Re: [sqlite] Committing to a savepoint

On Fri, Jan 14, 2011 at 2:16 AM, Charles Samuels wrote: > > Here's more or less what I need: > > A * sqlite gets some inserts > B * we're at a checkpoint, so everything after this point shouldn't get > committed now. So "savepoint SP" > C * insert some more into sqlite > D *

Re: [sqlite] how to put the database in cache without waiting the system do it by himself ?

it's not enalf :( On 1/14/2011 1:48 AM, Jean-Christophe Deschamps wrote: >> when i just launch th application, at the beginning the query can take >> around fews seconds... but after some time (10 - 20 minutes), it's take >> only few ms ! >> >> so i guess it's because the windows cache in memory

Re: [sqlite] How to track a record if the aliased id is changed

On 13 Jan 2011, at 17:27, Nicolas Williams wrote: > On Thu, Jan 13, 2011 at 10:59:29AM +, Simon Slavin wrote: >> His problem is that he doesn't know which rows are aliased to rowid, >> so he can't provide a list of column names. So the following might be >> closer >> >> CREATE TEMP TRIGGER

Re: [sqlite] How to track a record if the aliased id is changed

On Fri, Jan 14, 2011 at 1:20 PM, Philip Graham Willoughby < phil.willoug...@strawberrycat.com> wrote: > On 13 Jan 2011, at 17:27, Nicolas Williams wrote: > > > On Thu, Jan 13, 2011 at 10:59:29AM +, Simon Slavin wrote: > >> His problem is that he doesn't know which rows are aliased to rowid, >

Re: [sqlite] how to put the database in cache without waiting the system do it by himself ?

On Fri, Jan 14, 2011 at 1:18 PM, Vander Clock Stephane < svandercl...@yahoo.fr> wrote: > it's not enalf :( > > > Read the file! > > > If you know your queries in advance and it's some SELECT you can create a temporary query wrapped into "SELECT COUNT(*) FROM (your select here)" and execute it

Re: [sqlite] How to track a record if the aliased id is changed

On 14 Jan 2011, at 10:45, Max Vlasov wrote: > Phil, it's an interesting solution. I will probably use it, but the current > problem is that I have to resolve object context from non-oop function. It's > possible (maybe even by hard-coding pointer as one of parameters in the > trigger statement),

Re: [sqlite] How to track a record if the aliased id is changed

On Fri, Jan 14, 2011 at 1:57 PM, Philip Graham Willoughby < phil.willoug...@strawberrycat.com> wrote: > > On 14 Jan 2011, at 10:45, Max Vlasov wrote: > > > Phil, it's an interesting solution. I will probably use it, but the > current > > problem is that I have to resolve object context from

Re: [sqlite] How to track a record if the aliased id is changed

On 14 Jan 2011, at 11:06, Max Vlasov wrote: > I understand this, but in my case the tables are objects also. So when I > create a temporary trigger I should pass the table object context. Dead easy: make the object context required to find the table a parameter to the function, and put the

[sqlite] sqlite3 ltrim behaviour bug or feature?

Hi, the following seems wrong to me: bash-4.0# sqlite3 SQLite version 3.6.14.2 Enter ".help" for instructions Enter SQL statements terminated with a ";" sqlite> select ltrim("12300567","1230"); 567 sqlite> select ltrim("012300567","0123"); 567 sqlite> select ltrim("12300567","123"); 00567

Re: [sqlite] sqlite3 ltrim behaviour bug or feature?

Hi, this is the expected behaviour. See http://www.sqlite.org/lang_corefunc.html "The ltrim(X,Y) function returns a string formed by removing any and all characters that appear in Y from the left side of X. If the Y argument is omitted, ltrim(X) removes spaces from the left side of X. " The

Re: [sqlite] View with values in 1st table replaced by values in second table

Thank you both for the advice, I had never thought to join on the same table using 3 different names like that, will have to keep that in mind! On Thu, Jan 13, 2011 at 9:27 PM, Jay A. Kreibich wrote: > On Thu, Jan 13, 2011 at 01:44:12PM -0600, Josh Marell scratched on the >

[sqlite] Segfault from command-line app

I've never seen this before. In my application I executed the command INSERT INTO books DEFAULT VALUES and my application crashed inside the sqlite library. But I don't want you to try to debug my app so I tried the same thing in the command-line utility and got a crash with Segmentation

[sqlite] Bug report

Example sql-s how to reproduce: create table test (a integer not null, b float not null); insert into test values (1, 0); insert into test values (2, 0); insert into test values (3, 0); I actualy get values from other tables but in my example they sum to zero so to simplify you can try just

