@adobe/acc-js-sdk 1.1.14 → 1.1.16

package/compile.js

@@ -31,6 +31,7 @@ var resources = [
     { name: "./transport.js" },
     { name: "./xtkCaster.js" },
     { name: "./domUtil.js" },
+    { name: "./xtkJob.js" },
     { name: "./cache.js" },
     { name: "./entityAccessor.js" },
     { name: "./xtkEntityCache.js" },

package/docs/_data/navigation.yml

@@ -83,6 +83,8 @@
       link: methodLevelRepresentation.html
     - name: XTK Interfaces
       link: xtkInterface.html
+    - name: XTK Jobs
+      link: xtkJob.html
     - name: Handling timeouts
       link: timeouts.html
     - name: HTTP Headers

package/docs/_posts/2022-10-14-welcome.html

@@ -129,7 +129,7 @@ console.log(`About to call method '${methodName}' of scope '${scopeName}'`);
     returnedValue = undefined;
   }
   // If the API returns one parameter which is an object, the return it directly.
-  // If not (return value is a litteral), then use a JSON with the return parameter name
+  // If not (return value is a literal), then use a JSON with the return parameter name
   else if (outNames.length == 1) {
     if (result && typeof result == "object")
       returnedValue = result;

package/docs/changeLog.html

@@ -2,6 +2,20 @@
 layout: page
 title: Change Log
 ---
+<section class="changelog"><h1>Version 1.1.16</h1>
+  <h2>2022/11/29</h2>
+  
+  <li>Added support for xtk:job interface and job-related methods. See <a href="https://opensource.adobe.com/acc-js-sdk/xtkJob.html">Jobs documentation</a> for more details</li>
+</section>
+
+
+<section class="changelog"><h1>Version 1.1.15</h1>
+  <h2>2022/11/28</h2>
+  
+  <li>Fix an issue with SimpleJson serialization. See <a href="https://opensource.adobe.com/acc-js-sdk/simpleJson.html">SimpleJson documentation</a> for more details, and in particular the <b>Compatibility Notes</b> section</li>
+  <li>Fix credentials passing for fileUpload and getReportData methods. This is still a temporary fix until those APIs are made IMS compatible.</li>
+</section>
+
 <section class="changelog"><h1>Version 1.1.14</h1>
   <h2>2022/11/25</h2>
   
@@ -202,7 +216,7 @@ For breaking changes see the [migration guide](MIGRATION.md)
 <li><b>getEntityIfMoreRecent</b> now takes an additional parameter which allows to specify and force a representation (xml, json...)
 <li>New helper function <b>DomUtil.isArray</b> to test if a JavaScript object is an array
 <li>New <b>EntityAccessor</b> object which allows to get attributes and child elements from xml or json objects, regardless of their representation
-<li>New <b>escapeXtk</b> function to escape litteral values in Xtk expression. Can be used as a function or as a tagged template litteral
+<li>New <b>escapeXtk</b> function to escape literal values in Xtk expression. Can be used as a function or as a tagged template literal
 <li>New function <b>XtkCaster._variantStorageAttribute</b> which returns the name of a schema attribute used to store variant value types. The name of the attribute depends on the type: stringValue, longValue, etc.
 <li>Support for non static method that mutate the object on which they apply. For instance, the <b>xtk:queryDef#SelectAll</b>
 <li>Added samples in the samples/ folder
@@ -229,7 +243,7 @@ _Breaking changes in 1.0.0_
 <li>Changes in the <b>sdk.init</b>, <b>Client</b> constructor, and <b>logon</b> functions. Now using <b>ConnectionParameters</b> and <b>Credentials</b> objects to configure a Campaign connection
 <li>Client object members are now private: access to representation, etc. attributes is not allowed anymore except for <b>NLWS</b>, <b>XtkCaster</b>, and <b>DomUtil</b>
 <li>Access to the <b>sessionInfo</b> object after <b>logon</b> can be done via the new <b>getSessionInfo</b> call
-<li>Options cache internal strucutre change: option values in the cache are now object litterals containing the option value, type, and raw value (which may not be casted to the expected type yet)
+<li>Options cache internal strucutre change: option values in the cache are now object literals containing the option value, type, and raw value (which may not be casted to the expected type yet)
 <li>Connecting to mid-sourcing (or other Campaign instances which are defined by an external account) is now done with the <b>ConnectionParameters.ofExternalAccount</b> function. As a consequence, <b>getSecretKeyCipher</b> is now private and deprecated
 <li>CampaignException object signature changed (but was not previously exposed)
 <li>The client-side bundle is now generated in the <b>dist/bundle.js</b> file instead of <b>bundle.js</b>

