On Sunday 08 September 2013 19:47:12 Thomas Lübking wrote:
> On Sonntag, 8. September 2013 14:19:02 CEST, Pali Rohár wrote:
> > Ok, so I see that everybody is against loading plugins from
> > application directory.
> > 
> > What about this solution?
> > 
> > Load plugins from system path specified at compile time and
> > also from some application subdirectory (plugins). And
> > third posibility so specify custom path via env variable,
> > command line option or config option.
> 
> If you install plugins into the Qt plugin path, there's
> $QT_PLUGIN_PATH anyway. Loading from "./plugins/" is no
> different than loading from "." - basically just the plugin
> name changed from plugin_foo to plugin/foo and this is also
> superseded by an env strategy (since you can put "." into the
> inv list if you really want to)
> 

QPluginLoader (and my current code too) not using $QT_PLUGIN_PATH 
variable, so it will not work.

And my question is what do you think about searching for plugins 
in "./plugins/" directory (instead ".") and in $SOME_ENV_PATH 
directory? It is acceptable?

> > What I do not want is hardcoding build directory into
> > trojita executable when build service compiling package
> > (rpm, deb, ...).
> 
> I do not understand the usecase for this, but it's Jans
> decision anyway.
> 
> > loading plugins from rpaths? It is even possible to read
> > rpath in application?
> 
> Yes, unless you mean "in a portable way that does not include
> inspecting the ELF"
> http://stackoverflow.com/questions/2836330/is-there-a-way-to-> 
> inspect-the-current-rpath-on-linux
> 
> PE/COFF will require different treatment for sure.
> 

No, no, this is really horrible hack and from what I found 
windows does not support equivalent to rpath. So drop this.

-- 
Pali Rohár
[email protected]

Attachment: signature.asc
Description: This is a digitally signed message part.

Reply via email to