Re: plugin plans, -I

Felix Salfelder <[email protected]>
Newsgroups gmane.comp.gnu.gnucap.devel
Message-ID <[email protected]>
On Sat, Feb 14, 2015 at 11:07:21AM -0500, al davis wrote:
> but I do as much of that as I can with plugins, out of tree.

i'd be better off without 'variants' (and just plugins). just not there
yet.

> gnucap-uf is static linked, old pre-split configuration.

some plugins are compiled in mainly because of the s_dc output plugin
linkage problem. this blocks the switch to libgnucap.so and plugins.so
(this is really bad). i dont know how the output plugins are meant to
look like...

but not all plugins are compiled in. the architecture of -uf is still
the same, nevermind the file locations. the relevant parts have been
backported to the "autotools" branch. what i wrote about include
paths/variants etc. still applies there.

> Now, lib, main, modelgen, and plugins (apps directory) are
> considered separate, asynchronous.

same with "modules" in gnucap-uf, though most plugins are singled
out and part of extra packages.

cheers
felix

PS: i was wondering whether a symlink ${srcdir}/include/gnucap -> .
might reduce the headache of #include "gnucap/header.h". well, it feels
suboptimal, doesn't it?
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.