Release Notes

Day JCR Connector for MS SharePoint, Version 2.0.0

The release cycle for this release of Day JCR Connector for MS SharePoint (SharePoint connector) started on July 23, 2008, went through 18 iterations of development, quality assurance and bug fixing, and ended on September 24, 2009. 

Day JCR Connector for MS SharePoint 2.0.0 is generally available since September 28, 2009. This is a First Customer Shipping (FCS) release of this version of the connector.

SharePoint connector 2.0 is a major release launching a new era in Day JCR connector technology: "Connector 2.0". This release provides a new, streamlined internal repository architecture, write-enables the connector by supporting JCR Level 2 features, and supports the highest simplicity of installation, administration, and development via integrated product package based on Day Quickstart technology and CRX platform.

Being a key defining feature of the Day JCR connector products, this release provides support of the Content Repository for JavaTM technology API (JCR) standard which is defined by Java Specification Request 170 (JSR 170). The connector JCR-enables and Java-enables Microsoft SharePoint repositories, and allows applications to get access to valuable content, metadata, and content services managed in SharePoint repositories through a standardized API.

This release is a major milestone in Day’s JCR connector product strategy. The connector code base has been completely rewritten and based on a new internal architecture, based on Apache Jackrabbit SPI architecture. This new implementation allowed for achieving a clean and simplified modular internal architecture, and in effect for adding the JCR Level 2 features (write capability) to the supported JCR functionality. Product packaging has been also completely redesigned to match the CQ5 and CRX Quickstart-based and OSGi-based enterprise deployment capabilities.

This version combines in a unique way leading technology and architecture with ease of use for developers and system administrators.

The new internal architecture is based on the clean SPI API. The SPI persistence server component, which communicates with the SharePoint server, is stateless. This allows for dynamic instantiation of different JCR sessions connecting to different repositories by one connector instance, without the need to maintain persisted item state information.

The connector core provides a full JCR repository implementation, JCR-enabling the SharePoint server, and no longer requires CRX to run.

The product is packaged in CRX platform for simplifying development of applications using the connector, as well as installation and integration with other systems.

The connector is delivered as a Quickstart distribution, known from CQ5 and CRX. It makes the connector installation a breeze.

A GUI for configuring the connector instance is provided. In addition, the configuration can be done through updates to a special content node, as well as in the OSGi console.

The packaging of the connector core has been greatly enhanced. The core connector repository implementation is packaged into an OSGi bundle. Additionally, the bundle is wrapped into CRX / CQ5 Content Package. This packaging allows for the following deployments:

• Deployment into OSGi platforms (the case for CRX and CQ5). Connector repository can be managed and accessed using standardized OSGi APIs.
• Binding to JNDI. By default, the connector's repository object is bound to the CRX JNDI and available to JNDI-enabled applications.   
• Deployment into any Java system as a library. OSGi bundle can be used like a normal Java library, and the repository implementation instantiated programatically.

Additionaly, packaging into Content Package allows for very easy deployment into and portability between all Day applications (based on CRX / CQ5 platform).

The connector installed into CRX or CQ5 framework supports various user authentication configuration for flexibility in development and production deployments. The default, preconfigured SharePoint Login Module provides a transparent authentication with SharePoint credentials for the whole CRX connector instance. The standard CRX LDAP Login Module can be used to configure authentication against the common directory LDAP server.

Additionally, the connector provides an extensible API to allow for implementing custom authentication modules. This allows for extending the connector in projects to support SharePoint servers with customized authorization setups.

Our design goal for the release was to remove any dependency of the connector on custom, SharePoint-Side components, enhancing the native SharePoint API. The connector now only uses the native SharePoint web services API, and does not require installation of server-side components.

Note that this goal was only possible to realize in the connector after substantial improvements in SharePoint 2007 web services API (in the area of search and notification of changes) with regard to the older SharePoint 2003 version.

The benefits are:

• Ease of deployment and configuration. Note that in many cases IT policies disallow installation of any custom components on enterprise servers, so this broadens the application of the connector to these cases.
  
• Minimizing cost of IT operations of systems using the connector.
• Easier maintenance of the connector for future updates of SharePoint technologies
• Relying fully on scalability of the native platform. Custom server-side components can sometimes provide different scalability characteristics than the native server implementation, which might lead to problems.

The product documentation has been rewritten and enhanced to cover the new connector features and characteristics.

