Re: [sqlite-dev] SQLite packaging, again

2011-12-15 Thread Roger Binns
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 15/12/11 15:52, Jaroslaw Staniek wrote: > sqlite3_api is uninitialized, no reason it should be since I am not > using redefined API functions. It is initialized for dynamically loaded extensions. See this somewhat dated page: http://www.sqlite

Re: [sqlite-dev] SQLite packaging, again

2011-12-15 Thread Jaroslaw Staniek
On 14 December 2011 18:56, Roger Binns wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > On 14/12/11 09:34, Jaroslaw Staniek wrote: >> But I did not consider using Statically Linked Extensions via >> sqlite3_auto_extension. > > They aren't statically linked in the conventional sense. >

Re: [sqlite-dev] SQLite packaging, again

2011-12-14 Thread Roger Binns
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 14/12/11 09:34, Jaroslaw Staniek wrote: > But I did not consider using Statically Linked Extensions via > sqlite3_auto_extension. They aren't statically linked in the conventional sense. > The issue arise from the fact that there is at least one

Re: [sqlite-dev] SQLite packaging, again

2011-12-14 Thread Roger Binns
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Why are you posting to sqlite-dev? I don't see any relevancy. SQLite does not have collations builtin. There is an API where you can register whatever collations you want: http://www.sqlite.org/c3ref/create_collation.html There is another API wh

Re: [sqlite-dev] SQLite packaging, again

2011-12-14 Thread Jaroslaw Staniek
On 14 December 2011 18:13, Roger Binns wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Why are you posting to sqlite-dev?  I don't see any relevancy. > > SQLite does not have collations builtin.  There is an API where you can > register whatever collations you want: > >  http://www.sq