Re: [Discuss-gnuradio] Patching airprobe to compile with GNURadio 3.7

[Tom Rondeau]

On Sun, Nov 23, 2014 at 5:10 PM, jkkjjkjk kjjkjkjkj <address@hidden> wrote:

Hey, I've spent the last couple of days trying to get airprobe rtl GSM software to work with the new GNU Radio.
I made a patch to grcompat to fill in the missing functions:
https://github.com/genjix/grcompat
And have got ported the gsm module for gsm-receive/src/lib/ to work. And I've ported gsm_receive_rtl for all the changes in module structure in GNU Radio. Now the program opens but I'm stuck at one last step.

Here is the line giving me difficulty:
https://github.com/genjix/airprobe/blob/master/gsm-receiver/src/python/gsm_receive_rtl.py#L98

#self.connect(self.src, self.tuner, self.interpolator, self.receiver, self.converter, self.output)
self.connect(self.src, self.tuner, self.interpolator)

I've commented the original which gives me this error:

configure_receiver
Traceback (most recent call last):
File "./gsm_receive_rtl.py", line 232, in <module>
tb = top_block()
File "./gsm_receive_rtl.py", line 99, in __init__
self.connect(self.src, self.tuner, self.interpolator, self.receiver)
File "/usr/local/lib/python2.7/dist-packages/gnuradio/gr/top_block.py", line 131, in connect
self._connect(points[i-1], points[i])
File "/usr/local/lib/python2.7/dist-packages/gnuradio/gr/top_block.py", line 141, in _connect
(dst_block, dst_port) = self._coerce_endpoint(dst)
File "/usr/local/lib/python2.7/dist-packages/gnuradio/gr/top_block.py", line 152, in _coerce_endpoint
raise ValueError("unable to coerce endpoint")
ValueError: unable to coerce endpoint

Here is the source for the self.receiver type (a custom block for GSM receiving):

https://github.com/genjix/airprobe/blob/master/gsm-receiver/src/lib/gsm_receiver_cf.cc#L305

Any clues? What is the problem with gsm_receiver_cf that GNU Radio gives that error? It used to work with 3.6...

Thanks

This is why I'm against these kinds of compatibility headers. They just encourage the use of a project without ever updating it. As things continue to move in the main development, you can get left behind. Honestly, your work to fix the grcompat header would have been better spent updating airprobe. You've seen how it's just simple semantic/namespace changes by updating grcompat. Updating this code should not be difficult and would then make it more clear what kind of issues are going wrong. As it is, we have a third layer in between this code and GNU Radio obscuring things.

Tom