Re: [sqlite] Problem installing TCL bindings

2006-04-08 Thread Thomas Chust
On Sat, 8 Apr 2006, Miguel Bazdresch wrote: [...] It actually needs to say load /usr/lib/tcl8.4/sqlite3/libtclsqlite3.so Sqlite3 [...] Hello Miguel, if everything works correctly and the Tcl install script of SQLite3 did its job right, which it usually does, it should be unnecessary to l

Re: [sqlite] Problem installing TCL bindings

2006-04-08 Thread Miguel Bazdresch
On 4/7/06, Miguel Bazdresch <[EMAIL PROTECTED]> wrote: > 2. For some reason, the libraries needed for tcl interaction > (libtclsqlite3.so) are not installed by 'make install' I finally determined they *are* installed, to /usr/lib/tcl8.4/sqlite3. A mention of this somewhere on the website or READM

Re: [sqlite] Current julian day number

2006-04-08 Thread Thomas Chust
On Sun, 9 Apr 2006, ThomasChust wrote: [...] is there an elegant way to make the current julian day number the default value for a column in SQLite3? [...] Hello, sorry for the last post -- I figured out how it works myself: CREATE TABLE time(time REAL DEFAULT (julianday(current_timestamp)

[sqlite] Current julian day number

2006-04-08 Thread Thomas Chust
Hello, is there an elegant way to make the current julian day number the default value for a column in SQLite3? When I try to create a table like this CREATE TABLE time(time REAL DEFAULT julianday(current_timestamp)); I get of course a syntax error, because the calling of functions in the

Re: [sqlite] Small enhancement: BEGIN SHARED

2006-04-08 Thread Paul Bohme
[EMAIL PROTECTED] wrote: > Paul Bohme <[EMAIL PROTECTED]> wrote: > >> I have a small patch that adds "BEGIN SHARED" syntax in addition to >> IMMEDIATE and EXCLUSIVE. I have an application that requires a >> consistent view of the data across a number of individual statements. >> > > This

Re: [sqlite] Small enhancement: BEGIN SHARED

2006-04-08 Thread drh
Paul Bohme <[EMAIL PROTECTED]> wrote: > I have a small patch that adds "BEGIN SHARED" syntax in addition to > IMMEDIATE and EXCLUSIVE. I have an application that requires a > consistent view of the data across a number of individual statements. This is what plain old "BEGIN" does. OK, really BE

[sqlite] Re: Small enhancement: BEGIN SHARED

2006-04-08 Thread Igor Tandetnik
Paul Bohme wrote: I have a small patch that adds "BEGIN SHARED" syntax in addition to IMMEDIATE and EXCLUSIVE. I have an application that requires a consistent view of the data across a number of individual statements. What is ideal is a way to simply lock the database with a SHARED lock for t

Re: [sqlite] Efficient Query Processing

2006-04-08 Thread drh
Martin Pfeifle <[EMAIL PROTECTED]> wrote: > Is it also possible to force SQLite to retrieve all records according > to their pageid. Not as currently implemented. -- D. Richard Hipp <[EMAIL PROTECTED]>

[sqlite] Small enhancement: BEGIN SHARED

2006-04-08 Thread Paul Bohme
I have a small patch that adds "BEGIN SHARED" syntax in addition to IMMEDIATE and EXCLUSIVE. I have an application that requires a consistent view of the data across a number of individual statements. What is ideal is a way to simply lock the database with a SHARED lock for the duration, but t

[sqlite] Efficient Query Processing

2006-04-08 Thread Martin Pfeifle
Hello, We repeatedly have queries like select * from Tab where primary_key between ? and ?. SQLite retrieves all result-records according to their rowids (=primary key in our case). Is it also possible to force SQLite to retrieve all records according to their pageid. So the idea is that

Re: [sqlite] Persistent user-defined functions

2006-04-08 Thread Paul Bohme
Daniel Franke wrote: > Layering. Wrap sqlite3_* into your own set of functions. Create another > library, say libyourapp. Most functions will just forward the arguments to > sqlite, but others, e.g. yourapp_open_db() will not only open the database, > but also attach a couple of functions, which