Re: [sqlite] Date time input

On Wed, Oct 9, 2019 at 9:05 PM Keith Medcalf wrote: > > On Tuesday, 8 October, 2019 12:40, James K. Lowden < > jklow...@schemamania.org> wrote: > > >OK, but it needs a better name. What better place than here to debate > >that? ;-) > > >What the opposite of "Lite"? > > It's 'ite' as in 'stalagm

Re: [sqlite] Date time input

On Tuesday, 8 October, 2019 12:40, James K. Lowden wrote: >OK, but it needs a better name. What better place than here to debate >that? ;-) >What the opposite of "Lite"? I believe the PC euphemism is Big and Tall. SQLBAT3? -- The fact that there's a Highway to Hell but only a Stairway to

Re: [sqlite] Date time input

Tuesday, October 08, 2019, 7:39:40 PM, James K. Lowden wrote: > OK, but it needs a better name. What better place than here to debate > that? ;-) > SQLOfALot (providing you pronounce SQL as ess-queue-ell and not sequel) Graham ___ sqlite-users m

Re: [sqlite] SELECT uses index with SUBSTR but UPDATE doesn't

minor correction, char(255) should be x'ff' ... char(255) does not return the byte 0xff, but x'ff' is the byte 0xff ... where (name between :prefix and (:prefix || x'ff') and substr(name, 1, length(:prefix)) collate nocase == :prefix) -- The fact that there's a Highway to Hell but only a Stai

Re: [sqlite] SELECT uses index with SUBSTR but UPDATE doesn't

On Wednesday, 9 October, 2019 12:01, Jens Alfke said: >BETWEEN doesn't work well because it's inclusive, i.e. `BETWEEN 'foo' and >'fop'` doesn't work because it matches 'fop'. Coming up with the upper >end of a string prefix match is super annoying — `BETWEEN 'foo' and >'foo\xff' only works unti

Re: [sqlite] SELECT uses index with SUBSTR but UPDATE doesn't

> On Oct 9, 2019, at 10:02 AM, Keith Medcalf wrote: > > SUBSTR(name, 0, ?) is an expression, so unless you have an index on that > expression, then an index cannot be used to SEARCH for the rows. That's accurate in general. However, there _is_ a very similar special-case optimization for the

Re: [sqlite] [patch] avoid dynamic alloc in vdbeSorterSort(...)

Keith Medcalf wrote: > > On Monday, 7 October, 2019 14:58, Dominique Pellé < > dominique.pe...@gmail.com> wrote: > > >Here the allocated size is fixed (always 64 pointers), so alloca does > >not seem needed. > > >I wonder how many other functions could avoid dynamic allocation > >like this one (e

Re: [sqlite] SELECT uses index with SUBSTR but UPDATE doesn't

Unable to reproduce. In particular: >SELECT * FROM nodes WHERE SUBSTR(name, 0, ?) = ? >tells me that it can and will use the (primary key) index on the name >column. will not use the index. I can make it use an index by doctoring the table data to make the index scan cheaper than a table scan