It has been also enhanced to provide more conceptual overview of the connector architecture and functionality, as well as more in-depth explanation of product features.

All the above-mentioned enhancements were carefully designed and implemented, based on our own experience with the connector, as well as feedback from customers and partners. We kept in mind, as the goal and driving factor, the ease of use and efficiency for users of the product - developers, integration engineers, administrators, and finally end users, who are empowered to leverage and reuse the content stored in various different systems through integrated, streamlined user experience. We believe that we managed to achieve this goal.

Please note that the CRX used by the core connector JCR repository implementation as a platform (deployment, GUI, etc) is embedded in the product as an OEM module. CRX cannot be used independently of the connector module, or to provide the application with CRX core functionalities (like using the native JCR repository to manage content) without obtaining a separate license for it. The same holds for separately licensed CRX modules, like LDAP module or Virtual Repository.

If you plan to use the CRX embedded in the connector product in such a way, please obtain the respective license for CRX and the separately licensed CRX modules. Please contact Day Sales Group or your account manager at Day to clarify the licensing requirements.

If you plan to use the CRX repository in the connector in such a way for non-productive usage, please obtain a free CRX Developer license on www.day.com/downloadcrx.

The design principle of the JCR connectors is that they map content structures and available content services of the external repositories they support. The mapping takes into account only those content artifacts and services, which are relevant and defined in the JCR API. Additionally, the mapping covers the areas, which are available through used client libraries or protocols of the underlying platform.

For the SharePoint connector it means, that the connector maps the content and services to the extent supported by the native SharePoint web services, as the connector does not rely on custom additional server-side components.

This has a few important consequences:

• Sharepoint 2003 is no longer supported as it does not provide sufficient API support for search and observation
• Fulltext search only returns items, which are indexed by SharePoint server
• Set of searchable columns is limited
• A fixed set of node types is supported in searches
• Some searches require additional network roundtrip as connector does not receive the full information necessary to map results to JCR from the web services
• Observation is supported for SharePoint lists
  
• Observation must maintain some metadata in memory, as the information returned from web services would not allow for JCR compatible observation implementation. This results in growing memory requirements if too many lists are observed. The observable lists can be configured in the connector to manage this
• There is no guarantee of order of observation events

Please see product documentation for additional information regarding the JCR mapping details.

To upgrade the previous version of the connector, please note the configuration parameters, uninstal the previous version, install the new version, and apply the configuration to the new version.

Since the connector does not persist any content (apart from potential caches, which will be recreated), you don't have to upgrade or migrate any content.

This version of the connector is based on a completely reimplemented code base, so the list of code enhancements and bug fixes is not relevant.

Listed below are minor enhancements and fixes to the JCR mapping of the content (node naming, node types):

• Files and folders stored in document libraries are now represented as nt:file and nt:folder nodes, respectively. The SharePoint specific metadata for these items is added to these nodes through a mixin node type. This simplifies access to the mapped content to applications, which work with nt:folder/nt:file standard nodes
  
• Node names for those SharePoint artifacts, which are named by users (file / folder names, etc) is no longer prefixed with sp: prefix. This makes the mapping more natural.
• The mapping was extended to cover new types available in SharePoint 2007
• Some existing node types were modified according to the official Microsoft documentation of the SharePoint web services, released by Microsoft to comply with EU requirements during the connector development

Public Sites

• http://www.day.com/
• http://www.day.com/crx
• http://jcr.day.com/
• http://dev.day.com/
• http://dev.day.com/kb

Restricted Sites

• http://daycare.day.com/
• http://docs.day.com/
• http://dev.day.com/docs

E-Mail Addresses

• information@day.com
• oemsales@day.com

Copyright © 1993-2009 Day Management AG, Switzerland. All rights reserved.

The Communiqué® Software and the CRX Content Repository Extreme® Software are protected by copyright law and international treaties and are licensed (not sold). Their use is subject to a valid license agreement.

Any unauthorized use, copying or distribution may result in severe criminal or civil penalties and will be prosecuted to the maximum extent allowed by law.

DAY, the DAY logo, COMMUNIQUÉ, CONTENTBUS and CRX CONTENT REPOSITORY EXTREME are registered trademarks and service marks, or are trademarks and service marks, of Day Management AG in various countries around the world.

All other product names and company logos mentioned in the information, documents or other items provided or available herein may be the trademarks of their respective owners.