[sqlite] Export database into mySQL

[Igor Korot]

Hi,
Is it possible to load the SQLite schema from the .dump command into mySQL?
Or I will have to do that on per-table?

Thank you.
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] newbee questions

[Simon Slavin]

On 29 Nov 2016, at 12:22am, Scott Cason  wrote:

> Please forgive me if I seem a bit ignorant as I am trying to get my feet on 
> the ground.
> Does anyone know where I can get the information necessary to set up Eclipse 
> CDT for sqlite on a Mac? 



Simon.
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] newbee questions

[Scott Cason]

Please forgive me if I seem a bit ignorant as I am trying to get my feet on the 
ground.
Does anyone know where I can get the information necessary to set up Eclipse 
CDT for sqlite on a Mac?  I know that Mac users, almost all, use Xcode but I am 
really wanting to use Eclipse since I can run it on both Mac and PC.
Also, is the binary download for Mac 64 bit?  I thought that I ran across 
something somewhere that made me  think that I needed to compile the sources to 
enable 64 bit?.?.?  Maybe it’s just my memory going.
Thanks,Scott
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Read-only access which does not block writers

[Nico Williams]

On Mon, Nov 28, 2016 at 07:45:16PM +, Howard Chu wrote:
> Nico Williams wrote:
> >One of the problems with LMDB's locks is that readers need write
> >permission to the lock file, IIRC :( but at least it's a separate file.
> 
> We developed an alternate locking protocol for Postfix. It just uses
> standard fcntl() locks.

Yeah, I'm aware.

> http://www.postfix.org/lmdb_table.5.html
> 
> Of course nothing comes for free - with this approach, writers don't block
> readers, but readers block writers. That's the best you're going to get
> without custom lock protocols like LMDB uses natively.

TANSTAAFL, I know.  A proper COW DB with coalesce-and-rename-into-place
would need no locking of any kind for readers, at the price of needing
more storage and more I/O to do the vacuuming.
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Read-only access which does not block writers

[Howard Chu]

Nico Williams wrote:

On Fri, Nov 25, 2016 at 09:08:37AM +0100, Florian Weimer wrote:

I think you misunderstood what I was asking (see the start of the thread).
I need to support Byzantine readers which do not follow the locking
protocol.  Based on the documentation, LMDB uses locks to implement MVCC and
prevent premature page reuse.



There's a good use case for an LMDB-like COW DB such that readers need
never lock.  That would require vacuuming by writing a new file and
renaming it into place, which is a trade-off.

One of the problems with LMDB's locks is that readers need write
permission to the lock file, IIRC :( but at least it's a separate file.


We developed an alternate locking protocol for Postfix. It just uses standard 
fcntl() locks.


http://www.postfix.org/lmdb_table.5.html

Of course nothing comes for free - with this approach, writers don't block 
readers, but readers block writers. That's the best you're going to get 
without custom lock protocols like LMDB uses natively.


--
  -- Howard Chu
  CTO, Symas Corp.   http://www.symas.com
  Director, Highland Sun http://highlandsun.com/hyc/
  Chief Architect, OpenLDAP  http://www.openldap.org/project/
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Read-only access which does not block writers

[Nico Williams]

On Fri, Nov 25, 2016 at 09:08:37AM +0100, Florian Weimer wrote:
> I think you misunderstood what I was asking (see the start of the thread).
> I need to support Byzantine readers which do not follow the locking
> protocol.  Based on the documentation, LMDB uses locks to implement MVCC and
> prevent premature page reuse.

There's a good use case for an LMDB-like COW DB such that readers need
never lock.  That would require vacuuming by writing a new file and
renaming it into place, which is a trade-off.

One of the problems with LMDB's locks is that readers need write
permission to the lock file, IIRC :( but at least it's a separate file.

Nico
-- 
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users