Re: [sqlite] Tell if index is used for query?

2011-03-17 Thread Simon Slavin
On 18 Mar 2011, at 6:51am, Michael Parker wrote: >> http://www.sqlite.org/datatype3.html > > Ahh, interesting. Since the date granularity is days and does't > include the time, I'll be changing date and uid to use REAL and TEXT, > respectively. INTEGER and TEXT ? Or did I misunderstand you. R

Re: [sqlite] Tell if index is used for query?

2011-03-17 Thread Michael Parker
Hi Simon, thanks for the quick and thorough reply! Inlining my response... On Thu, Mar 17, 2011 at 11:27 PM, Simon Slavin wrote: > > On 18 Mar 2011, at 6:06am, Michael Parker wrote: > >> CREATE TABLE CalendarData ( >>  date date NOT NULL, >>  uid char(40) NOT NULL, >>  json text NOT NULL, >>  PRI

Re: [sqlite] UTF-8 strings & ICU

2011-03-17 Thread Simon Slavin
On 18 Mar 2011, at 5:07am, ashish yadav wrote: > Can you please also provide some code sample for same ? > > i am not able to understand, your point "When a custom collation is > registered, it may indicate whether it wants its strings in UTF-8, UTF-16 or > either." The ICU extension for SQLi

Re: [sqlite] Tell if index is used for query?

2011-03-17 Thread Simon Slavin
On 18 Mar 2011, at 6:06am, Michael Parker wrote: > CREATE TABLE CalendarData ( > date date NOT NULL, > uid char(40) NOT NULL, > json text NOT NULL, > PRIMARY KEY (date, uid) > ); > > I just want to confirm that the index for the primary key efficiently > handles queries using WHERE or BETWEE

[sqlite] Tell if index is used for query?

2011-03-17 Thread Michael Parker
Hi all, Sorry in advance about this newbish question. Rest assured I've checked the FAQ, Wiki, and Googled before coming to you now empty-handed. I have a really simple schema: CREATE TABLE CalendarData ( date date NOT NULL, uid char(40) NOT NULL, json text NOT NULL, PRIMARY KEY (date, ui

Re: [sqlite] UTF-8 strings & ICU

2011-03-17 Thread ashish yadav
Hi Igor, Thanks a lot for info. Can you please also provide some code sample for same ? i am not able to understand, your point "When a custom collation is registered, it may indicate whether it wants its strings in UTF-8, UTF-16 or either." Thanks & Regards Ashish On Fri, Mar 18, 2

Re: [sqlite] UTF-8 strings & ICU

2011-03-17 Thread Igor Tandetnik
On 3/18/2011 12:27 AM, ashish yadav wrote: > My application is using UTF-8 strings ie database contain UTF-8 strings. > For sorting requirement of application ,i use ICU which is integrated with > sqlite3. > > Now ICU support UTF-16 encode string for its operation. > > Can any one please tell me

[sqlite] UTF-8 strings & ICU

2011-03-17 Thread ashish yadav
Hi, My application is using UTF-8 strings ie database contain UTF-8 strings. For sorting requirement of application ,i use ICU which is integrated with sqlite3. Now ICU support UTF-16 encode string for its operation. Can any one please tell me how to deal with this situation ? If ICU/Sqlite3

Re: [sqlite] How to detect existing TRANSACTION before performingSAVEPOINT?

2011-03-17 Thread Afriza N. Arief
On Thu, Mar 17, 2011 at 8:05 PM, Richard Hipp wrote: > On Thu, Mar 17, 2011 at 12:30 AM, Igor Tandetnik >wrote: > > > Afriza N. Arief wrote: > > > What is the correct way of detecting existing transaction? > > > > sqlite3_get_autocommit > > > > Correct. Or, you could just always use SAVEPOINT

Re: [sqlite] How to delete all indexes

2011-03-17 Thread Simon Slavin
On 17 Mar 2011, at 9:00pm, Jeff Archer wrote: > On Wed, Mar 16, 2011 at 5:59 PM, Simon Slavin wrote: >> You should almost never be creating indexes on the fly. Bear in mind that >> if SQLite finds a search that would be better with an index, it creates it >> itself and it is far better at wor

Re: [sqlite] How to delete all indexes

2011-03-17 Thread Jeff Archer
On Wed, Mar 16, 2011 at 5:59 PM, Simon Slavin wrote: > You should almost never be creating indexes on the fly.  Bear in mind that if > SQLite finds a search that would be better with an index, it creates it > itself and it is far better at working out the best index than you are.  The > only di

[sqlite] Yes, I checked out SqLite Maestro

2011-03-17 Thread Matt Young
I thought is just about matches everything I need for managing my databases. I gave it a thumbs up. That is all from a small researcher with gobs of data. ___ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listin

[sqlite] AUTO: Bret Patterson/Austin/IBM is out of the office (returning 03/21/2011)

2011-03-17 Thread Bret Patterson
I am out of the office until 03/21/2011. I'm out of the office but checking email once or twice a day and will respond to any high importance issues as quickly as possible. Note: This is an automated response to your message "sqlite-users Digest, Vol 39, Issue 17" sent on 3/17/2011 6:00:01.

Re: [sqlite] Optimizing list retrieval with a FTS3 table

2011-03-17 Thread Travis Orr
Scott, Thank you for clarifying the inefficiency of FTS3 when not using a MATCH criteria. Unfortunately there are other use cases that do require the MATCH criteria so the FTS3 is required. I believe reading everything into a temp table would consume too much memory as in an ideal situation we

Re: [sqlite] How to detect existing TRANSACTION before performingSAVEPOINT?

2011-03-17 Thread Richard Hipp
On Thu, Mar 17, 2011 at 12:30 AM, Igor Tandetnik wrote: > Afriza N. Arief wrote: > > What is the correct way of detecting existing transaction? > > sqlite3_get_autocommit > Correct. Or, you could just always use SAVEPOINT which works the same as BEGIN if you are not already in a transaction.

[sqlite] Lemon and errors

2011-03-17 Thread Igmar Palsenberg
Hi, I'm attempting to rewrite a buggy hand-written parser with a re2c / lemon one. I've ran into a strange thing (which is/was actually a bug in the lexer), but I can't determine why lemon finishes parsing, instead of throwing up a parser error. The stripped-down parser : [igmar@devel native]