Re: [Discuss-gnuradio] Python block help

[Marcus Müller]

I spoke too soon – general blocks seem to be well-supported in python. My memory might come from a time where gr_modtool didn't have a template for general blocks. So, yeah, you're right, we have the magic in place needed to let Python classes build upon the simple block (non-fixed-rate).

So, what I did was use gr_modtool to make a new module, and then went ahead and used `gr_modtool add -l python -t general` to add a new python general block.

Best regards,

Marcus

On 05/26/2017 02:22 AM, Zach Morris wrote:

Hmm, interesting. The code below suggests that the Python interface used to support general blocks. I wonder if it could still be implemented by extending basic_block. 

Do you know if there are any good tutorials on using the consume() or forecast() functions in Python? The Python tutorials I've seen seem to stop right after mentioning that those blocks exist.

Zach

On May 25, 2017 2:25 PM, "Marcus Müller" <address@hidden> wrote:

Hi Zach,

not sure the Python interface supports general blocks (ie. blocks where
the number of output items is not a multiple of the input items), I hope
someone else can comment; I'd recommend implementing this as a C++
block; the "Guided Tutorials" on http://tutorials.gnuradio.org should
prepare you to do that.

Best regards,

Marcus


On 25.05.2017 18:08, Zach Morris wrote:

> Hello Marcus,
>
> I realize this is a couple years later but I am attempting to do the same
> type of arbitrary ratio block, where samples above a threshold are passed
> and samples below that threshold are dropped.
>
> When I tried to subclass gr.block (as below, and in  this tutorial from 2014
> <https://github.com/guruofquality/grextras/wiki/Blocks-Coding-Guide#arbitrary-ratio-block>
> ) in an embedded Python block, GRC threw an error saying gr.block could not
> be found. Does this method still work, or is there an updated method with
> basic_block to implement arbitrary ratio blocks in Python?
>
> Greetings from Menlo Park, CA,
>
> Zach
>
>
>
> Marcus Müller-3 wrote
>> Hi Bob,
>>
>> I think you've introduced the "j" variable to keep count of how many
>> items you're going to produce, but then just tell the scheduler you've
>> produced as many items as he offered you to do. Replace
>>     self.produce(0,len(out0))
>> by
>>     self.produce(0,j).
>> Also, you consume ninput_items every for loop iteration, so
>> ninput_items^2 per work run. That's not right; do it only once, after
>> the loop.
>>
>> Greetings,
>>
>> Marcus
>>
>> On 12/24/2014 12:21 PM, bob wole wrote:
>>> Hi list,
>>>
>>> I am writing a custom python block that should take complex input,
>>> check magnitude of incoming samples and if the magnitude is greater
>>> than a threshold value, the block should pass that sample otherwise
>>> the block just drop the samples. As this is an arbitrary ratio block I
>>> derived it from gr.block and set_auto_consume(False).
>>>
>>> However I get intermittent zeros in output stream of my custom block.
>>> Below is the code
>>>
>>> from gnuradio import gr
>>> import gnuradio.extras
>>> import math
>>> import numpy as np
>>>
>>>
>>> class sdr_pass_valid(gr.block):
>>>     """
>>>     """
>>>     def __init__(self,threshold):
>>>         gr.block.__init__(
>>>             self,
>>>             name = "VALID",
>>>             in_sig = [np.complex64],
>>>             out_sig = [np.complex64],
>>>         )
>>>     self.set_auto_consume(False)
>>>
>>>     self.threshold =  threshold
>>>     def forecast (self,noutput_items,ninput_items_required):
>>>     for i in range(len(ninput_items_required)):
>>>         ninput_items_required[i] = noutput_items
>>>
>>>     def work(self, input_items, output_items):
>>>
>>>         in0 = input_items[0][:len(output_items[0])]
>>>     out0= output_items[0]
>>>         nread = self.nitems_read(0) #number of items read on port 0
>>>         ninput_items = len(in0)
>>>     j=0
>>>     for i in range(0,ninput_items):
>>>         if np.absolute(in0[i]) >= self.threshold :
>>>             out0[j] = in0[i]
>>>             j = j + 1
>>>         self.consume(0,ninput_items)
>>>     self.produce(0,len(out0))
>>>         return 0
>>>
>>>
>>> --
>>> Bob
>>>
>>>
>>> _______________________________________________
>>> Discuss-gnuradio mailing list
>>>
>> Discuss-gnuradio@
>>> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>>
>> _______________________________________________
>> Discuss-gnuradio mailing list
>> Discuss-gnuradio@
>> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>
>
>
>
> --
> View this message in context: http://gnuradio.4.n7.nabble.com/Python-block-help-tp51706p64046.html
> Sent from the GnuRadio mailing list archive at Nabble.com.
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio


_______________________________________________
Discuss-gnuradio mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio