Kbase P152798: What are the reasons behind the code change in datavis.p from Progress 9.1C to OpenEdge 10.2A
| Autor |
Progress Software Corporation - Progress |
| Acesso |
Público |
| Publicação |
9/22/2009 |
|
Status: Unverified
GOAL:
What are the reasons behind the code change in datavis.p from Progress 9.1C to OpenEdge 10.2A
GOAL:
Is there any thing wrong with reverting the code to its Progress 9.1C version?
FACT(s) (Environment):
Windows
Progress 9.x
OpenEdge 10.x
FIX:
The datavis.p internal procedure resetRecord code changes from Progress 9.1C to OpenEdge 10.2A include replacing the statement:
Run displayRecord. Progress 9.1C with the statement IF cNewRecord <> 'NO':U THEN RUN displayRecord IN TARGET-PROCEDURE. in OpenEdge 10.2A.This change is one of many in this area (undo/reset/cancel). The basis for this change is that the datasource's resetRow, which is called from datavis.p resetRecord, is now publishing dataAvailable. This publish reaches the viewer and causes the viewer to call displayRecord in itself.
Reverting the code to its Progress 9.1C version should not do any harm, but (in theory) may cause the display to happen twice. An extra display may be harmless, in particular since this case only applies to resetRecord, but there are cases were developers may consider an extra display unacceptable for performance reasons, for example when using LOBs or when using Dynamic lookups (without caching on an AppServer).