Idea

The problem I see with this is that it doesn't scale well for inner joins. Do you add two columns to the result if it's an inner join of two tables, and do you automatically name them based on the table name, and rank them based on their rank in the FROM? I guess to me this solution just isn't in keeping with the design goals of SQL. It's just weird that we would haphazardly add columns to the result table. Shouldn't that be encoded into the SQL query itself? I've searched several times online to see if there is some SQL-standard keyword that you can use to get the row of the table. It looks like there isn't, likely because usually each row has a unique id field, and you always go off of that id, so you don't need to worry about a row number. If you follow that unique id guideline, you can always add a column to your source table that is just there to tell you what row it is. The other option is to use FlexSim's custom $ syntax to essentially do it yourself.

query("SELECT Col1, Col2, $2 AS RowNum FROM $1", reftable("GlobalTable1"), getrank($iter(1)));

The problem with this solution is that it doesn't work if you have a bundle table (getrank($iter(1)) doesn't work).

At this point I'm not convinced we should add this as a parameter of dumpquery(). My preference would be to get $iter() to work somehow for bundle tables, and/or to add a FlexSim-specific SQL keyword that gets the table row. It seems the broader SQL community doesn't make much stink about not having a keyword, but in FlexSim, lots of interfacing with tables has to do with accessing tables via row numbers, so maybe our situation is different.