v3.0.3.0
Release 3.0.3.0
ALL VERSIONS
[Enhancement]
- Enhanced synchronization log for FI_T/UC_T tags.
- Performance enhancements in set rules consolidation step of synchronization.
- Several UI enhancements on Run Manager.
[Fixes]
- Blank spaces after the Minus sign are removed in exclusion rules.
- Solving a case with multiple time slice definitions selected. See details
- Super region is now case insensitive.
- Veda assigns dimension based on the element in FI_T/UC_T col headers and table tags. The allocation will be done in the following priority order: attribute,currency,lim_type,region,side,time_slice,year,commodity.
[New]
- Veda version introduced in the title of LST files.
- Attrib_cond/Val_cond feature was purely a process filter so far. It now works as a process-commodity filter if the attribute in question has process and commodity indexes.
ADVANCED VERSION
- Reporting variables that are generated to support SANKEY diagrams are not shown in the list of variables.
- Multi-user support in Results and Reports views.