What is the preferred way to create SQLite databases? Ideally, I would like
to have the entire schema in a file, and then generate the DB from that, as
we make changes to the schema often.
We're currently using something called SQLite Database Browser, but I am
becoming concerned that the
My fix is to REMOVE the lines of code I mentioned. The malloc is done
outside of this method already.
> -Original Message-
> From: Alexandre Balaban [mailto:[EMAIL PROTECTED]
> Sent: Thursday, November 29, 2007 11:13 AM
> To: email@example.com
> Subject: [sqlite]
I have finally tracked down a bug that has plagued by vxWorks port. It
appears that our file system's implementation of ftruncate() does not like
to "truncate" a file larger than its current size, and returns EINVAL for
My question...assuming that I can't get ftruncate to
Is there any documentation on how one would run the SQLite tests? I'm
thinking it would be an excellent test for my particular platform port. I'm
seeing the test directory in CVS, but not exactly sure what to do with the
Ah, yes. My mistake. Thanks.
> The locking style has absolutely nothing to do with
> threads. Locking style is what kind of file locks
> are used to prevent separate processes from updating
> the same database as the same time. SQLite uses
> posix advisory locks by default. Not all
I was using the noLocking style. Is that essentially what you get with
SQLITE_THREADSAFE = 0?
> I recommend that you *not* use LOCKING_STYLE. It is
> disabled by default. Dan and I have a note to go back
> and get the LOCKING_STYLE stuff working, but we have not
I am currently migrating over from 3.4 to 3.5.2. I am creating a custom
VFS, but it is essentially the same as unix.
As I was going through the code, there was a particular section that made me
think that it may not have been migrated correctly.
In the fillInUnixFile() method (and assuming
Ah, that makes sense. Thanks!
> I was talking about using the sqlite download page at
> If you right click on a link (like the one for
> sqlite3-3_5_2.zip under
> Precompiled Binaries For Windows) and select "copy the link location"
Ahh...I think I figured it out. Clicking the "Show Milestones" button was
To unsubscribe, send email to [EMAIL PROTECTED]
Thanks for the reply.
How do I know what version of the file was part of 3.4.2 and what is part of
3.5? I'm seeing version numbers of 1.171, for example.
> Go to the download page, copy the link for the file you want
> (with the
> current 3..5.2 version
Could someone please remind me how I can download SQLite code for version
3.4.2 from CVS? Was there some sort of tag made that I can use?
I'm hesitant to upgrade to 3.5.x just because I'm not sure how much it has
been test driven and we are close to release. Maybe I am being overly
I'm currently using SQLite 3.4.1 and have been reading with interest the
large database thread. I learned about the soft heap limit feature and was
considering using it.
While reading about the bug fixes for 3.4.2 and soft heap limit, I came
across a sample of setting the soft heap limit to
Thanks, John, for your reply. We also had the same idea about wanting to
avoid the SQLITE_BUSY possibilities, and just have callers wait.
So, I think if I understand your response correctly, by having THREADSAFE=0
we have effectively commented out any code that would be trying to use
We have an application that wraps SQLite 3.4.1. Due to some vxWorks
platform issues with non-standard posix behavior, we are considering
compiling SQLite with THREADSAFE = 0, and then having our application make
sure that only one thread can access SQLite at a time. I was wondering
We are currently creating databases using this open source tool. According
to the documentation, it is using version 3.3.5 of SQLite. I was wondering
if there are any problems with creating a database with the tool, but then
using the database with an application that is using SQLite
No, not a soft link. :)
Based on other posts I have read about threading performance and SQLite, it
seems like most people like to use a single thread. I'm going to change our
application to use a system-wide mutex for thread synchronization and see if
that improves our results. I'm still
There is a .lock file for each database. From my understanding, that should
prohibit 2 connections from using the same database at the same time.
However, that is not the situation I am wondering about. I am specifically
wondering if database activity on a connection to DB 1 would have
After rereading your post, I wanted to confirm something. In your example
below, are thread1 and thread2 connected to the same database, or different
databases? In my scenario, the threads are connected to different
databases, so I'm not sure if it is the same situation.
Thank you for your insight. Looks like we have some redesign scheduled for
> -Original Message-
> From: Andre du Plessis [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, August 15, 2007 5:05 AM
> To: firstname.lastname@example.org
> Subject: RE: [sqlite]
I've got an application that has many different SQLite databases. Each
database connection is opened in its own thread. Each database has only one
I created some test cases that create a database and schema on the fly and
perform various SELECT, INSERTS, UPDATES on it. The
No, we are actually filling in the parameter with a valid integer value. I
was just trying to say it was a parameter.
> Are you actually searching for records where F is the string "?"
> If so, why don't you try WHERE F="?" instead of leaving it with the
> ? unquoted.
My final post on this issue...vxWorks has confirmed a bug in their fstat()
implementation. I'm hoping for a quick resolution. Thanks for the help.
> Just a follow-up for the group...
> Richard was able to determine that my platform is having
> trouble with an
> fstat() call to get
Just a follow-up for the group...
Richard was able to determine that my platform is having trouble with an
fstat() call to get the size of files on my file system. Specifically, the
call is returning a size of 0 when it should be 11K. I am investigating the
vxWorks platform and have found that
No, I am not confident that my file locking code is working. For vxWorks, I
needed to alter the .lock functionality, as it wouldn't compile. Pretty
much changes such as calling stat() instead of lstat(), etc. However, for
this example, I do know that the two threads operate serially (ie,
I have some odd results to report on the db counter.
I put some diagnostic code in our database wrapper class to write out the db
counter whenever a statement is executed. It will print out for every
statement executed whether the statement is a SELECT or UPDATE or BEGIN
Hi Richard and Joe-
It probably is some application error in our code. What puzzles me is that
calling sqlite3_enable_shared_cache(1) appears to fix (or at least change)
the behavior. Not sure if that triggers any ideas as to what our problem
Your suggestion of looking at the
Thanks for the quick reply. Unfortunately, we are developing code on the
vxWorks platform, so I don't think sample code would be of use. We have
seen the problem for some time now...at least from 3.3.12. The submitter of
the ticket appears to have the exact same scenario as us. Hopefully he
We have a scenario where we have two different database connections to the
same database. Each database connection is running on a separate thread,
but in this situation, we are only using one connection at a time. We are
finding that sometimes one database connection will do a "select" on
Mail list logo