On Mon, Feb 24, 2003, Chris Hawks wrote:

> You're more familiar with the uncompress code (and palm databases)
> than I, so, how about this plan for uncompressing (possibly nested)
> tables??

I have a very pragmatic view; if it works then use it ;-)

It is always possible to improve (or even completely rewrite) the
code later.

> Uncompress creates a new record in the uncompress database, saves
> the record # and returns the handle.

OK, so far.

> table.c locks the handle and callsGet UncompressTableIndex() in
> uncompress to get the record #.

This part I have some doubts about. I think it would be better to use
the UID for the table record as an "index", i.e. the uncompress code 
keeps a map with the <UID, uncompress record#> info, so that later
when the table code wants to release a record it only needs to provide
the UID. Just a matter of encapsulation (i.e. no one but the uncompress
code really needs access to the record#)

/Mike

_______________________________________________
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Reply via email to