Re: [sqlite] wrong expression

On Thu, Jul 5, 2012 at 3:06 AM, wrote: > > > > In file os_win.c in line 2052 must be: > if( locktype==PENDING_LOCK && res ){ > instead of: > if( locktype==EXCLUSIVE_LOCK && res ){ > No. Code is correct as written. The SQLite core never actually requests a PENDING lock. PENDING

Re: [sqlite] SQLite4 nit/question

On Thu, Jul 5, 2012 at 7:23 PM, Nico Williams wrote: > In the key encoding page you talk about encoding E as a varint, and in > some cases -E as a varint, but the page on varint says they are > unsigned. I assume that "-E" == ~E + 1 (i.e., two's complement of E), > and

Re: [sqlite] Bug regarding HAVING aggregation and typecast comparisons

> sqlite> SELECT test_base.id, SUM(test_join.value) FROM test_base LEFT > JOIN test_join ON test_join.base_id = test_base.id GROUP BY > test_base.id HAVING SUM(test_join.value)='0'; > sqlite> > > The last statement generates no results. There is no mention of a > special case for HAVING so I

Re: [sqlite] Database sharing across processes

On Thu, Jul 5, 2012 at 10:54 AM, Jonathan Haws wrote: > I am fairly new to database development and I am working on an embedded > system where we are utilizing SQLite to manage some files and other > information that is being shared between processes. What I am doing

[sqlite] Bug regarding HAVING aggregation and typecast comparisons

sqlite> CREATE TABLE test_base(id INTEGER NOT NULL PRIMARY KEY); sqlite> CREATE TABLE test_join(id INTEGER NOT NULL PRIMARY KEY, base_id INTEGER NOT NULL REFERENCES test_base(id), value NUMERIC NOT NULL); sqlite> INSERT INTO test_base VALUES(1); sqlite> INSERT INTO test_join VALUES(1, 1, 0);

[sqlite] Database sharing across processes

I am fairly new to database development and I am working on an embedded system where we are utilizing SQLite to manage some files and other information that is being shared between processes. What I am doing is I have the SQLite amalgamation source code that I am compiling into each binary

[sqlite] wrong expression

In file os_win.c in line 2052 must be: if( locktype==PENDING_LOCK && res ){ instead of: if( locktype==EXCLUSIVE_LOCK && res ){ ___ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] seeking working PDO-Sqlite with php form example

Can anybody explain to me how to set up my sqlite database file with my Linux server? To my limited knowledge, i have uploaded the following files in a directory on my server: 1. sqlite.db 2. form.php Do i have to also upload the command shell?? I am confident in the php and sqlite code and

Re: [sqlite] Inserting from another table...

The insert statement below should insert one row into table PP_VIEWER_SETTINGS. Does it do that? Is it what you called "does not work"? To insert several rows you need to write a huge join of dblookup to itself, so your insert statement should look like this: insert into PP_VIEWER_SETTINGS (...)

[sqlite] Inserting from another table...

I am working on converting my system table from one form to another. The old form was one row per value with a category/key/value (DBLookup) , the new form is a separate column for each value (PP_VIEWER_SETTINGS). I am trying to create an insert statement to run when the new table is created,

[sqlite] SQLite4 nit/question

In the key encoding page you talk about encoding E as a varint, and in some cases -E as a varint, but the page on varint says they are unsigned. I assume that "-E" == ~E + 1 (i.e., two's complement of E), and that ~E == one's complement of E. Nico --

Re: [sqlite] Standalone Lua regex that can be used in sqlite

On Jul 4, 2012, at 11:00 AM, Domingo Alvarez Duarte wrote: > I did a modification to the LUA regex code to allow using it without > LUA dependency so it can be used with sqlite as regex function. Well done. Lua [1] and SQLite are made for each other :) [1] http://www.lua.org/about.html#name

Re: [sqlite] Multiple connections to in-memory database

On Thu, Jul 05, 2012 at 09:03:54AM -0400, Pavel Ivanov scratched on the wall: > So this feature shouldn't work for you. From my first message: > > > But this possibility was > > introduced in SQLite 3.7.13. So your asp.net provider should be > > compiled with the latest version of SQLite,

