I have a dataset of 1 million objects to analyze by property and frequency. The data are organized as a master worksheet that records object id and properties, and a satellite set of worksheets that contain frequency data for various subsets of the objects.
My question is how to associate the correct object properties with the frequency data for graph construction. I have tried to use the wks.join method to construct a new worksheet that merges the frequency and property data for the object id's common to both worksheets on a single row in the new worksheet (i.e. a SQL inner join on 2 tables), but cannot figure out the correct syntax.
win -a IndexList; // Index list of object ids (X) & attributes (YYY) %I=%H; // store name of index worksheet win -a Frequency1;// now get frequency data (Y) with ids (X) worksheet -d Joined; %H!wks.joinmode=0; // Set to append mode %H!wks.join(%I); // Add data to worksheet
Worksheet Joined will have id's in (X) and (X2), but out of sync, since the set of (X) is a subset of (X2).
Is the raw data of 1 million objects located in some database that you are now all reading into Origin?
If in database, and if you have OriginPro 7.5 you can use COM client programming to access ADO object and then specify search string with inner joins etc to bring in just the right data you want into Origin and then plot etc?
In Pro, there is an example showing how to do this: \Samples\COM Client\ADOSample.c
can Origin perform SQL-style inner joins?
Posted - 08/04/2005 : 7:09:30 PM
