Accessing custom properties from CS through API

I have found that I am unable to access custom properties (and even some basic properties like "Description") using the CMIS API. I have been trying to use the ELS API to create new objects and update existing objects with extended properties, but even with the ELS API I am not finding those properties. For example, there is a "Manager Name" field that I can see in Content Server, but invoking getObject() for an object does not display the Manager Name anywhere.

At some point, we'll be adding a lot of other fields to the objects, but I haven't quite figured out how to expose those properties and update those properties programatically.

Can someone confirm whether this is possible through the web service API (CMIS, ELS, or anything else)?

Thank you,

Faraz

Find more posts tagged with

Comments

eLink User

Hi Faraz.

I’m not sure which properties you are talking about.

ELS API.

Basic attributes like Name and Description are accessible through set and getDisplayName(), set and getComment() on the Element instance.

Type properties are accessible through set and getProperties() on the Element instance.

Categories are accessible through set and getCategories() on the Element instance. On the Category instance the call set and getProperties().

The Element instance is accessible though the ELService.getObject family calls and updatable through ELService.updateObject().

Type properties are not exposed at all in case they are not filled with values. However if such properties are present in the correspondent Type description you should be able to set new values for them.

CMIS.

Here is possible to manage only type properties and the object name. No category properties are exposed/managed yet. However there’s already proposals to deliver this feature in the next version.

Use the CmisObjectService to manage those properties by using getObject, getProperties and updateProperties.

Please notice that while addressing the same objects and types ELS and CMIS use two different notations/mappings for type property IDs. Please use the right ones by exploring the correspondent belonging types.

I hope this helps.

Stefano

From: eLink Entry: ELS General Forum [mailto:el_general_forum2@elinkkc.opentext.com]
Sent: Mittwoch, 21. September 2011 23:58
To: eLink Recipient
Subject: Accessing custom properties from CS through API

Accessing custom properties from CS through API

Posted by faraz.i.abbasi@accenture.com (Abbasi, Faraz) On 2011/09/21 17:57

At some point, we'll be adding a lot of other fields to the objects, but I haven't quite figured out how to expose those properties and update those properties programatically.

Can someone confirm whether this is possible through the web service API (CMIS, ELS, or anything else)?

Thank you,

Faraz

[To post a comment, use the normal reply function]
Forum:	ELS General Forum
Livelink Server:	knowledge-wlweb01

Faraz_Abbasi

Hi Stefano,

Thank you. I don't see the getCategories() and set/getProperties() methods in the ELS Web Service API. Am I missing something?

Basically, what I need to do is add a document to Content Server using the API which includes a few fields of metadata specific to our application. So I guess this needs to be implemented as a Category on the document, with the specific metadata fields as a property of the Category.

So what I need to figure out is how to create a new document with the appropriate Category assigned along with the associated metadata. Can this be achieved with a single Web Service call, or do I need to make multiple requests to achieve the result I need?

Once the document is available in Content Server with all the required metadata, I will need a process to retrieve the document based on one of the metadata fields. My plan there was to write this specific field to the Nickname property (in addition to the appropriate field in the Category definition), and use the getObjectByNickname() method to retrieve the ObjectId. I would then retrieve the actual content using the getContent() method in the ContentPortBinding.

Does this approach make sense?

Thank you,
Faraz

eLink User

Hi Faraz,

<![if !supportLists]>è <![endif]> “I don't see the getCategories() and set/getProperties() methods in the ELS Web Service API. Am I missing something?”

Are you binding the ELIB Web Services correctly?

Those call are related to the one object retrieved. You won’t find them directly on the Enterprise Library Service interface but on the Element instance.

<![if !supportLists]>è <![endif]>“Basically, what I need to do is add a document to Content Server using the API which includes a few fields of metadata specific to our application. So I guess this needs to be implemented as a Category on the document, with the specific metadata fields as a property of the Category.”

Right!

<![if !supportLists]>è <![endif]>“So what I need to figure out is how to create a new document with the appropriate Category assigned along with the associated metadata. Can this be achieved with a single Web Service call, or do I need to make multiple requests to achieve the result I need?”

You will need multiple ELIB interface calls to prepare your objects to create. However only the creation calls (elService.createFolder(folder), elService.createItem(item), elService.createCategoryType(parentId, catType)) will generate a server round-trip.

