Re: [sqlite] BUG Report on sqlite 3.6.20 "Error in SQL parser between sqlite3.3.4 and sqlite3.6.20"

On Dec 30, 2009, at 4:52 AM, Wilson, Ronald wrote: > I get the same error in 3.6.18, so probably the same solution > applies in 3.6.20. I got the query to work with a sub-select. Changed between 3.6.6 and 3.6.7 from the looks of things. > > SQLite version 3.6.18 > Enter ".help" for

Re: [sqlite] fts3 issue with tokenizing of content during a query

On Dec 30, 2009, at 6:25 AM, Nick Hodapp wrote: > Hi - > > I'm using sqlite 3.6.21 with this > patch >, > which I found in this forum a few weeks ago. I'm also using a custom > tokenizer which I wrote. > > My scenario is

Re: [sqlite] Archive Search Engine

On 30 Dec 2009, at 2:56am, Roger Binns wrote: > Bill Marvin wrote: >> It would be very helpful if there was a search engine for the sqlite-user >> mailing list archive. > > Gmane does a pretty good job and gives you other ways of reading the list > (eg NNTP/news, RSS): > >

Re: [sqlite] Archive Search Engine

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Bill Marvin wrote: > It would be very helpful if there was a search engine for the sqlite-user > mailing list archive. Gmane does a pretty good job and gives you other ways of reading the list (eg NNTP/news, RSS):

Re: [sqlite] .dump of utf16 database

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Kevin Ryde wrote: > Perhaps it doesn't matter to anything. The only effect is what happens behind the scenes. If you primarily use the - -16 interfaces to bind and retrieve text then the database also being utf16 means you avoid SQLite doing a

[sqlite] fts3 issue with tokenizing of content during a query

Hi - I'm using sqlite 3.6.21 with this patch, which I found in this forum a few weeks ago. I'm also using a custom tokenizer which I wrote. My scenario is this: I am storing XHTML in the database, and I want to FTS-enable

[sqlite] .dump of utf16 database

I was trying some .dump round-trips like sqlite3 old.db .dump | sqlite3 new.db and noticed if old.db is utf16, ie. pragma encoding=utf16, then new.db doesn't get that but instead is utf8. Is that intentional? Would slipping a pragma into the .dump output preserve the db coding? (I don't

Re: [sqlite] BUG Report on sqlite 3.6.20 "Error in SQL parser between sqlite3.3.4 and sqlite3.6.20"

I get the same error in 3.6.18, so probably the same solution applies in 3.6.20. I got the query to work with a sub-select. SQLite version 3.6.18 Enter ".help" for instructions Enter SQL statements terminated with a ";" sqlite> BEGIN TRANSACTION; sqlite> CREATE TABLE basica( ...> x,

Re: [sqlite] selective result columns

I'm not clear. Were you able to use Simon's syntax link: http://sqlite.awardspace.us/syntax/sqlitepg09.htm To correct your syntax? SQL works with sets, and will always return the same number of columns, but you *can* use sql to force a NULL, an empty string, or a space to be returned for a

Re: [sqlite] selective result columns

I don't have access to that level of software, so that's the problem. Thanks for your response - I like the power of select case, and was hoping that case could be applied in other areas as well. Simon Slavin-3 wrote: > > > On 29 Dec 2009, at 8:34pm, nomorecaddy wrote: > >> I'm looking for

[sqlite] Possible error when using overloaded name "oid" inside a trigger

There appears to be a problem with the use of an explicit column named "oid" from inside an 'instead of insert' trigger on a view. This seems to have been introduced in version 3.6.18 as the results differ from those of version 3.6.17. SQLite version 3.6.21 behaves the same as: SQLite version

Re: [sqlite] selective result columns

On 29 Dec 2009, at 8:34pm, nomorecaddy wrote: > I'm looking for a SQL query that returns a variable number of > columns. Many of my columns contain NULL data, and I want to avoid showing > the column altogether in that case. That is something that must be handled by your software. There is no

Re: [sqlite] undefined reference to `readline'

Hi, Am 29.12.09 19:28, schrieb Angelo: > sqlite-3.6.16/conftest.c:37: undefined reference to `readline' > collect2: ld returned 1 exit status. > > So it's clear, the linker does not find the readline function , but I don't > understand why. Do you have the readline library including headers

[sqlite] BUG Report on sqlite 3.6.20 "Error in SQL parser between sqlite3.3.4 and sqlite3.6.20"

Hello, I detect this problem because a program using sqlite command line works on sqlite.3.3.4 but it does not anymore using sqlite3.6.20 PROBLEM TITLE: Unjustified Error joining tables in sqlite3.6.20 (in previous version sqlite3.3.4 OK) OR Error in SQL parser

