Add timeout parameter to yarp detect and yarp where

[Nicogene]

trafficstars

Is your feature request related to a problem? Please describe. When you do yarp detect or yarp where and in your .conf file you have an unreachable IP, it get stuck forever, because it is not possible to set timeout argument to the Contact and the default timeout is -1:

https://github.com/robotology/yarp/blob/ce6f008c86ba58a652b04b6bffdc2b46e9b157e2/src/libYARP_os/src/yarp/os/Contact.cpp#L60

This problem can be reproduced as:

yarp conf 1.2.3.4 10000

And then it get stuck like this:

Looking for name server on 1.2.3.4, port number 10000
If there is a long delay, try:
  yarp conf --clean

This is particular annoying if you use this commands in scripts because you have to stop everything, run yarp conf --clean and then restart the scripts again.

Describe the solution you'd like Add a timeout argument for the two companion commands

[Nicogene]

If you agree on this solution I can open the PR by myself

[randaz81]

I tried your test case with ip 1.2.3.4, but i cannot reproduce it. Both yarp detect and where exit correctly with a messagr after few seconds. Maybe there is something more that I dont know?

[randaz81]

I wonder if you were using some virtualized environment or WSL or something else which may block the multicast packets sent to find if a server is running on the network. (just an hypothesis, because I cannot reproduce your issue)

[Nicogene]

I tried your test case with ip 1.2.3.4, but i cannot reproduce it. Both yarp detect and where exit correctly with a messagr after few seconds. Maybe there is something more that I dont know?

Sorry for the late reply! If I remember correctly this issue was on the laptop of the robot (then a Ubuntu 20.04) but I can double-check