<![if !supportLists]>è <![endif]>“Once the document is available in Content Server with all the required metadata, I will need a process to retrieve the document based on one of the metadata fields. My plan there was to write this specific field to the Nickname prope! rty (in addition to the appropriate field in the Category definition), and use the getObjectByNickname() method to retrieve the ObjectId. I would then retrieve the actual content using the getContent() method in the ContentPortBinding. Does this approach make sense?”

This can make sense as long you generate unique nicknames for each object. Nicknames must be unique on the same system. You could consider them as a secondary key to the objectIds.

You can find a detailed documentation about ELIB Web Services interface in the Knowledge Center here.

https://knowledge.opentext.com/knowledge/llisapi.dll/Open/18073705

This is the 10.2.0 version you can start with. The 10.2.1 documentation is not released yet. Still you will be able to access to all information you mentioned.

Please have a look at the “Programming Guide for the Metadata API” to learn how to get along with the “Enterprise Library Web Services”.

You might want have a look at the “Programming Sample” package, which delivers C# and Java examples as well as the WS client stubs (elib.jar) and a “Eclipse Project Installation Guide”.

You will be able to grab code snippets to accomplish document and properties retrieval/creation/update.

I realized the example doesn’t contain a Category related code snippet. However once you get a ELIB object (Element) you will be able to perform such operations as well.

List<Category> categories = element.getCategories();

Category category = categories.get(0);

List<Property> properties = category.getProperties();

String displayName = category.getDisplayName();

String id = category.getId();

To create a category do the following

CategoryType catType = new CategoryType();

catType.setDisplayName("MyCat");

catType.setId("MyCatID");

catType.setVersion(3);

String categoryId = elService.createCategoryType("categoryParentId", catType);

I hope this helps you.

Stefano

Faraz_Abbasi

Hi Stefano,

Thank you for the detailed reply. We were originally planning on integrating using only the CMIS Web Services, and when that approach was deemed impossible (due to the inability to access additional Categories and Attributes), I started looking at the Enterprise Library Service API. I was hoping that adding documents could be done through a single Web Service (the integration is part of a larger BPEL process running on WebSphere Process Server). If I am reading this correctly however, I would basically need to create a local object in a Java or C# application to be able to access the detailed properties of the document object for both writing and reading.

Can you confirm whether my understanding above is correct?

Thank you!

Faraz

eLink User

Hi Faraz,

yes, your understanding is correct.

Typically you would access an object like following.

        //--------------------------------------------------------

        // Create the selector to be used for object accesses.

        //--------------------------------------------------------

        // The selector defines which parts of the fetched Element

        // instance will be populated. This allows a client to

        // reduce network traffic by requesting only data it is

        // interested in. Here we indicate that we want to get

        // just classification and metadata informations.

        //--------------------------------------------------------

        Selector  selector  = new Selector();

        selector.setClassifications(false);

        selector.setMetadata(true);

        selector.setAcl(true);

        selector.setSecClearanceInfo(false);

        List<String> itemIDs   = new ArrayList<String>();

        itemIDs.add(itemId); // Your Id here

        List<Element> elements = elService.getObjects(itemIDs, selector);

        String dispName = element.getDisplayName();

        String id = element.getId();

        String parentId = element.getParentId();

        List<Property> properties = element.getProperties();

        List<Category> categories = element.getCategories();

…

To create a document act like following.

        Item elem = new Item();

        elem.setType(typeName);

        elem.setParentId(parentId); // The parent folder id

        elem.setDisplayName(elemName);

        elem.setPrimaryClassification(primaryClassification);

        //--------------------------------------------------------

        // Setup the variant key.

        //--------------------------------------------------------

        // To create a master variant the major version number has

        // to be set to -1. Positive version numbers are only used

        // when a rendition variant is created.

        // As the master variant that we are going to create holds

        // pictures in several different image formats we do not

        // use an image MIME type as variant rendition but choose

        // application/octet-stream as variant rendition.

        //--------------------------------------------------------

        VariantKey variantKey = new VariantKey();

        variantKey.setMajorVersion(-1);

        variantKey.setRendition("application/octet-stream");

        //--------------------------------------------------------

        // Create the Variant instance.

        //--------------------------------------------------------

        Variant myVariant = new Variant();

        // ... defining which type ...

        myVariant.setType(variantType);

        // ... which item belongs to ...

        myVariant.setParentId(parentId);

        // ... and associating the variant key. No need to associate neither classification

        // (which is defined at item level already) nor a display name (automatically generated)

        myVariant.setKey(variantKey);

        //--------------------------------------------------------

        // Inserting properties.

