[KRRM-76] Rewrite eDocLite to use the KRAD Created: 29/Sep/09 Updated: 05/Apr/13
|Project:||Kuali Rice Roadmap|
|Fix Version/s:||Unspecified Future Release|
|Reporter:||Eric Westfall||Assignee:||Eric Westfall|
|Remaining Estimate:||Not Specified|
|Time Spent:||Not Specified|
|Original Estimate:||Not Specified|
|Rice Theme:||Development Ease of Use|
|Functional Justification :||eDocLite is still implemented using the old pre-Rice workflow Cascading Stylesheets and look-and-feel. This results in inconsistencies between EDL-based applications and KRAD-based applications.|
|Technical Justification:|| EDL duplicates some of the functionality that KRAD components like the data dictionary provide. It also duplicates lots of other stuff including validation, field rendering, and others.
In addition to this, EDL could also benefit from many of the features that KRAD has (like security, KIM permission integration, etc.) which it does not currently support if it was better integrated with the framework. Bridging the gap here results in some technical challenges, but those should be researched and cataloged as part of the effort to address this roadmap item.
|Impact if not Implemented:||eDocLite and KRAD will continue to have to evolve in parallel, resulting in additional duplication of functionality as we have to continue to enhance both frameworks.|
|Priority - KFS:||Low|
|Priority - KC:||Low|
|Priority - KS:||No Priority|
|Priority - Rice:||Medium|
Development Ease of Use, Ease of Implementation, Project Standardization
|Effort Estimate:||Very High ~ 2500 hrs|
eDocLite is actually very similar to what the KRAD does, except it's all configured in XML that is "installed" to the Rice standalone server as opposed to Data Dictionary files that are "loaded" at system startup time.
|Comment by Gary Prohaska (Inactive) [ 28/Jan/11 ]|
what's the motivation for doing this? Especially if the effort estimate is epic?
|Comment by Eric Westfall [ 08/Feb/11 ]|
Regarding this, I think it only makes sense to do this after some of the KRAD functionality is in place. I think that some of the functionality planned for that will help us bridge the gaps between EDL and the current KNS framework.
|Comment by Matt Sargent [ 01/Jun/12 ]|
After talking to Jerry Neal and Eric Westfall, the change should be to KRAD instead of KNS. As such I've updated this JIRA to reflect that
|Comment by Matt Sargent [ 05/Apr/13 ]|
From Eric: I think the effort estimate on this may end up being on the high side nowadays thanks to work that the KRAD team has done on the User Interface Framework and work that is being done on the new krad-data module for 2.3/2.4. Two of the thornier technical challenges (dynamic loading/reloading of data dictionary/UI configuration at runtime and the ability to leverage schemaless data models with dynamic type support backed by a custom datastore) are being addressed in the implementation of the UIF and the design of krad-data. Still, it doesn't hurt to keep the estimate high for now until the functionality required to support this has been implemented in full.