Tom Eulenfeld

Hi, sorry, I cannot provide a good reference. I only googled my way through the theory. If you find a good reference, please post it here. Thanks.

Hi, thanks for the short report. Probably nobody tried to use the filter functions with matrices up til now. Seems like a nice addition. We could also add the `axis`...

Hi Liam, thanks for reporting this issue. This behaviour could definitely be improved. It's not only a problem for data of different stations and one event, but also for a...

> What do you mean by not correctly ordered? Considering a stream of traces `ST1..Z, ST2..N, ST1..E, ST2..Z, ST1..N, ST2..E` rotate will mix data of different stations.

> it looks like more magic introduced by this PR, I think I disagree, because this PR introduces minimal changes. It just looks up the baz and inci in the...

Maybe I was a bit fast. This code is present at multiple places in obspy: ``` numeric_level = getattr(logging, loglevel.upper(), None) if not isinstance(numeric_level, int): raise ValueError('Invalid log level: %s'...

I narrowed the scope of this PR. It now just adds the command line script. The original issue seems to be resolved.

It is not clear to me why the new warning is raised.

I reproduced the bug with your example code. I am baffled. I do not know why, but adding this line in `Response.recalculate_overall_sensivitity()` solves the issue: ```diff --- a/obspy/core/inventory/response.py +++ b/obspy/core/inventory/response.py...

I'll close this issue as it appears to be a conda packaging problem which hopefully will vanish as time goes bye.