package/docs/escaping.html

@@ -4,7 +4,7 @@ title: Escaping
 ---
 
 
-<p>It's common to use variables in query conditions. For instance, in the above example, you'll want to query an account by name instead of using the hardcoded <b>ffda</b> name. The <b>expr</b> attribute takes an XTK expression as a parameter, and <b>ffda</b> is a string litteral in an xtk expression.</p>
+<p>It's common to use variables in query conditions. For instance, in the above example, you'll want to query an account by name instead of using the hardcoded <b>ffda</b> name. The <b>expr</b> attribute takes an XTK expression as a parameter, and <b>ffda</b> is a string literal in an xtk expression.</p>
 
 <p>To prevent xtk ingestions vulnerabilities, you should not concatenate strings and write code such as expr: "@name = '" + name + "'": if the value of the name 
   parameter contains single quotes, your code will not work, but could also cause vulnerabilities.
@@ -13,19 +13,19 @@ title: Escaping
 
 <h1>sdk.escapeXtk</h1>
 
-<p>The <b>sdk.escapeXtk</b> can be used to properly escape string litterals in xtk expressions. The function will also surround the escaped value with single quotes.</p>
+<p>The <b>sdk.escapeXtk</b> can be used to properly escape string literals in xtk expressions. The function will also surround the escaped value with single quotes.</p>
 
 <p>You can use string concatenation like this. Note the lack of single quotes around the value.</p>
 <pre class="code">
 { expr: "@name=" + sdk.escapeXtk(name) }
 </pre>
 
-<p>or a template litteral</p>
+<p>or a template literal</p>
 <pre class="code">
 `{ expr: "@name=${sdk.escapeXtk(name)}" }`
 </pre>
 
-<p>The <b>escapeXtk</b> function can also be used to create tagged string litterals. This leads to a much shorter syntax. Note that with this syntax, only the parameter values of the template litteral are escaped</p>
+<p>The <b>escapeXtk</b> function can also be used to create tagged string literals. This leads to a much shorter syntax. Note that with this syntax, only the parameter values of the template literal are escaped</p>
 <pre class="code">
 sdk.escapeXtk`{ expr: "@name=${name}" }`
 </pre>

package/docs/simpleJson.html

@@ -4,7 +4,7 @@ title: SimpleJson format
 ---
 
 <p>The <b>SimpleJson</b> format is the format used by default by the SDK to conver between Campaign XML and JSON</p>
-
+<p></p>
 
 <ul>
   <li>XML attributes are JSON properties (without the “@” sign)</li>
@@ -27,6 +27,7 @@ title: SimpleJson format
 </pre>
 
 
+<h1>XML Root</h1>
 <p>The XML root element tag is automatically determined by the SDK as it's generating the XML, usually from the current schema name.</p>
 
