I’ve downloaded and experimented with ChainBuilder ESB recently and wanted to relay my experiences to the Mule community. Caveat: While I am biased in favour of Mule, I did try to keep an open-mind while working with it …

1) When installed, you can configure the ESB by launching its IDE. I was curious to see what the IDE looked like as I am ambivalent on the subject of IDEs for ESBs. On the one hand, you need a certain level of control that the raw XML gives you but on the other hand, certain simple and common tasks really should be automated. I was disappointed to see that the Chainbuilder IDE includes Eclipse. Not that this is a bad thing, mind, but I already have Eclipse set up and configured so why can’t I just have a few plug-ins for my existing installation?

2) The IDE is slow. Maybe it was just my machine but regular Eclipse worked like a charm so I’m not too sure that it’s the hardware.

3) I followed the Getting Started Guide to set up a simple example. (Aside: Why is it a PDF? Surely simple HTML is faster and easier to use … ?)

4) Once you set up projects, the first thing that you notice is a large number of source folders for your project. I like the idea of having these auto-created as it helps organise the novice but that does not mean that you’ll know what each folder is there for until you actually need it.

5) Under the source “formats” folder, there are a few default items that you can work with natively. One of these is HL7 – and given the various number of formats, it comes with 6 of them provided for you. That’s a cool option. In Mule, it’s the equivalent of having a transformer that can convert HL7 data for you and most of these items are available on the MuleForge rather than the actual product.

6) Using the Component Flow Editor, I was able to customise how I wanted to read files. In Mule I would setup the file connector and modify the autoDelete and streaming attributes as well as the endpoint itself. Here, I had a wizard which presented me with the same options. The resultant XML is easy to understand and, as a novice with the tool, the wizard helped me out with the settings I would want to configure. I like this.

<fileConsumer sourceDir="C:/Documents and Settings/Antoine/Desktop"
 stageDir="C:/Documents and Settings/Antoine/Desktop/stage"
 holdDirectory="C:/Documents and Settings/Antoine/Desktop/hold">
      <fileCompletion/>
      <Scheduler/>
      <fileReply 
replyDir="C:/Documents and Settings/Antoine/Desktop/reply"
 replyFilePattern=""/>
    </fileConsumer>

7) The XML is not colour coded. I know it’s a matter of installing the appropriate plugin into this customised version of Eclipse but surely this is something that could be done out of the box? It’s as if using ChainBuilder is meant to be through the visual interface only. I don’t mind that, but I want to be able to control things myself too

Verdict – All in all, the sight and smell of an IDE and colourful moving parts is exciting at first but the number of buttons and clickable options means that I spent more time figuring out what the interface was supposed to do than actually building something. Maybe I’m just too used to XML or maybe I expected hand-holding to be available only if I need it rather than the other way round. On the plus side though, having several data formats available for easy transformation does make life easier if you work with them.

If you enjoyed this post, make sure you subscribe to my RSS feed!

Tags: ChainBuilder, comparison, newbie

This entry was posted on Monday, July 20th, 2009 at 10:00 AM and is filed under Java, Mule. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.