Re: [sqlite] Rowids, bit-arrays, recycling, and performance

2011-03-10 Thread Richard Hipp
On Thu, Mar 10, 2011 at 11:01 AM, Jay A. Kreibich wrote: > On Thu, Mar 10, 2011 at 10:37:14AM -0500, Richard Hipp scratched on the > wall: > > On Thu, Mar 10, 2011 at 10:31 AM, Dave White >wrote: > > > >1. Do I understand this correctly? > > >

Re: [sqlite] Rowids, bit-arrays, recycling, and performance

2011-03-10 Thread Jay A. Kreibich
On Thu, Mar 10, 2011 at 10:37:14AM -0500, Richard Hipp scratched on the wall: > On Thu, Mar 10, 2011 at 10:31 AM, Dave White wrote: > >1. Do I understand this correctly? > > No. I think you may be confusing SQLite and PostgreSQL. And even > PostgreSQL

Re: [sqlite] Rowids, bit-arrays, recycling, and performance

2011-03-10 Thread Simon Slavin
On 10 Mar 2011, at 3:31pm, Dave White wrote: > 1. Do I understand this correctly? Not for SQLite. It doesn't do keying arrays so don't worry about gaps in arrays or anything like that. > 2. If necessary, I can manually recycle rowids to fill in these gaps, > and maintain a

Re: [sqlite] Rowids, bit-arrays, recycling, and performance

2011-03-10 Thread Richard Hipp
On Thu, Mar 10, 2011 at 10:31 AM, Dave White wrote: > > We are considering using rowids as a unique and stable ID for all records. > I know that in order to do this, we need to prevent recycling by using > "primary key autoincrement". > > The concern is that over time,

[sqlite] Rowids, bit-arrays, recycling, and performance

2011-03-10 Thread Dave White
We are considering using rowids as a unique and stable ID for all records. I know that in order to do this, we need to prevent recycling by using "primary key autoincrement". The concern is that over time, as records are added and removed, there will appear large gaps in the rowids of a