 <pre class="code">
@@ -34,15 +35,21 @@ XML:     &lt;root/>
 JSON:    {}
 </pre>
 
-<p>XML attributes are mapped to JSON attributes with the same name, whose litteral value can be a string, number, or boolean. There's no "@" sign in the JSON attribute name.
-Values in JSON attributes can be indifferently typed (ex: number, boolean), or strings (ex: "3" instead of just 3) depending if the conversion could determine the attribute type or not. 
-API users should expect and handle both value and use the `XtkCaster` object to ensure proper conversion when using.</p>
+
+
+
+<h1>XML Attributes</h1>
+<p>XML attributes are mapped to JSON attributes with the same name, whose literal value can be a string, number, or boolean. There's no "@" sign in the JSON attribute name.
+Values in JSON attributes can be either typed (ex: number, boolean), or strings (ex: "3" instead of just 3) depending if the conversion could determine the attribute type or not. 
+API users should expect and handle both value and use the <b>XtkCaster</b> object to ensure proper conversion.</p>
 
 <pre class="code">
 XML     &lt;root hello="world" count="3" ok="true"/>
 JSON:   { hello:"world", count:"3", ok:"true" }
 </pre>
 
+
+<h1>XML Elements</h1>
 <p>XML elements are mapped to JSON objects</p>
 
 <pre class="code">
@@ -64,7 +71,9 @@ XML:    &lt;root>&lt;item id=1/>&lt;item id=2/>&lt;/root>
 JSON:   { item: [ { id:"1" }, { id:"2" } ] }
 </pre>
 
-<p>Text of XML element is handle with the <b>$</b> sign in the JSON attribute name, or with a child JSON object name <b>$</b></p>
+
+<h1>Text nodes</h1>
+<p>Text of XML element is handled with the <b>$</b> sign in the JSON attribute name, or with a child JSON object name <b>$</b></p>
 
 <p>Text of the root element</p>
 <pre class="code">
@@ -86,3 +95,67 @@ XML:    &lt;root>&lt;item>Hello&lt;child id="1"/>
         &lt;/root>
 JSON:   { item: { $: "Hello", child: { id:"1" } }
 </pre>
+
+<p>Normally Campaign will not generate XML with elements containing multiple sibling text nodes. If this should happen, the SDK will consider them as a single text value, i.e. it will concatenate the contents of each text and CDATA node as if there was only one. However, whitespaces are processed independently for each text node.
+</p>
+
+<h1>Whitespaces</h1>
+<p>In XML documents, whitespaces can be either significant or insignificant. The SDK uses the following rules to strip whitespaces, which correspond to how Campaign specifically uses XML</p>
+<p> </p>
+<ul>
+<li>The characters " ", "\t", "\r" and "\n" are considered whitespaces</li>
+<li>CDATA sections are kept unchanged, i.e. whitespaces are preserved exactly</li>
+<li>For TEXT nodes directly under the root node are, whitespaces will be trimmed as explained below</li>
+<li>For element nodes which only contain text and possibly attributes (for instance &lt;node&gt;value&lt;/node&gt;</li>, the text is kept unchanged, i.e. whitespaces are preserved exactly</li>
+<li>For element nodes which have child elements, whitespaces will be trimmed as explained below</li>
+<li>For the root element, whitespaces are always trimmed</li>
+</ul>
+<p> </p>
+<p>
+  Whitespace trimming consists of removing all the leading and trailing whitespaces of each text node and concatenating all text node values. If the resulting value is empty, the text node is ignored. The rationale is to remove insignificant spaces created when formatting XML documents
+</p>
+
+
+<h1>Exceptions</h1>
+
+<h2>When an attribute has the same name of an element</h2>
+
+<p>If an element contains another element and an attribute which have the same name, the attribute name is prefixed with the "@" character</p>
+<pre class="code">
+XML:    &lt;root timezone="Europe/Paris">
+          &lt;timezone>Europe/Paris&lt;/timezone>
+        &lt;/root>
+JSON:   { "@timezone": "Europe/Paris", 
+          timezone: { "$": "Europe/Paris" } }
+</pre>
+
+
+<h1>Arrays of text elements</h1>
+<p></p>
+<pre class="code">
+  XML:    &lt;root>
+            &lt;item>One&lt;/item>
+            &lt;item>Two&lt;/item>
+          &lt;/root>
+  JSON:   { item: [
+            { $: "One" },
+            { $: "Two" }
+          ] }
+  </pre>
+  
+
+
+<h1>Compatibility notes</h1>
+
+<p>
+  A few gaps were identified regarding SimpleJson format when converting from XML to JSON. They were fixed in release 1.1.15 and may introduce some behavior changes for cases which were ambiguous
+</p>
+<p> </p>
+<ul>
+  <li>When XML element has both a text content and attributes or child elements</li>
+  <li>When root element has only text</li>
+  <li>When XML element has both an attribute and a child element with the same name</li>
+  <li>When array contains element which only have text</li>
+  <li>When XML is formatted (and therefore has insignificant whitespaces)</li>
+  <li>XML collections (i.e. when root node ends with “-collection”) would sometimes contain a JSON property “#text”</li>  
+</ul>