        //--------------------------------------------------------

        List<Property> propertyList = myVariant.getProperties();

        Property property = new Property();

        property.setName(propertyName);

        List<String> valueList = property.getValues();

        for(String value : propertyValues)

            valueList.add(value);

        propertyList.add(property);

        //--------------------------------------------------------

        // Create Blob instances.

        //--------------------------------------------------------

        Blob blob;

        blob  = new Blob();

        blob.setMimeType(blobMimeType);

        blob.setName(blobName);

        // Attaching the blob to the variant element

        myVariant.getBlobs().add(blob);

        //--------------------------------------------------------

        // Create an item handle to create the item and the first

        // variant.

        //--------------------------------------------------------

        String handle = elService.createItemHandle(

            item,

            Collections.singletonList(variant),

            false);

        //--------------------------------------------------------

        // Upload files for Blob instances using the data

        // transfer service.

        //--------------------------------------------------------

        // pass blob name as URL parameter

        String blobName = URLEncoder.encode(blob.getName(), "UTF-8");

        handle = handle + "?blobName=" + blobName;

        // if handle is an item handle pass the index of the

        // variant as URL parameter

        if(variantIndex >= 0) {

            handle = handle + "&index=" + variantIndex;

        URL               url;

        HttpURLConnection urlCon;

        File              file;

        url = new URL("http://&quot;

            + serverNamePort

            + "/ot-transfer/write/"

            + handle);

        file   = new File(fileName);

        urlCon = (HttpURLConnection)url.openConnection();

        urlCon.setRequestMethod("PUT");

        urlCon.setDoInput(true);

        urlCon.setDoOutput(true);

        urlCon.addRequestProperty("Content-Type"  , blob.getMimeType());

        urlCon.addRequestProperty("Content-Length", "" + file.length());

        OutputStream os = null;

        InputStream  is = null;

        try {

            os = urlCon.getOutputStream();

            is = new FileInputStream(fileName);

            copyStream(is, os);

            int code = urlCon.getResponseCode();

            if(code != HttpURLConnection.HTTP_OK) {

                throw new IOException(urlCon.getResponseMessage());

        finally {

            if(is != null) {

                try {is.close();} catch(IOException ioe) {/* Ignore */}

            if(os != null) {

                try {os.close();} catch(IOException ioe) {/* Ignore */}

        //--------------------------------------------------------

        // Commit item creation. This actually the roundtrip to the server.

        //--------------------------------------------------------

        CommitItemResult cir = elService.commitItem(itemHandle);

Please refer also to the Sample Program here https://knowledge.opentext.com/knowledge/piroot/elws/v100200/elws-pgd/en/html/code-clientexample.htm#CLIENTEXAMPLE-INITELEMENT-1 expecially concerning MTOM (big) file upload/download procedures.

Regards,

Stefano.

Faraz_Abbasi

Thanks again, I greatly appreciate your responsiveness.

In this product specification for the Content Server SDK, it mentions that Web Services exist for "updat(ing) object categories and attributes". Is this referring to the Enterprise Library Service API? Because what I've learned thus far seems to indicate that Object Categories and Attributes are not directly editable through Web Services, but rather through C#/Java libraries only.

I apologize for going in circles on this but I just want to make sure I clearly understand how all these pieces fit together. Our Use Cases are relatively simple, and I want to make sure that we're doing it the right way, leveraging the right tools and processes.

eLink User

Hi Faraz,

the document you are referring to describes another interface, the Open Text Content Server Web Services (CWS).

For Enterprise Library documentation please use the links I provided so far.

In general the Enterprise Library interface has the following capabilities, which I guess could be interesting for you.

<![if !supportLists]>· <![endif]>Powerful and more abstracted interface (less round-trips).

<![if !supportLists]>· <![endif]>Typed document model.

<![if !supportLists]>· <![endif]>Typed metadata.

<![if !supportLists]>· <![endif]>Records Management integration.

<![if !supportLists]>· <![endif]>Direct access to Open Text Archive Server (Fast content upload/download). Optional.

What you get delivered is a WSDL file for client code generation you could use not only for C#/Java client class generation but also with any other platform/framework supporting Web Services capabilities.

Stefano