Is it possible to set PBS directly similar as PBA without setting many criteria with a lot value pairs?
What CR or TR tickets(for reference) for this ability to be implemented?
Update:At last true PBA will be possible in Bentley Map SS4 - http://communities.bentley.com/products/geospatial/desktop/f/5924/p/85188/240358.aspx#240358
Yes, this will happen, and I think it is a good thing, but there is a price to pay for this.
Since this is implemented at the API level :
1. You will have to create your own code to generate the PBS (not PBA). This will hide all the information as to 'how' the symbology was set.
2. The computational handoff could be quite intense, if there are a lot of different decisions to be made in generating the PBS values.
3. The logic for the PBS is built into the code, thus any minute changes will require a code change, compile and retesting.
I say this because I have been involved in projects where ALL the placement/edit for features was done via customized MDL code along with parameters set in the database. The process/debugging/testing could be quite intense and requires someone who can uinderstand MDL and the supported languages, along with the feature definitions and databases. Especially when the symbology settings were placed in a database table.
Thanks for your answer it explains a lot. I still quite don't get it if it works for PBA why it can't be implemented the same way for PBS to directly set symbology from properties without much logic. Ok if properties are wrong then also symbology will be but it is then data issue and not PBS.
Think 3 properties with 15 different possibilities each. What is the greatest possible number of checks, I code, to determine the level of the feature ?
I realize the following is not a typical situation.
I know of a company that is implementing this new functionality. Their original feature table, based upon the different symbology possibilities was over 20k entries at one time. Granted they have it worked down some, to a more manageable level and this new functionality will help them out quite a bit. The code to determine this is quite intense for their feature sets. It is not trivial.
This is definitely needed. I tend to think sometimes a redesign of the features/attributes may go a long way in assisting in the CADD rendering off the data.
We need this for simple cases that level name is the same as property or combination of properties, there are no checks needed.