package/docs/transport.html

@@ -27,7 +27,7 @@ title: The Transport Protocol
 
 <p>If the request is successful, a promise is returned with the result payload, as a string.</p>
 
-<p>If the request fails, the promise is rejected with an error object with class <b>HttpError</b>, a litteral with the following attributes:</p>
+<p>If the request fails, the promise is rejected with an error object with class <b>HttpError</b>, a literal with the following attributes:</p>
 <ul>
     <li><b>statusCode</b> is the HTTP status code, such as 404, 500, etc.</li>
     <li><b>statusText</b> is the HTTP status text coming with the error</li>
@@ -39,7 +39,7 @@ title: The Transport Protocol
 <p>The transport can be overriden by using the <b>client.setTransport</b> call and passing it a transport function, i.e. an async function which</p>
 
 <ul>
-    <li>Takes a <b>Request</b> object litteral as a parameter</li>
+    <li>Takes a <b>Request</b> object literal as a parameter</li>
     <li>Returns a the request result in a promise</li>
     <li>Returns a rejected promise containing an <b>HttpError</b> in case of failure</li>
 </ul>

package/docs/xtkInterface.html

@@ -17,4 +17,4 @@ const delivery = client.NLWS.nmsDelivery.create({ label: "Hello" });
 await delivery.newInstance();
 </pre>
 
-<p>There are 2 common interfaces: <a href="{{ site.baseurl }}/xtkPersist.html">xtk:persist</a> and <a href="">xtk:jobInterface</a> which are documented below.</p>
+<p>There are 2 common interfaces: <a href="{{ site.baseurl }}/xtkPersist.html">xtk:persist</a> and <a href="{{ site.baseurl }}/xtkJob.html">xtk:jobInterface</a> which are documented below.</p>

