Re: Python embedded block stops working after consume()

Yeah, I know it's strange and I shouldn't need to use any of those. But, the -2 in return is because that's the value of WORK_CALLED_PRODUCE which doesn't exist in Python, so that's why we put the -2 directly. And so this doesn't terminate flowgraph. At least that's what I understand.

On Thu, Oct 28, 2021 at 1:41 PM Jeff Long <willcode4@gmail.com> wrote:

You should not call produce() or consume() in a sync block, actually. The framework does those for you, using the return value from work(). If you return -2, you're manually terminating the flowgraph from this block? Glad it's working for you, in any case.

On Thu, Oct 28, 2021 at 6:24 AM Verónica Toro Betancur <vetorobe@gmail.com> wrote:

Nevermind now, I figured it out!

It turns out that using produce() instead of consume() does the trick, but I couldn't say why. So, the last three lines should be

        output_items[0][:] = input_items[0]
        self.produce(0, len(output_items[0]))
        return -2

Although, in the following block, I just returned len(output_items[0]) without using consume() or produce() and it works fine. So, apparently, my code is a bit random, but I'm happy it works now. :)

Thanks for your help, Jeff!

Best regards,
Verónica

On Thu, Oct 28, 2021 at 8:47 AM Verónica Toro Betancur <vetorobe@gmail.com> wrote:

I see but no, I'm not processing a finite number of samples. I'm generating a message periodically with a Message Strobe, then the message is encoded and modulated as a WiFi signal using the modules in: https://github.com/bastibl/gr-ieee802-11. Now, I'm trying to detect the signals myself and then decode them with the blocks from the same gr-ieee802-11 repository. The block I'm implementing in python looks like this:

class short_sync_blk(gr.sync_block):
    def __init__(self):
        gr.sync_block.__init__(self,name='Custom short preamble detection', in_sig=[np.complex64], out_sig=[np.complex64])

        <+define some variables+>

    def work(self, input_items, output_items):
        print(input_items[0])   # Here is where I see that, after one iteration, input_items[0] is filled with zeros
        <+some processing+>

        <+insert a tag+>
        output_items[0][:] = input_items[0]
        self.consume_each(len(output_items[0]))
        return len(output_items[0])

Is this enough? Or should I put all the actual code in here with all the connected blocks?

Thanks in advance.

Best regards,
Verónica