Re: Trigger Buffer Question

["Nelson C. Smith" <ncsmith@xxxxxxxx>]

I'm in the process of converting my trigger procedures over from the Big Array overlay method (32,767 1-byte elements) to one in which I use pointer math to find the offsets.  The question that comes up is what does data management put into the trigger buffer when the file record is really big?  Say I have 32,767 one-byte fields in a single record?  Does it write the 96 fixed bytes plus four sets of 32,767 byte variable fields (the before buffer, the before nullmap, the after buffer, and the after nullmap)? 

 

If so, then the big array overlay method I've been using would not have worked anyway for big files like this would it, since it depends on starting out with a data structure and overlays it with an array of only 32767 bytes?  What will data management write to the buffer when we get the really big fields (blobs, clobs, etc)? 

 

If I switch over to calculating pointers to the four variable areas based on the trigger buffer's beginning address plus the offsets (plus 1), will this be able to handle those really big fields?  Or is the record in the buffer just going to contain some sort of pointer to that 2 meg. image in the blob?