package/docs/xtkJob.html

@@ -0,0 +1,131 @@
+---
+layout: page
+title: XTK Jobs (xkt:job, xkt:jobInterface)
+---
+<p>This section describes how long running jobs are handled in ACC and ACC api. There are 2 schemas for jobs</p>
+<ul>
+  <li>xtk:job which represents an acutal job, and which is implemented by several objects such as nms:delivery</li>
+  <li>xtk:jobInterface which contains methods to submit jobs and get their statuses. Other schemas do not actually implement the xtk:jobInterface interface</li>
+</ul>
+
+<p>In term of SOAP calls, running a job is simply executing a SOAP call synchronously or asynchronously in a separate thread or process. However, calls must be made on both the xtk:jobInterface and job entity schema. The SDK provides a helper interface to submit and handle jobs</p>
+
+<h1>Simple Jobs</h1>
+  
+<p>
+  Simple jobs are non-static methods with no parameters, such as the nms:delivery#Prepare method which is used to prepare a delivery. 
+  Such jobs can be executed synchronously (xtk:jobInterface#Execute) or asynchronously (xtk:jobInterface#Submit). Both methods return
+  a job id which can be used to retreive (poll) more about the job.
+<p>
+    
+<p>
+  First, we need to retreive a delivery, which is the object upon which to run the Prepare method, but also implements xtk:job interface. 
+  The Prepare method requires a complete delivery object, so we use SelectAll
+</p>
+
+<pre class="code">
+  const  queryDef = {
+    schema: "nms:delivery",
+    operation: "get",
+    select: { node: [ { expr: "@id" } ] },
+    where: { condition: [ { expr:`@internalName='DM19'` } ] }
+  }
+  const  query = NLWS.xtkQueryDef.create(queryDef);
+  await query.selectAll();
+  const delivery = await query.executeQuery();
+</pre>
+
+<p>
+  Now we can create a job for the Prepare method using the <b>client.jobInterface</b> function which returns a XtkJobInterface object. 
+  We pass it a job specification containing the schema, method, and object
+</p>
+
+<pre class="code">
+  const job = await client.jobInterface({
+    xtkschema: 'nms:delivery',
+    method: 'Prepare',
+    object: delivery
+  });
+</pre>
+
+<p>
+  Finally we can submit the job and get it's status, progress, result, etc.
+</p>
+
+<pre class="code">
+  await job.submit();
+  var status = await job.getStatus();
+</pre>
+
+
+<h1>Soap calls</h1>
+<p>
+  The <b>SubmitSoapCall</b> method allows to run more complex jobs with parameters. The principle is the same, for instance calling the <b>PrepareProof</b> method
+  which takes a boolean parameter
+</p>
+
+<pre class="code">
+  const job = await client.jobInterface({
+    xtkschema: 'nms:delivery',
+    method: 'PrepareProof',
+    object: delivery,
+    args: [ false ]
+  });
+</pre>
+
+
+<h1>Job Status</h1>
+<p>
+  The status of a job is made of 3 pieces of information
+</p>
+<ul>
+  <li>The job status code</li>
+  <li>Job logs</li>
+  <li>Job properties which are arbitrary key value pairs and also contain progress information</li>
+</ul>
+
+<p>
+  For convenience, the SDK provides a <b>getStatus</b> function which will fetch and return the full job status object with strong
+  typing. When getStatus is called several time, each call will fetch the most recent status, and new logs since the previous call.
+</p>
+
+<table>
+  <thead>
+    <tr><th>Status code</th><th>Description</th></tr>
+  </thead>
+  <tbody>
+    <tr><td><b>0</b></td><td>Being edited</td></tr>
+    <tr><td><b>2</b></td><td>Running: execution is in progress</td></tr>
+    <tr><td><b>3</b></td><td>Canceling: a cancel request was submitted and the job will be canceld as soon as possible</td></tr>
+    <tr><td><b>4</b></td><td>Canceled</td></tr>
+    <tr><td><b>5</b></td><td>Finished: the job has finished successfully and getResult can be called</td></tr>
+    <tr><td><b>6</b></td><td>Error: the job failed. Details about the error can be found in the logs</td></tr>
+    <tr><td><b>7</b></td><td>Pause pending: a pause request was submitted and the job will be paused as soon as possible</td></tr>
+    <tr><td><b>8</b></td><td>Pause: the job is paused</td></tr>
+    <tr><td><b>9</b></td><td>Purge pending: a purge request was submitted and the job will be purged as soon as possible</td></tr>
+  </tbody>
+</table>
+
+
+<h1>Job Progress</h1>
+<p>
+  The progress of a job is available as 2 integers: current and max which represent the current progression and max value for progression.
+  Both values are returned by the getStatus call as properties (i.e. <b>properties.progress.current</b> and <b>properties.progress.max</b>).
+  The SDK provides the <b>getProgress</b> function to retreive the progress as a percentage (and a valid number)
+</p>
+
+<pre class="code">
+  // Returns progress as a percentage in the [0..1] range
+  var progressPercent = job.getProgress();
+</pre>
+
+
+<h1>Job Result</h1>
+<p>
+  Typically, a client would call the getStatus method every few seconds to get updates on the progress. When a job is finished and successfull,
+  one can get the result using the <b>getResult</b> function
+</p>
+
+<pre class="code">
+  var result = await job.getResult();
+</pre>

package/docs/xtkQueryDef.html

@@ -74,7 +74,7 @@ const extAccount = await query.executeQuery();
 
