Personalized Information Links are essentially our approach to row level security.
Personalized Information Links are essentially our approach to row level security. Most importantly, they're simple to implement and manage. Check out how to create them.
(16) Answers
As pointed out, this technique does not work for cached analysis of large data sets that still needs row level data control and I have not see a good way around it.
What we have done is create one analysis with all data in it and schedule this report on the web that is refreshed as needed.
Then create a COPY of this report so that the data table has the same GUID and apply personalization to the new report.
Then whenusers access the report with personalization on it, the web player uses the data alread loaded in memory to create the restricted data view rather than making a new DB request.
Clumsy but it worked.
1 Comment
Hi djwillia,
You don't need to create separate reports to use this technique (achieving rowlevel security on preloaded data).
It boils down to putting your big table in the WebPlayer memory through scheduled update, then limit the data viewed by the user by doing a personailzed read of data to populate a "security table". Third step will be to create an inner join between these two tables, thus limiting the data exposed to the user viewing the analysis to the only contain the allowed subset.
Hope this helps.
1 Comment
Hi djwillia,
You don't need to create separate reports to use this technique (achieving rowlevel security on preloaded data).
It boils down to putting your big table in the WebPlayer memory through scheduled update, then limit the data viewed by the user by doing a personailzed read of data to populate a "security table". Third step will be to create an inner join between these two tables, thus limiting the data exposed to the user viewing the analysis to the only contain the allowed subset.
Hope this helps.
If the security table is a different table from the main table which needs to be cached, and the row level security is achieved by creating the a relation (set to include filtered rows only) between the security table and the main data table then a 6.5 report should work fine when it is in the cache with Scheduled Updates. 6.5 has a feature which allows you to force a refresh for any data table for each user even if the report is cached. This allows you to reload the smaller security table each time someone opens the report. Go to Edit > Data Table Properties and click on the Scheduled Updates tab
Hi All,
I've tried to follow the instructions, but I get an error. Below is a video of what I'm doing - and the error I get. Any insights into what I'm doing incorrectly would be much appreciated.
Thanks,
Olga
Error message: Unable to import data. See the details for more information.
ImportException at Spotfire.Dxp.Data:
Failed to execute data source query. (HRESULT: 80131500)
Stack Trace:
at Spotfire.Dxp.Data.DataSourceConnection.ExecuteQuery2()
at Spotfire.Dxp.Data.DataFlow.Execute()
at Spotfire.Dxp.Data.DataFlow.DataFlowConnection.ExecuteQueryCore2()
at Spotfire.Dxp.Data.DataSourceConnection.ExecuteQuery2()
at Spotfire.Dxp.Data.Producers.SourceColumnProducer.<>c__DisplayClass15.b__12()
at Spotfire.Dxp.Framework.ApplicationModel.Progress.ExecuteSubtask(String title, ProgressOperation operation)
at Spotfire.Dxp.Data.Producers.SourceColumnProducer.GetColumnsAndProperties(DataSourceConnection connection)
at Spotfire.Dxp.Data.Persistence.DataItem.PerformUpdate(SourceColumnProducer producer, DataSourceConnection connection)
at Spotfire.Dxp.Data.Persistence.DataItem.Update(SourceColumnProducer producer, DataSourceConnection connection)
at Spotfire.Dxp.Data.Persistence.DataPool.d__2.MoveNext()
at Spotfire.Dxp.Data.Producers.SourceColumnProducer.ConfigureAndLoadData()
