Re: [Discuss-gnuradio] Analyzing an Arbitrary Length Sequence in a Python Block

[Kevin Reid]

On Sun, Oct 15, 2017 at 10:59 PM, Gilad Beeri (ApolloShield) <address@hidden> wrote:

I'm writing a Python block that, in general, does the following:

Input: a stream of numbers (assume integers for the sake of discussion).

Output: a stream of numbers

… 

Note: The sequences can be of arbitrary lengths.

If you need to look at an arbitrary length of samples, it would likely be best for your block to store the samples it receives in a buffer it manages, not try to convince GNU Radio to store them for you.

More narrowly, if the only _sample_ you want to output is the max value (you talked about "tagging the max value" but your example output sequences also didn't contain any "untagged" samples, so I'm not sure if that's what you actually meant), you should be able to just track the max-so-far value and not even need an unbounded buffer.