Kepler

The PDF seems to render "nil" as "nil" whereas the AWT rendering (from Apache-FOP) sometimes shows a huge integer instead. I believe it's something to do with how the AWT is using components- maybe the column (int) is confused by this? Also seeing some "NaN" entries in a different column (double)

After some digging, I'm going to conclude that this is being introduced by the "deep copy" serialization technique for creating a ReportInstance from a ReportLayout.

2147483647 = Integer.MAX_VALUE like IntToken.NIL

added special handling when we have IntToken.NIL in the report instance

NEED TO COMMIT TO BRANCH

This is also happening for DoubleTokens - but they show as NaN rather than nil - which I believe is acceptable.
BooleanTokens loose their nil-ness and become false which may be misleading in some situations - but typically it corresponds to rows where there are null data and you can infer that false is inconclusive. At any rate, we're stuck with that.

committed to the branch.

Original Bugzilla ID was 4217