Re: [sqlite] selective result columns

Thanks, I'm looking for a SQL query that returns a variable number of columns. Many of my columns contain NULL data, and I want to avoid showing the column altogether in that case. Simon Slavin-3 wrote: > > > On 29 Dec 2009, at 6:29pm, nomorecaddy wrote: > >> Is it possible to run an SQL

Re: [sqlite] selective result columns

On 29 Dec 2009, at 6:29pm, nomorecaddy wrote: > Is it possible to run an SQL query in sqlite that displays columns only on a > condition? For example: > select col1, (case when 1==2 then col2) from myTable > > In this case, I only want to show one column (col1) I'm not sure I understand your

[sqlite] selective result columns

Is it possible to run an SQL query in sqlite that displays columns only on a condition? For example: select col1, (case when 1==2 then col2) from myTable In this case, I only want to show one column (col1) Thanks -- View this message in context:

[sqlite] undefined reference to `readline'

Hi everybody, sqlite 3.6.16 on linux debian. Using the amalgamation src, I compile sqlite using configure --enable-readline. But using the shell, I haven' t the history function. The config.log file says: sqlite-3.6.16/conftest.c:37: undefined reference to `readline' collect2: ld returned 1

[sqlite] BUG Report -- schema.test does not check for authorization in build

Test schema-13.1 fails with Error: {authorization not available in this build} Test needs to be bracket with ifcapable auth { do_test schema-13.1 { set S [sqlite3_prepare_v2 db "SELECT * FROM sqlite_master" -1 dummy] db function hello hello db function hello {} db auth auth db auth

Re: [sqlite] Archive Search Engine

But there is a search engine on: http://www.mail-archive.com/sqlite-users%40sqlite.org/info.html It is right at the top. Adam On Tue, Dec 29, 2009 at 12:02 PM, Bill Marvin wrote: > > It would be very helpful if there was a search engine for the sqlite-user > mailing

[sqlite] Archive Search Engine

It would be very helpful if there was a search engine for the sqlite-user mailing list archive. My question might have already been answered, but currently using the archive I have to manually look through the threads month by month. It is like finding a needle in a haystack! Bill

[sqlite] Requirements for index-aware INSERT SELECT

The code I use can calculate data flow for sql queries (summing xRead iAmt in VFS) and I noticed that many variations of INSERT SELECT led to very big data flow (multiplication of the db size). I thought that such queries can be optimized if both tables are indexed accordingly and finally the

Re: [sqlite] SQLitedb problem

Because the data was deleted. If the data was deleted it doesn't mean, that the data is not more in the database. It just not overwritten or not cleaned by an vacuum command. Is that so hard to believe? Artur -- Am 28.12.2009, 19:09 Uhr, schrieb Othman Guessous

Re: [sqlite] SQLitedb problem

I understand now. Thank you Pavel for the link. But i don't understand why i don't receive the answers to my gmail inbox. However, I joined the mailing list. So i'll check my registration. I thank everyone for the responses. I will read them with attention. Othman. On Mon, Dec 28, 2009 at 6:23

Re: [sqlite] SQLitedb problem

Thanks, for your remark. I paid no attention to the mailing list. In fact, I wanted to respond to the list. But, i doesn't receive on my inbox answers from the mailing list. Is there any answers? Othman. On Mon, Dec 28, 2009 at 6:05 PM, Pavel Ivanov wrote: > 1. Don't

Re: [sqlite] SQLitedb problem

Hello, Thanks you for your reply. Ok i will explain my sqlite's problem with more details. The problem concern the address book sqlitedb file from my iPhone. There is some data (contacts) from this file don't appear on my Iphone. And i verfy that by using a sqlite application browser : SQLite

Re: [sqlite] Using incremental BLOB I/O when processing result set

> Its a shame that a BLOB handle is not returned as a result of a query > rather than the BLOB itself. Let me correct you. It's not a shame, it's database specifics and it has its good points. For me personally I'd hate if SQLite returned me some abstract handle if I requested blob value. Note:

Re: [sqlite] Which is faster raw file I/O or sqlite BLOB

It depends on numerous facts. I'd say if size of all your files is measured in megabytes and you're not dealing with thousands of files in the same directory then raw file I/O will be faster. If size of files is measured mostly in tens or hundreds of bytes and you need thousands and millions of

Re: [sqlite] cross-database time function.

Hello! On Tuesday 29 December 2009 04:21:07 Simon Slavin wrote: > I agree that this is often an acceptable alternative. But > > * it's hard to decipher if you're reading the data by eye SQLite internal juliandays format is not human readable too. > * the system does not deal with leap

[sqlite] Which is faster raw file I/O or sqlite BLOB

Hi, I am planning to use BLOB to store file contents. So can you please suggest, which is faster 'raw file I/O' or 'sqlite BLOB'. Thank you in advance. -H ___ sqlite-users mailing list sqlite-users@sqlite.org