Is this a collector bug, or is Sony overloading spell_id?

[tbigg]

I just noticed that about 6500 records in dumpeditems have recently changed the value of spellid from -1 to 65535. Since there is no spell with that ID, I'm just wondering if this is really the way the data is being sent now, or is this a bug in collector? Maybe from treating a signed int as an unsigned?

If this is really the way the data is being sent now, then Sony may have overloaded this field with some new non-spell flag value. Or they just decided to globally replace -1 with 65535.

[kalipsh]

Or maybe from having treated an always-unsigned int as a signed int before and they just fixed it?

[Doodman]

(unsigned short)-1 = 65535
Of for you windows peeps:
(uint16)-1 = 65535

Are you seeing this in the raw collected packets or in the database?

[tbigg]

I'm seeing it in the downloaded dumpeditems.sql file. I have not looked at the raw data from collected packets.