 
 <h1>Escaping</h1>
-<p>It's common to use variables in query conditions. For instance, in the above example, you'll want to query an account by name instead of using the hardcoded <b>ffda</b> name. The <b>expr</b> attribute takes an XTK expression as a parameter, and <b>ffda</b> is a string litteral in an xtk expression.</p>
+<p>It's common to use variables in query conditions. For instance, in the above example, you'll want to query an account by name instead of using the hardcoded <b>ffda</b> name. The <b>expr</b> attribute takes an XTK expression as a parameter, and <b>ffda</b> is a string literal in an xtk expression.</p>
 
 <p>To prevent xtk ingestions vulnerabilities, you should not concatenate strings and write code such as expr: "@name = '" + name + "'": if the value of the name 
   parameter contains single quotes, your code will not work, but could also cause vulnerabilities.

package/package-lock.json

@@ -1,6 +1,6 @@
 {
   "name": "@adobe/acc-js-sdk",
-  "version": "1.1.14",
+  "version": "1.1.16",
   "lockfileVersion": 1,
   "requires": true,
   "dependencies": {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@adobe/acc-js-sdk",
-  "version": "1.1.14",
+  "version": "1.1.16",
   "description": "ACC Javascript SDK",
   "main": "src/index.js",
   "homepage": "https://github.com/adobe/acc-js-sdk#readme",

package/src/campaign.js

@@ -28,7 +28,7 @@ const { Util } = require("./util.js");
   static INVALID_CREDENTIALS_TYPE(type, details)          { return new CampaignException(undefined, 400, 16384, `SDK-000000 Invalid credentials type '${type}'`, details); }
   static CANNOT_GET_CREDENTIALS_USER(type)                { return new CampaignException(undefined, 400, 16384, `SDK-000001 Cannot get user for Credentials of type '${type}'`); }
   static CANNOT_GET_CREDENTIALS_PASSWORD(type)            { return new CampaignException(undefined, 400, 16384, `SDK-000002 Cannot get password for Credentials of type '${type}'`); }
-  static INVALID_CONNECTION_OPTIONS(options)              { return new CampaignException(undefined, 400, 16384, `SDK-000003 Invalid options parameter (type '${typeof options}'). An object litteral is expected`); }
+  static INVALID_CONNECTION_OPTIONS(options)              { return new CampaignException(undefined, 400, 16384, `SDK-000003 Invalid options parameter (type '${typeof options}'). An object literal is expected`); }
   static INVALID_REPRESENTATION(representation, details)  { return new CampaignException(undefined, 400, 16384, `SDK-000004 Invalid representation '${representation}'.`, details); }
   static CREDENTIALS_FOR_INVALID_EXT_ACCOUNT(name, type)  { return new CampaignException(undefined, 400, 16384, `SDK-000005 Cannot created connection parameters for external account '${name}': account type ${type} not supported`); }
   static BAD_PARAMETER(name, value, details)              { return new CampaignException(undefined, 400, 16384, `SDK-000006 Bad parameter '${name}' with value '${value}'`, details); }

package/src/client.js

@@ -36,6 +36,7 @@ const request = require('./transport.js').request;
 const Application = require('./application.js').Application;
 const EntityAccessor = require('./entityAccessor.js').EntityAccessor;
 const { Util } = require('./util.js');
+const { XtkJobInterface } = require('./xtkJob.js');
 const qsStringify = require('qs-stringify');
 
 /**
@@ -173,16 +174,7 @@ const clientHandler = (representation, headers, pushDownOptions) => {
                         return callContext;
 
