Posts Tagged ‘integration’

Improve Document Management in SharePoint with jQuery

May 16, 2010 1 comment
Today I will present a project I did with a Human Ressources consulting company, ISMAT, for who I integrated a document management process in SharePoint Online. This post covers the context and describes how the developed artifact – a copy operation plugin for SharePoint Online – works. The innovative part of this plugin is that it combines the use  of multiple data sources (SharePoint Web Services and HTML source code) while offering end-user controls that are integrated inside the SharePoint GUI. It was developed with the jQuery JavaScript library. The project described here was also part of a report published by the SharePoint Competency Center, and can be downloaded here: Microsoft SharePoint : on-premise or online?.

While implementing a SharePoint document management system for ISMAT, I had to customize it to meet the business needs. Indeed, in order to sustain an internal business process, an extension was written in JavaScript and jQuery which worked with the SharePoint Online Web Services. This extension allowed ISMAT to copy multiple documents to clients’ folders, while adding business metadata along the way. This allowed collaborators to get copies of generic teaching material ready to be personalized for each customer. It was entirely integrated into the user interface of SharePoint Online, which proved the extensibility of the service and its ability to sustain ISMATs’ particular business needs (see Picture 1: Screen capture of the Documents Copy Plugin in action).

The objective was to automate a part of the clients’ document management process that was entirely manual before. Indeed, it was the source of classification problems like duplicates or lost information. The implementation of this plugin had four main advantages for ISMAT:

  • Increase of productivity by reducing the number of manual actions to perform by the user to customize teaching materials for a customer.
  • Decreased risks linked to misfiled documents, like information loss or duplicates, thanks to the automatic mapping of customers’ metadata.
  • Less training necessary for the end user, as the process is fully integrated into SharePoint Onlines’ user interface.
  • Minimized maintenance: the plugin was designed to be easily installable, configurable and portable. Indeed, just a line in a content editor web part is necessary to enable it, and the plugin requires almost no configuration. Furthermore, it is entirely cross browser compatible and can’t possibly crash SharePoint Onlines’ servers as the code is executed on the client side.
The plugin works like this: Checkboxes are added near the documents displayed in a Document Library WebPart. They use HTML structural patterns that are in Document Library WebParts to detect the place where to insert the UI elements (see schema below: How UI elements are injected).
Clicking on any of those checkboxes generates a new “Copy to” button that appears in the Web Part’s toolbar. When the user moves his mouse over this new button, a list of clients is revealed in a drop down menu. These clients all come from a list of clients that resides in the same SharePoint Site collection. Each time the page is loaded, the plugin fetches the updated list of clients.
Once the user clicks on a client name in the menu, the selected documents are copied to a clients’ library that resides on another site. These documents are updated with business metadata along the way. Another particularity of this plugin is that it uses SharePoint Onlines’ Web Services as well as structural HTML patterns as data sources (see picture below: jQuery plugin for SharePoint bidirectional interactions).

This alternative way of extending SharePoint Onlines’ Standard functionalities proves to be particularly interesting for small companies like ISMAT, who do not have enough employees to be able to benefit from the dedicated variant of the service which allows custom code deployment.

Using external business data with SharePoint Online Part 1: Scenario and conception of a solution

January 17, 2010 2 comments

The first project I would like to put forward on this blog concerns the integration of external business data with Microsoft SharePoint Online. This theme concerned a part of my Bachelor’s thesis that I did last summer which was named “Electronic Documents Management Optimization In A Consulting Firm”.  I will share the parts that I found the most interesting and challenging, in my opinion.


First of all I would like to put some context on the topic that I will cover. The company I worked for was a small “deskless” RH consulting business that managed all of its information (addresses, meetings, projects, etc) through the web, on a custom built PHP-MySQL CRM (Customer Relationship Management) server. It was hosted by a local ISP on a linux server.

It soon became apparent that such a CRM wouldn’t be able to support a centralized document management system, which the company increasingly needed. Consequently, they had to find a solution available on the market.

The company finally settled for Microsoft’s cloud based solution, SharePoint Online, which is part of the Business Productivity Online Suite, to manage their documents. It turned out to be perfectly suited for such a small deskless company. It offered great functionality fora very low cost (thanks to the Software As A Service model, where you pay only for what you use), and low maintenance once set up.

Conception of a solution

One of the challenges of SharePoint Online is that there isn’t any obvious way to integrate external business data from a non-Microsoft source.

Abstract schema of a CRM, Connector and SharePoint Online interacting

This was however a very important point in my project, because alot of the company’s documents had to be organized by client’s names. And guess what: the clients names had to be the same in the CRM’s MySQL database and a SharePoint Online’s list.

But thanks to SharePoint Online’s Web Services, it was possible to build a connector in Java that that made the link between the CRM and SharePoint.

It is important to note that the connector can’t be actioned from within SharePoint Online, as it is impossible at this time to deploy custom code on SharePoint’s cloud-based server. Therefore, clients information would need to be entirely manipulated from within the CRM (and therefore the connector), and the SharePoint clients list protected from users manual access.

Java was chosen because it could directly be deployed on the company’s rented LAMP server.

The point here is to show you that if don’t have a Microsoft infrastructure (Access database, Windows OS that can deploy C# code), but would like to integrate existent business data with a low-priced SharePoint Online solution, it is possible. I’ll use PHP and Java as examples.

How it works

The objective was to have the same names of the clients in the CRM and in a SharePoint Online’s clients list. The designed solution worked like this:

SharePoint Online Connector interaction with external business data

  1. A collaborator types in a new client’s information in an HTML form on the CRM, from his computer.
  2. The Apache server, which is running PHP, adds the client’s information in the local MySQL database, while sending the same info to a “Java to SharePoint connector” running on Tomcat (this is done with a PHP-Java Bridge).
  3. The Java connector processes the received data and encapsulates it into an XAML request that is sent SharePoint Online’s ASMX 2.0 Lists Web Service.
  4. The data is added in the list (which name was detailed in a properties file used by the connector). The list in SharePoint Online is the most restricted as possible for SharePoint Users: the whole point of building a connector is lost if clients information is manipulated from more than one place, for the simple reason that SharePoint Online can’t push those information outside of its walls!
  5. SharePoint Online sends the result back to the connector, which is then passed to PHP, which in turn displays the result to the end user.

The Java code used to build the connector and the implementation of the bridge with PHP will be described in further posts. Your comments are welcome!