at Spotfire.Dxp.Framework.DocumentModel.Node.ConfigureSubTree()
at Spotfire.Dxp.Framework.DocumentModel.Node.b__10(Node node)
at Spotfire.Dxp.Framework.DocumentModel.Node.ConfigureSubTree()
at Spotfire.Dxp.Framework.DocumentModel.Node.b__10(Node node)
at Spotfire.Dxp.Framework.DocumentModel.Node.ConfigureSubTree()
at Spotfire.Dxp.Framework.DocumentModel.Node.b__10(Node node)
at Spotfire.Dxp.Framework.DocumentModel.State.NodeState.ForEachChild(IDocumentNodeChild[] children, Action`1 action)
at Spotfire.Dxp.Framework.DocumentModel.Node.ConfigureSubTree()
at Spotfire.Dxp.Framework.ApplicationModel.Progress.ExecuteSubtask(String title, IndeterminateProgressFormatter progressFormatter, ProgressSpeedFormatter speedFormatter, ProgressOperation operation)
at Spotfire.Dxp.Framework.DocumentModel.DocumentNode.ConfigureAndAttachFromNew()
at Spotfire.Dxp.Framework.DocumentModel.DocumentNode.AttachSubTreeWhileExecuting(UndoableNode newOwner)
at Spotfire.Dxp.Framework.DocumentModel.DocumentNode.AttachToUndoableNode(UndoableNode newOwner)
at Spotfire.Dxp.Framework.DocumentModel.UndoableKeyedCollection`2.<>c__DisplayClass6.b__5()
at Spotfire.Dxp.Framework.DocumentModel.Node.Spotfire.Dxp.Framework.DocumentModel.INode.InternalTransaction(Executor executor)
at Spotfire.Dxp.Framework.DocumentModel.UndoableKeyedCollection`2.Insert(Int32 index, TNode item)
at Spotfire.Dxp.Data.DataTableCollection.<>c__DisplayClass20.b__1f()
at Spotfire.Dxp.Framework.DocumentModel.Node.Spotfire.Dxp.Framework.DocumentModel.INode.InternalTransaction(Executor executor)
at Spotfire.Dxp.Data.DataTableCollection.Add(String dataTableName, DataSource dataSource, Boolean autoCreateFilters, Boolean includeInAxisControls)
at Spotfire.Dxp.Forms.Data.Import.TableContext.DataFlowBuilderContext.<>c__DisplayClass3.b__0()
at Spotfire.Dxp.Framework.ApplicationModel.Progress.ExecuteSubtask(String title, ProgressOperation operation)
at Spotfire.Dxp.Forms.Data.Import.TableContext.DataFlowBuilderContext.AddDataTables(DataManager dataManager, Form baseForm, List`1& addedTables, String& errorMessage)
InformationModelException at Spotfire.Dxp.Data:
Failed to execute query: El nombre de columna ''opolyakov'' no es v--lido. (HRESULT: 80131500)
Stack Trace:
at Spotfire.Dxp.Data.InformationModel.InternalInformationModelManager.DataStream.GetNextBlock()
at Spotfire.Dxp.Data.InformationModel.InternalInformationModelManager.DataStream.Read(Byte[] buffer, Int32 offset, Int32 count)
at Spotfire.Dxp.Internal.Utilities.SeekableStream.Read(Byte[] buffer, Int32 offset, Int32 count)
at System.IO.BinaryReader.FillBuffer(Int32 numBytes)
at System.IO.BinaryReader.ReadUInt16()
at Spotfire.Dxp.Data.Formats.Sbdf.SbdfSectionHeader.ReadSectionType(BinaryReader reader, SbdfSectionTypeId expectedType)
at Spotfire.Dxp.Data.Import.SbdfDataRowReader.DelayedInit()
at Spotfire.Dxp.Data.Import.InformationLinkDataSourceConnection.ExecuteQueryCore2()
at Spotfire.Dxp.Data.DataSourceConnection.ExecuteQuery2()
InformationModelServiceException at Spotfire.Dxp.Services:
Failed to execute query: El nombre de columna ''opolyakov'' no es v--lido. (HRESULT: 80131509)
Stack Trace:
at Spotfire.Dxp.Services.WebServiceBase`1.InvokeService[T](ServiceMethod`1 serviceMethod, ExceptionFactoryMethod exceptionFactoryMethod)
at Spotfire.Dxp.Data.InformationModel.InternalInformationModelManager.DataStream.GetNextBlock()
Hi Pavan, Sorry, I thought I had translated in the audio. Yes, the name of the column opolyakov is not valid, but I don't know why it would think that is the name of the column. Those are Entity and Userid.
I think these are the screenshots you are asking for.
Thanks again,
Olga
| Attachment | Size |
|---|---|
 Teacher Table table screenshot.PNG | 77.62 KB |
| Lookup Info Link Screenshot.PNG | 59.3 KB |
| Teacher Roster Info Link table screenshot.PNG | 66.94 KB |
| Lookup table screenshot.PNG | 42.89 KB |