                     // get Schema id from namespace (find first upper case letter)
-                    var schemaId = "";
-                    for (var i=0; i<namespace.length; i++) {
-                        const c = namespace[i];
-                        if (c >='A' && c<='Z') {
-                            schemaId = schemaId + ":" + c.toLowerCase() + namespace.substr(i+1);
-                            break;
-                        }
-                        schemaId = schemaId + c;
-                    }
-                    callContext.schemaId = schemaId;
+                    callContext.schemaId = Util.schemaIdFromNamespace(namespace);
 
                     const caller = function(thisArg, argumentsList) {
                         const callContext = thisArg["."];
@@ -546,16 +538,14 @@ const fileUploader = (client) => {
                     }
                     const data = new FormData();
                     data.append('file_noMd5', file);
-                    //TODO: Needs to be refactored after cookie issue get resolved.
                     client._makeHttpCall({
                         url: `${client._connectionParameters._endpoint}/nl/jsp/uploadFile.jsp`,
                         processData: false,
-                        credentials: 'include',
                         method: 'POST',
                         data: data,
                         headers: {
-                            'x-security-token': client._securityToken,
-                            'Cookie': '__sessiontoken=' + client._sessionToken,
+                            'X-Security-Token': client._securityToken,
+                            'X-Session-Token': client._sessionToken,
                         }
                     }).then((okay) => {
                         if (!okay.startsWith('Ok')) {
@@ -1475,11 +1465,21 @@ class Client {
         const result = [];
         const schemaId = callContext.schemaId;
 
-        var schema = await that.getSchema(schemaId, "xml", true);
+        var entitySchemaId = schemaId;
+        if (schemaId === 'xtk:jobInterface')
+            entitySchemaId = callContext.entitySchemaId;
+
+        // Get the schema which contains the method call. Methods of the xtk:jobInterface interface are handled specificaaly
+        // because xtk:jobInterface is not actually a schema, but just an interface. In practice, it's used as an xtk template
+        // rather that an xtk inheritance mechanism
+        const methodSchemaId = schemaId === 'xtk:jobInterface' ? 'xtk:job' : schemaId;
+        var schema = await that.getSchema(methodSchemaId, "xml", true);
         if (!schema)
             throw CampaignException.SOAP_UNKNOWN_METHOD(schemaId, methodName, `Schema '${schemaId}' not found`);
         var schemaName = schema.getAttribute("name");
-        var method = that._methodCache.get(schemaId, methodName);
+
+        // Lookup the method to call
+        var method = that._methodCache.get(methodSchemaId, methodName);
         if (!method) {
             // first char of the method name may be lower case (ex: nms:seedMember.getAsModel) but the methodName 
             // variable has been capitalized. Make an attempt to lookup method name without capitalisation
@@ -1489,13 +1489,16 @@ class Client {
         }
         if (!method) {
             this._methodCache.put(schema);
-            method = that._methodCache.get(schemaId, methodName);
+            method = that._methodCache.get(methodSchemaId, methodName);
         }
         if (!method)
             throw CampaignException.SOAP_UNKNOWN_METHOD(schemaId, methodName, `Method '${methodName}' of schema '${schemaId}' not found`);
-        // console.log(method.toXMLString());
 
-        var urn = that._methodCache.getSoapUrn(schemaId, methodName);
+        // Compute the SOAP URN. Again, specically handle xtk:jobInterface as it's not a real schema. The actual entity schema
+        // would be available as the entitySchemaId property of the callContext
+        var urn = schemaId !== 'xtk:jobInterface' ? that._methodCache.getSoapUrn(schemaId, methodName)
+            : `xtk:jobInterface|${entitySchemaId}`;
+
         var soapCall = that._prepareSoapCall(urn, methodName, false, callContext.headers, callContext.pushDownOptions);
 
         // If method is called with one parameter which is a function, then we assume it's a hook: the function will return
@@ -1512,13 +1515,13 @@ class Client {
             if (!object)
                 throw CampaignException.SOAP_UNKNOWN_METHOD(schemaId, methodName, `Cannot call non-static method '${methodName}' of schema '${schemaId}' : no object was specified`);
 
