Hmm, I was wrong about the above issue (strictValidationFailed) getting fixed while I was in the middle of things. It's an odd one.
On sheetA, my table definition succeeded when I removed strict 'entries from a list' validation from a header cell. When attempting to put the validation back to prove my point, I could set the validation, but it proved no obstacle to the table definition - the validation just got removed. SheetB has an 'original' validation that's the same type (alhough different values allowed in the list), and this is again preventing a table being defined. I expect if I remove the validation, my table definition will succed. I probably won't bother filing a bug about this, since I don't really want my header cells to have a failed validation anyhow. But it does illustrate the wierdness that we out here in userland face. Had I filed a bug, you wouldn't have been able to reproduce it, because it's 'special' to my spreadsheet. I'll keep a copy of the sheet to hand in case you guys want to submit it for forensic analysis.
