Source Code Archiving

Stank1964

Well Known Member
Okay, so it's 2013. We've all probably been using E1 for ever. Can someone tell me if there are any options to archive the source code as objects get promoted? I've never seen a CMS that doesn't do this. We have a graet thing, okay maybe it's not so great, in OMW, but I really need to see the versions of the code and how the code has changed historically. Especially when it's darn near impossible to clearly comment code changes in a way that makes the change clear.

Any thoughts?

Thanks to one and all!
 
Bruno,

This is great information, but what is Clear Case and how can I get some info about this? Is this the IBM Rational product?

Thanks!

Stan
 
It's fairly standard to use an OMW backup pathcode (This is setup in OMW). Essentially when we promote code from PY to PD, the PD objects are archived into the BK910 pathcode and then the PY objects are moved to PD. In case there is a code issue/bug we can retrieve code from BK910 and restore. This is all automated (one time setup by CNC) and that's it. In this scenario there is no need for any 3rd party system's...
 
Wow, that is really interesting. Never knew anything like that was possible. This appears to be "out-of-the-box" supported by Oracle??? In other words, not a 3rd party add-on. I wonder if it can be configured to work with MS Team Foundation Server or is it only supported with ClearCase? TFS is what we use for all or non-JDE code and it would be nice to actually have true version control of some kind.

It looks like though, that the only true actual source that gets stored in version control is the C code. For ER/specs etc. it looks like it just stores the equivalent of printing out the code so you couldn't truly retrieve/get Pxxxx specs from n number of versions ago and load into FDA, etc. Still better than nothing, and most of our business logic is encapsulated in C code anyway.

As an aside. We do have our OMW activity rules set that when we promote something to the PD path code it backs off the current version of the object to PDSave. So at any point in time we can go back at least one version for any given object. However we lose all visibility to past versions beyond that.
 
Yes, it's IBM Rational.
It's an expensive product.
I started some years ago to work with Jira,a very affordable product for project management, issue tracking, collaborative work and others stuff. It play well with txt file and could be well integrated with JDE in order to get a good control over software version.

If someone is interested to collaborate in order to build a solution available to all JD Edwards ecosystem (for free) based on Jira I'm available to invest my time setting up a web space for sharing and lead the project.
 
Yes, it prints the source code into text file, and it's great, you can store, compare thousand of versions, indexing the source and search for a specific string in millisecond (I have done this configuring Splunk for monitoring all kind of JDE logs and configurations file).
 
Back
Top