Re: [sqlite] When to call VACUUM - revisited

On Thu, Jul 05, 2012 at 01:29:18PM +0100, Simon Slavin scratched on the wall: > > On 5 Jul 2012, at 9:34am, _ph_ wrote: > > > I already read your previous replies, but to revisit my scenaro: > > > > - My OS is "sensitive to fragmentation" > > - We are running with

Re: [sqlite] Standalone LUA regex that can be used in sqlite

On Wed, Jul 04, 2012 at 11:00:55AM +0200, Domingo Alvarez Duarte scratched on the wall: > Hello ! > > I did a modification to the LUA regex code to allow using it without > LUA dependency so it can be used with sqlite as regex function. Very handy! > It's very light and small thus a good

Re: [sqlite] When to call VACUUM - revisited

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 05/07/12 05:29, Simon Slavin wrote: > Fragmentation of the database file on disk is something that SQLite > can't control, However you can ask SQLite to do things that will mitigate fragmentation by extending the file in larger blocks. See the

Re: [sqlite] Strange issue with sqlite 3.7.9

On 05/07/12 01:05, Richard Hipp wrote: On Wed, Jul 4, 2012 at 3:05 PM, Alan Chandlerwrote: The commit referenced by that page: http://www.sqlite.org/src/info/b23ae131874bc5c621f0 went into 3.7.9. So the problem was probably introduced in 3.7.9, not 3.7.10.

Re: [sqlite] Multiple connections to in-memory database

Thank you so much Pavel. I will try with the new version. From: Pavel Ivanov To: T Ü Cc: General Discussion of SQLite Database Sent: Thursday, July 5, 2012 4:03 PM Subject: Re:

Re: [sqlite] Multiple connections to in-memory database

So this feature shouldn't work for you. From my first message: > But this possibility was > introduced in SQLite 3.7.13. So your asp.net provider should be > compiled with the latest version of SQLite, otherwise it won't work. Pavel On Thu, Jul 5, 2012 at 8:56 AM, T Ü

Re: [sqlite] Multiple connections to in-memory database

It returns 3.6.23.1 From: Pavel Ivanov To: T Ü Cc: General Discussion of SQLite Database Sent: Thursday, July 5, 2012 3:40 PM Subject: Re: [sqlite] Multiple connections to in-memory

Re: [sqlite] Multiple connections to in-memory database

On Thu, Jul 5, 2012 at 8:37 AM, T Ü wrote: > By trying I found out that SQLiteConnection("Data > Source=:memory:;cache=shared"); worked. > In a single aspx.page at cs code, first I open an in-memory database > connection > SQLiteConnection conn = new

Re: [sqlite] Multiple connections to in-memory database

By trying I found out that SQLiteConnection("Data Source=:memory:;cache=shared"); worked. In a single aspx.page at cs code, first I open an in-memory database connection         SQLiteConnection conn = new SQLiteConnection ( "Data Source=:memory:;cache=shared" );     conn.Open(); than

Re: [sqlite] When to call VACUUM - revisited

On 5 Jul 2012, at 9:34am, _ph_ wrote: > I already read your previous replies, but to revisit my scenaro: > > - My OS is "sensitive to fragmentation" > - We are running with auto-vacuum enabled, so the freelist_count is usually > small (not a good indicator) Ah. If

Re: [sqlite] Multiple connections to in-memory database

On Thu, Jul 5, 2012 at 7:46 AM, T Ü wrote: > I have an asp.net application. > I open a sqlite in-memory connection with SQLiteConnection conn = new > SQLiteConnection ( "Data Source=:memory:" ); command. > I read that by using cache=shared parameter, I can make that

[sqlite] Multiple connections to in-memory database

I have an asp.net application. I open a sqlite in-memory connection with SQLiteConnection conn = new SQLiteConnection ( "Data Source=:memory:" ); command. I read that by using cache=shared parameter, I can make that in-memory database reachable from other connections. 1.What is the way of

Re: [sqlite] When to call VACUUM - revisited

Hi Simon, I already read your previous replies, but to revisit my scenaro: - My OS is "sensitive to fragmentation" - We are running with auto-vacuum enabled, so the freelist_count is usually small (not a good indicator) but fragmentation supposedly gets worse -We use sqlite as application