pkg-config usage for sqlite3

2019-08-12 Thread Kevin J. McCarthy
It looks like building a static binary doesn't work with the sqlite3 option, due to a couple extra dependencies needed in that case. One common, and increasingly prevalent, solution is to use pkg-config and the related autoconf macros (e.g. PKG_CHECK_MODULES). However, this isn't without issu

Re: pkg-config usage for sqlite3

2019-08-12 Thread Oswald Buddenhagen
On Mon, Aug 12, 2019 at 09:15:57AM -0700, Kevin J. McCarthy wrote: > One common, and increasingly prevalent, solution is to use pkg-config > and the related autoconf macros (e.g. PKG_CHECK_MODULES). > However, this isn't without issues, for example dealing with libraries > installed in non-standa

Re: pkg-config usage for sqlite3

2019-08-12 Thread Kevin J. McCarthy
On Mon, Aug 12, 2019 at 07:11:20PM +0200, Oswald Buddenhagen wrote: On Mon, Aug 12, 2019 at 09:15:57AM -0700, Kevin J. McCarthy wrote: One common, and increasingly prevalent, solution is to use pkg-config and the related autoconf macros (e.g. PKG_CHECK_MODULES). However, this isn't without is

Re: pkg-config usage for sqlite3

2019-08-12 Thread Eike Rathke
Hi Kevin, On Monday, 2019-08-12 10:51:56 -0700, Kevin J. McCarthy wrote: > > why would supplying a non-standard search path to pkg-config files be > > any more problematic than a non-standard path to the library itself? > > By all means, please enlighten me. My understanding from starting to >

Re: pkg-config usage for sqlite3

2019-08-12 Thread Kevin J. McCarthy
On Tue, Aug 13, 2019 at 02:39:35AM +0200, Eike Rathke wrote: Yes. Or rather, if "no" then do not check for it, and break if --enable-autocrypt was specified as well. PKG_CHECK_MODULES generates code to output a message if the .pc file wasn't found and to adjust PKG_CONFIG_PATH or how things can

Re: pkg-config usage for sqlite3

2019-08-12 Thread Petr Pisar
On Mon, Aug 12, 2019 at 07:05:19PM -0700, Kevin J. McCarthy wrote: > There's no new issue here with sqlite3. :-/ > Actually I can see an issue with the sole fact that sqlite3 is a new dependency. Before introducing autocrypt to mutt, there was no need for it. I can see that mutt can already use

Re: pkg-config usage for sqlite3

2019-08-13 Thread Kevin J. McCarthy
On Tue, Aug 13, 2019 at 07:42:19AM +0200, Petr Pisar wrote: I can see that mutt can already use Berkely DB, GDBM, lmdb or qdbm for caching headers. Wouldn't be possible to use one of them for autocrypt key storage? I did consider that, but the implementation uses more than simple key-value s

Re: pkg-config usage for sqlite3

2019-08-13 Thread Petr Pisar
On Tue, Aug 13, 2019 at 06:46:48AM -0700, Kevin J. McCarthy wrote: > >Or maybe vice versa implement header caching using sqlite3. Some people > >get nervous when they see a program links to multiple libraries of the > >same class of use. People packaging software do not like it. An > >abundance

Re: pkg-config usage for sqlite3

2019-08-18 Thread Matthias Andree
On 12.08.19 18:15, Kevin J. McCarthy wrote: It looks like building a static binary doesn't work with the sqlite3 option, due to a couple extra dependencies needed in that case. One common, and increasingly prevalent, solution is to use pkg-config and the related autoconf macros (e.g. PKG_CHECK_M