and even her phone. Now, add in the fact that I want to continue to manage my iTunes library from our PC and allow her to manage hers from her new MacBook Pro and we have good reason to move to home sharing with separate Apple IDs for each of us.
My question is the logistics of doing so. Not only do we need to replicate her existing library on a new Apple ID, but then we need to migrate all the necessary files over to her MacBook Pro. Quite honestly, I don't even know where to begin. Is it possible to associate her existing library with her new Apple ID without having to create her library from scratch? Then is there anyway to migrate only the items in her library over to her MacBook Pro as opposed having to move all of our music & apps? Where do I begin?
Thanks!

Okay...I misunderstood and thought we needed separate ID's...but I see we actually need to use the same ID which simplifies things in my mind a bit. But still some followups...
1) What do you mean by restore the library on her computer? Simply open her library file in iTunes?
2) On our current machine...media is actually split between a couple folders but all under the My Music folder. But there are multiple subfolders (iTunes, Her iPhone, and Amazon MP3). I presume as long as I copy over the entire My Music folder there shouldn't be any access issues when the library is opened on her machine?
3) Is there any easy way to get rid of all the extra files that she doesn't have in her library? For instance, all my music and apps that she doesn't actually want, is there any easy way of separating those out from the files that are actually linked to in her library? Vice versa for me...once we've go her library set up on her machine, any easy way to purge all the extra files from my PC that aren't actually in my library?
Thanks!

Using a single itunes library with multiple iphones?

My wife got an iPhone 3GS recently and I already have a 4S. I plan to use an existing iTunes install on a single PC to manage them both. I see that Apple support articles say it's worth keeping iTunes libaries separate (with separate Windows user accounts) when there are multiple iPhones using iTunes on one PC.
I like to manually manage music, videos and podcasts in my iTunes library rather than use synching. I like the idea of having a shared pool of music, videos and apps for us both to use. I know apps that are purchased on my iPhone or my wife's are linked to a specific apple ID so that means if she transfers any apps that were bought on my phone originally, her phone will ask her for my apple ID password when they get updates. But sharing passwords with eachother isn't a problem.
I can create a second user account for my wife easily, but I kind of have the feeling I'd prefer to share a single library as I say.
Are there any other reasons I need to keep 2 user accounts and 2 iTunes libraries going when I manually manage the transfer of content?
Cheers.

MikeBelfast wrote:
I can create a second user account for my wife easily, but I kind of have the feeling I'd prefer to share a single library as I say.
Adn this can easily be done.
Move your /Music/iTunes/ folder to a Shared directory (such as /Users/Public/ folder) and everyone can use this single library.
Each user must quit iTunes before the other can open it.
Just hold Shift, launch iTunes, select Choose library... and select the iTunes folder in the shared directory.
The big advantage with separate Windows user accounts, data such as photos, calendars, emails, and everything else will be separate.

Using xpath.evaluate(): xpath exp with multiple namespaces

Similar Messages

Maybe you are looking for