Wednesday, April 30, 2014

Resco javascript Fetch all account example

I have recently been working on a Proof-of-concept for a customer with the mobile platform. It is the first time I have been working deeply with it and I really like it. The woodford customization tool is very powerful and easy to use. It also allows customizers to add custom components such as offline html and js-components that can be used to extend the built in functionality where the standard forms do not meet the customer’s requirements.

In this case my customer wanted me to prove that it was possible to create an overview screen for an account showing lots of related information on a single screen. As Resco does not intrinsically support subgrids, like Microsoft Dynamics CRM does, you have to create this with a custom html which loads this data from the offline Resco storage. In theory this shouldn’t be too hard. It proved to be a bit more of a challenge, especially since the Resco Woodford manual contained some errors why I thought it might be a good idea to publish an example of how a complete code example of how to fetch a list of accounts wrapped in a working html-page that can just be copy-pasted into a file and tested out. This was one of the issues I had as just getting a simple base-line to work from was quite a lot of work.

So, without any more bells and whistles, here is the code:

<!DOCTYPE html>
    <title>Account List</title>
<script type="text/javascript" src="JSBridge.js"></script>
<script type="text/javascript">

function writeError(text) {
var error = document.getElementById("errortext");
error.innerHTML = error.innerHTML + "
+ text;

function callbackAccountsLoaded(res)
    writeError("In callbackAccountsLoaded")
function callbackAccountsFailure(res)

function onAccountLoad() {
    try {

        var account = new MobileCRM.FetchXml.Entity("account");
        account.filter = new MobileCRM.FetchXml.Filter();
        account.filter.where("statuscode", "eq", 1); // which have “Active” status

        var fetch = new MobileCRM.FetchXml.Fetch(account);
        fetch.execute("Array", function (res) { callbackAccountsLoaded(res); }, function (err) { callbackAccountsFailure(err); }, null);
    catch (e)
        errortext("Error: " + e.message);
<body onload="onAccountLoad();">
<div style="height:300px;" id="errortext"></div>

Some of the important points here are:

1.      account.filter.where needs to be used instead of using account.where directly as indicated in the example in the Woodford manual. It also needs to be instantiated with the “MobileCRM.FetchXml.Filter();”

2.      The fetch.execute is referencing two callback functions, one for success and one for failure. Make sure these references exist and have the correct number of arguments.

3.      I have not found any decent way of debugging javascripts within Resco. I would be more than happy to be corrected here. Using alerts will not work on all platforms (Resco indicates alerts do not work on Windows 8 RT for instance). Hence I have created the errortext-method which outputs text in a log-file manner to a div at the end. Make sure that the height is enough.

4.      To be able to load information on the currently edited object, if the iFrame is shown in the context of an account for instance, then MobileCRM.UI.EntityForm.requestObject() has to be used.

5.      The first parameter of fetch.execute indicates how the result is to be formatted. In this example I have chosen “Array”, but there are other choices like “JSON” and “XML” as well. Please refer to the Woodford manual for more information on this.

6.      Do try to use try-catch around a lot of this as any errors will otherwise just not show when run. If there is an error, you can display the message and at least get an idea of what was wrong.

7.      When developing, I like to use Windows 8/8.1 and maybe a CRM Online/IFD environment as that will allow me to install the Windows 8 version of the Resco CRM to try it out. The different platform clients for Resco are 95% identical, some UI differences and I have seen some other differences as well, but in general, it will speed up development a lot and you will only have to do the final development tests on the iPad/Android device.

If I get a lot of interest in this, I will publish more on this subject. As mentioned earlier, I havn’t found that much community content on Resco, the mobile client I currently feel is the premier choice, as the Microsoft Mobile client is a bit simpler in its design, features and also customizabilities.

Gustaf Westerlund
MVP, CEO and owner at CRM-konsulterna AB

Tuesday, April 22, 2014

Server side Sync Trouble shooting

I have been working with a Proof-of-concept for the lates couple of Days and have been setting up the server side sync for a CRM on-premise system. First of all I would just like to share that it is a really nice new feature, I really like the fact that we now can have an option to not have to use the email router. With the new spring release, it will also be available for CRM Online, but only CRM-Online to Exchange Online (O365) not hybrid Environments (ex. CRM Online to Exchange on-premise or vice versa).

However, I did run into some issues as none of the emails that I sent to the email account connected to the queue showed up in the queue.

The Server Sync is handled by the MSCRMAsyncService and if you switch on logging in CRM (which can be done using adding the registry key "TraceEnabled" and setting it to 1 or using Power Shell)  you will get the logging of the server sync in the async tracing loggs. It turned out that I had a missmatch in email addresses which CRM apparently didn't like and you can only have one email address for a queue, but it was easily fixed once I knew the problem. I just wonder how I would troubleshoot this if I got this in CRM Online... hopefully MSFT will be adding additional logging to the server sync functionality.

Gustaf Westerlund
MVP, CEO and owner at CRM-konsulterna AB