Re: [sqlite] Segfault from command-line app

Here is part of the report from the crash of my own application: Exception Type: EXC_BAD_ACCESS (SIGSEGV) Exception Codes: KERN_INVALID_ADDRESS at 0x0010 Crashed Thread: 0 Dispatch queue: com.apple.main-thread Thread 0 Crashed: Dispatch queue: com.apple.main-thread 0

Re: [sqlite] Segfault from command-line app

I don't know if the lack of a semicolon at your END statement could be causing this? Or is that a typo? This works just fine in 3.7.4 Michael D. Black Senior Scientist Advanced Analytics Directorate Northrop Grumman Information Systems From:

Re: [sqlite] Bug report

Radovan Antloga wrote: > create table test (a integer not null, b float not null); > > update test > set b = 0.0 / 0; > > you will get error: test.b may not be null Appears to be working as intended. Division by zero produces null, which cannot be stored in test.b .

Re: [sqlite] Segfault from command-line app

On 14 Jan 2011, at 2:53pm, Black, Michael (IS) wrote: > I don't know if the lack of a semicolon at your END statement could be > causing this? Or is that a typo? Oops. Sorry, that semi-colon is actually there, and the command-line tool shows another prompt. Faulty cut-and-paste into the

Re: [sqlite] Segfault from command-line app

AFAIK, this was fixed in versions after 3.6.23. See http://www.sqlite.org/src/info/f3162063fd and http://www.mail-archive.com/sqlite-users@sqlite.org/msg51588.html. Pavel On Fri, Jan 14, 2011 at 9:48 AM, Simon Slavin wrote: > I've never seen this before.  In my

Re: [sqlite] Bug report

S, Igor Tandetnik piše: > Radovan Antloga wrote: >> create table test (a integer not null, b float not null); >> >> update test >> set b = 0.0 / 0; >> >> you will get error: test.b may not be null > Appears to be working as intended. Division by zero produces null, which

Re: [sqlite] Segfault from command-line app

On Fri, Jan 14, 2011 at 9:48 AM, Simon Slavin wrote: > I've never seen this before. In my application I executed the command > > INSERT INTO books DEFAULT VALUES > > and my application crashed inside the sqlite library. But I don't want you > to try to debug my app so I

Re: [sqlite] Segfault from command-line app

On 14 Jan 2011, at 3:32pm, Richard Hipp wrote: > Thanks for the bug report, and for including a reproducible test script with > the report. Thanks to Michael, Pavel and Richard. I've now updated my app to the latest amalgamation. Simon. ___

Re: [sqlite] How to track a record if the aliased id is changed

On Fri, Jan 14, 2011 at 10:20:22AM +, Philip Graham Willoughby wrote: > On 13 Jan 2011, at 17:27, Nicolas Williams wrote: > > On Thu, Jan 13, 2011 at 10:59:29AM +, Simon Slavin wrote: > >> CREATE TEMP TRIGGER [UpdatedRowIdTrigger] AFTER UPDATE ON TestTable > >> BEGIN > >> UPDATE

[sqlite] Procedural SQL techniques (Re: How to track a record if the aliased id is changed)

On Fri, Jan 14, 2011 at 10:53:56AM -0600, Nicolas Williams wrote: > But that function would have to track those changed rowids somewhere. > The trigger I posted does exactly that, using SQLite3's own primitives > (a temp table in this case): Speaking of which, the lack of procedural programming

Re: [sqlite] Committing to a savepoint

> There are some challenges, for example to allow arbitrary undo > operations we should also log transaction boundaries since undoing to some > points in between not only makes little sense, but also dangerous. But I > think if implemented with those challenges solved, such implementation would >

Re: [sqlite] Committing to a savepoint

On Fri, Jan 14, 2011 at 8:55 PM, Pavel Ivanov wrote: > > There are some challenges, for example to allow arbitrary undo > > operations we should also log transaction boundaries since undoing to > some > > points in between not only makes little sense, but also dangerous. But

Re: [sqlite] Committing to a savepoint

Max, I understand your point of view. But you assume that any insert ends up with call to xWrite in the VFS (so that VFS could track the changes). But it's not that way in SQLite. xWrite is called only during commit and it's provided with full database pages, not individual changes. Pavel On

[sqlite] FTS3 bug

Hi all, I've found a defect in FTS3 where it will accept matching with the string "" (two quotes) but not """ (three quotes). I first came across the issue on Android, but have since reproduced it in Python. 1) I assume this is the place to post FTS3 bug reports, please let me know otherwise.