-            const rootName = schemaId.substr(schemaId.indexOf(':') + 1);
+            const rootName = entitySchemaId.substr(entitySchemaId.indexOf(':') + 1);
             object = that._fromRepresentation(rootName, object, callContext.representation);
             // The xtk:persist#NewInstance requires a xtkschema attribute which we can compute here
             // Actually, we're always adding it, for all non-static methods
             const xmlRoot = object.nodeType === 9 ? object.documentElement : object;
             if (!xmlRoot.hasAttribute("xtkschema"))
-                xmlRoot.setAttribute("xtkschema", schemaId);
+                xmlRoot.setAttribute("xtkschema", entitySchemaId);
             soapCall.writeDocument("document", object);
         }
         const parametersIsArray = (typeof parameters == "object") && parameters.length;
@@ -1527,7 +1530,7 @@ class Client {
             var param = DomUtil.getFirstChildElement(params, "param");
             var paramIndex = 0;
             while (param) {
-                   const inout = DomUtil.getAttributeAsString(param, "inout");
+                const inout = DomUtil.getAttributeAsString(param, "inout");
                 if (!inout || inout=="in") {
                     const type = DomUtil.getAttributeAsString(param, "type");
                     const paramName = DomUtil.getAttributeAsString(param, "name");
@@ -1570,9 +1573,9 @@ class Client {
                             // Hack for workflow API. The C++ code checks that the name of the XML element is <variables>. When
                             // using xml representation at the SDK level, it's ok since the SDK caller will set that. But this does
                             // not work when using "BadgerFish" representation where we do not know the root element name.
-                            if (schemaId == "xtk:workflow" && methodName == "StartWithParameters" && paramName == "parameters")
+                            if (entitySchemaId == "xtk:workflow" && methodName == "StartWithParameters" && paramName == "parameters")
                                 docName = "variables";
-                            if (schemaId == "nms:rtEvent" && methodName == "PushEvent")
+                            if (entitySchemaId == "nms:rtEvent" && methodName == "PushEvent")
                                 docName = "rtEvent";
                             // Try to guess the document name. This is usually available in the xtkschema attribute
                             var xtkschema = EntityAccessor.getAttributeAsString(paramValue, "xtkschema");
@@ -1636,7 +1639,7 @@ class Client {
                         else if (type == "DOMDocument") {
                             returnValue = soapCall.getNextDocument();
                             returnValue = that._toRepresentation(returnValue, callContext.representation);
-                            if (schemaId === "xtk:queryDef" && methodName === "ExecuteQuery" && paramName === "output") {
+                            if (entitySchemaId === "xtk:queryDef" && methodName === "ExecuteQuery" && paramName === "output") {
                                 // https://github.com/adobe/acc-js-sdk/issues/3
                                 // Check if query operation is "getIfExists". The "object" variable at this point
                                 // is always an XML, regardless of the xml/json representation
@@ -1794,11 +1797,10 @@ class Client {
                 url: `${this._connectionParameters._endpoint}/report/${callContext.reportName}?${encodeURI(`_noRender=true&_schema=${callContext.schema}&_context=${callContext.context}&_selection=${callContext.selection}`)}&_selectionCount=${selectionCount}`,
                 headers: {
                     'X-Security-Token': this._securityToken,
-                    'Cookie': '__sessiontoken=' + this._sessionToken, 
+                    'X-Session-Token': this._sessionToken,
                     'Content-Type': 'application/x-www-form-urlencoded'
                 },
                 method: 'POST',
-                credentials: 'include',
                 data : qsStringify(callContext.formData)
             };
             
@@ -1871,6 +1873,15 @@ class Client {
         const result = this._toRepresentation(doc);
         return result;
     }
+
+    /**
+     * Creates a Job object which can be used to submit jobs, retrieve status, logs and progress, etc.
+     * @param {Campaign.XtkSoapCallSpec} soapCallSpec the definition of the SOAP call
+     * @returns {Campaign.XtkJobInterface} a job
+     */
+    jobInterface(soapCallSpec) {
+        return new XtkJobInterface(this, soapCallSpec);
+    }
 }