I keep needing to load poorly defined data into Hana to do some analysis, but the analysis is not quite defined up front.Many of the sources have a large number of columns, and the first step is often to work out which columns might be interesting.
I’ve created a simple page to show a summary of the fields in a table. It’s nothing which can’t be done other ways, but it’s a nice fast way to get a feel for some data, and start to see which columns might be worth further investigation.
It has mostly been used for time series data, and so it can also show a graph of selected columns, although this function is a bit more dependant on the actual data (e.g. it takes an average of the value to show each point), and is less well developed than the list of columns!
There are three parts – some xsjs to get the columns from the database and return it as JSON, similarly for the actual data in time buckets, and a SAPUI5 page to show it to the user.
- To see graphs, select one or more rows in the table, and click “Show Graphs”.
- To change the information displayed about columns, select “Names Only”, “Min/Max” or “All Stats”. You have to have at least Min/Max for the graphs to work, and All Stats can be a bit slow.
- To show the full table (rather than paging), click Maximise
- To see the table as text (to cut and paste elsewhere), click CSV.
This is not a production tool – although the user still needs to logon and have authorisation to see the data, it’s still a potential route for inappropriate access!
You need to put the three attached files into an xs project on Hana. Then call the html with parameters SCHEMA, TABLE and TIMESTAMP. The first two parameters are the schema of the table, and the tablename, and the TIMESTAMP parameter is the field to use as the timestamp for the graphs.
For example (replace server with you server, and DataAnalysis with your package name):