Re: [sqlite] [EXTERNAL] "Optional" incoming parameters for virtual tables

On Thu, 18 Apr 2019 at 09:48, Hick Gunter wrote: > So the QP is attemopting to determine which cost product is less: > > My guess is it will probably choose the undesired, invalid plan. A linear > cost penalty for desired but optional inputs is probably not going to cut > it. In this case, the re

[sqlite] Page Size question

When I get my act together and get un-lazy about it, I want to take a large set of "small" files and put them into a SQLite database blobs in a table, then have my app run some kind of CRC check (CRC32/MD5SUM/SHA1/custom/whatever - Haven't decided on a CRC or just byte-for byte comparisons) to iden

Re: [sqlite] Page Size question

On 4/18/19, Stephen Chrzanowski wrote: > I'm wondering > if there is going to be any optimization internally from SQLite by making > the page size, say, 180k? ( SQLite handles storing 200KB blobs just fine. It has been doing so routinely for time out of mind. Do not stress over the details of h

Re: [sqlite] Reporting two more interesting queries with performance regression

On 4/6/19, Jinho Jung wrote: > > +--++ > | Query| Time | > +--++ > | 10002.sql (v3.23)|789 | > | 10002.sql (v3.27.2) | 1270 | > +--++ > | 10052.sql (v3.23)| 3094 | > | 10052.sql (

[sqlite] SQLITE_MAX_MMAP_SIZE 2GB default

Hi there, My apologies if this has been answered in the past, but is there a reason for the ~2GB default for SQLITE_MAX_MMAP_SIZE ? #ifndef SQLITE_MAX_MMAP_SIZE ... # define SQLITE_MAX_MMAP_SIZE 0x7fff /* 2147418112 */ It makes sense for 32bit platforms, but for