Arno Brinkman wrote:

>It's done per field/item, so the delta's will be still small and possible:
>
>edsNull, edsIntLen5, <data>, edsCompression, edsZip, <data>, edsIntLen4, <data>,
>edsCompression, edsZip, <data>, ...
>
>The <data> inside edsCompression can contain compressed :
>- edsUtf8Count2, <data>
>- edsOpaqueCount3, <data>
>- or whatever other type
>
>
>

To elaborate on my previous post, I think this would work better:

edsNull, edsIntLen5, <data>, edsCompression, edsZip, edsUtf8Count2,
<data>

This would say that the <data> following the edsUtf8Count2 is utf-8 when
decompressed from zip encoding. The two byte count field would reflect
the compressed length. There may be merit in encoding the decompressed
length somewhere, but I'm not sure about this.