Re: [HACKERS] Planning to force reindex of hash indexes

2003-09-06 Thread jearl
Tom Lane <[EMAIL PROTECTED]> writes: > "Mendola Gaetano" <[EMAIL PROTECTED]> writes: >> "Tom Lane" <[EMAIL PROTECTED]> wrote: >>> I've found a number of infelicities in the hash index code that >>> can't be fixed without an on-disk format change. > >> How can we avoid this kind of mess for the fut

Re: [HACKERS] Planning to force reindex of hash indexes

2003-09-05 Thread Bruce Momjian
Mendola Gaetano wrote: > "Tom Lane" <[EMAIL PROTECTED]> wrote: > > "Mendola Gaetano" <[EMAIL PROTECTED]> writes: > > > "Tom Lane" <[EMAIL PROTECTED]> wrote: > > >> I've found a number of infelicities in the hash index code that can't > be > > >> fixed without an on-disk format change. > > > > > How

Re: [HACKERS] Planning to force reindex of hash indexes

2003-09-05 Thread Mendola Gaetano
"Tom Lane" <[EMAIL PROTECTED]> wrote: > "Mendola Gaetano" <[EMAIL PROTECTED]> writes: > > "Tom Lane" <[EMAIL PROTECTED]> wrote: > >> I've found a number of infelicities in the hash index code that can't be > >> fixed without an on-disk format change. > > > How can we avoid this kind of mess for the

Re: [HACKERS] Planning to force reindex of hash indexes

2003-09-05 Thread Tom Lane
"Mendola Gaetano" <[EMAIL PROTECTED]> writes: > "Tom Lane" <[EMAIL PROTECTED]> wrote: >> I've found a number of infelicities in the hash index code that can't be >> fixed without an on-disk format change. > How can we avoid this kind of mess for the future ? Build a time machine, go back fifteen

Re: [HACKERS] Planning to force reindex of hash indexes

2003-09-05 Thread Mendola Gaetano
"Tom Lane" <[EMAIL PROTECTED]> wrote: > I've found a number of infelicities in the hash index code that can't be > fixed without an on-disk format change. The biggest one is that the > hashm_ntuples field in hash meta pages is only uint32, meaning that > hash index space management will become co

Re: [HACKERS] Planning to force reindex of hash indexes

2003-09-03 Thread Bruce Momjian
Tom Lane wrote: > I've found a number of infelicities in the hash index code that can't be > fixed without an on-disk format change. The biggest one is that the > hashm_ntuples field in hash meta pages is only uint32, meaning that > hash index space management will become confused if the number o

[HACKERS] Planning to force reindex of hash indexes

2003-09-02 Thread Tom Lane
I've found a number of infelicities in the hash index code that can't be fixed without an on-disk format change. The biggest one is that the hashm_ntuples field in hash meta pages is only uint32, meaning that hash index space management will become confused if the number of entries exceeds 4G. I