discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Discuss-gnuradio] TypeError for gr_vector_float arg


From: John Murphy
Subject: [Discuss-gnuradio] TypeError for gr_vector_float arg
Date: Tue, 5 Aug 2014 17:08:31 -0400

Trying to get my OOT block to accept a vector of floats for filter taps.
I used a couple in-tree blocks (mostly ofdm_carier_allocator and 
fir_interpolator) for a quick reference. And I have fir interp/decim in my 
flowgraph and they have no issues at runtime. The flowgraph ran fine with my 
OOT block before I decided to try adding support for modifying the filter taps 
from GRC.

Now I am getting the following runtime error when starting the flowgraph in GRC:

Traceback (most recent call last):
  File "/usr/local/share/gnuradio-InTree-modules/ofdm_tx_rx.py", line 518, in 
<module>
    tb = ofdm_tx_rx()
  File "/usr/local/share/gnuradio-InTree-modules/ofdm_tx_rx.py", line 314, in 
__init__
    self.comso_ofdm_rx_0 = comso.ofdm_rx((sync_taps), nfft, ncp, nrp, nhst, 
thr, sel)
  File "/usr/local/lib64/python2.7/site-packages/comso/comso_swig.py", line 
398, in make
    return _comso_swig.ofdm_rx_make(*args, **kwargs)
TypeError: in method 'ofdm_rx_make', argument 1 of type 'gr_vector_float const 
&'

It is really hard to find similar errors on the web, but there are a couple. 
None exact and not many where a solution is indicated. However, it seems likely 
that SWIG has a problem of some sort converting a tuple of floats in python 
(which is typically what the taps for filters look like that I have seen so 
far) to a vector<float> in C++. Or, rather, it has a problem unless you take 
the required steps. I think there is the the equivalent of an include missing 
in SWIG? Something dot i with some syntax? I am brand new to SWIG and python.

I tried to add this capability to an existing OOT block of mine with a ton of 
existing code and did not want to start over with gr_modtool which perhaps 
would have built the SWIG interfaces properly if it had known there would be a 
vector parameter involved. But that is probably what I will try next, just have 
to save off and paste stuff back in. I don't know, I am guessing that may be 
the case and the reason why this is such a rarely found error.

But does anyone know what I need to do in order to make this run?



reply via email to

[Prev in Thread] Current Thread [Next in Thread]