Re: [ros-users] REP 116 pluginlib extension

Top Page
Attachments:
Message as email
+ (text/plain)
Delete this message
Reply to this message
Author: User discussions
Date:  
To: User discussions
Subject: Re: [ros-users] REP 116 pluginlib extension
On Tue, Nov 22, 2011 at 2:55 PM, Dirk Thomas <> wrote:
>> - There is a strict separation of build products from what gets
>> checked out of source control; that is, the build and source
>> directories for packageN will be different.  libpackageN.so and
>> packageN/somepython.py will exist in completely different places
>> before installation, and these places will be different when "built"
>> and after installation.
>>
>> -  "/path/to/stacks/stackN/packageN" won't exist, per se.  But
>> /path/to/share/packageN/anything_you_like may exist (if packageN
>> chooses to install things that way).  The point is that /path/to will
>> be the CMAKE_INSTALL_PREFIX, and under this prefix things will be
>> installed in as FHS-ish a scheme as possible.
>
> What package-path will be returned in this "installed" scenario?
> Either when using C++: ros::package::getPath("packageN") or in Python:
> rospkg.RosPack().get_path('packageN')
>


One answer would be "/opt/ros/VERSION" is returned.. and things
speicifc to packageN are installed thereunder in predictable
locations. Which begs the question... why ask rospack for these paths
if you can calculate them a priori?

-t