Hi Andres,

thanks for the detailed explanations!

For other interested parties - there is also some discussion going on on

BTW, I was going to suggest using Dataless as an alternative (supported
by WebDC), but I'm unable to get a usable Inventory object from Dataless:

from obspy import read_inventory
inv =

<class 'obspy.core.inventory.inventory.Inventory'>

from import Parser
sp = Parser("Package_1475048499259-GFZ_73152721_inventory.dlsv")
inv = sp.get_inventory()

<type 'dict'>

The parser way predates the inventory objects and we never had the
urgency/time to make sure (X)SEED hooks into the whole
`read_inventory()` machinery which would indeed be very nice to have.

ObsPy can still work with dataless SEED files but usage is a bit
different, e.g. instrument correction:

AFAICS, (ArcLink) Inventory XML is not supported by read_inventory().
This is surprising for me, because
obspy.clients.arclink.client.Client.get_inventory() actually works with
this format. I suppose it would not be too difficult to refactor this


That's for historic reasons - it does not actually parse to inventory
objects as these also did not exist at the time the arclink client was
written. Most people should be able to just replace their use of arclink
with the fdsn web services.

Nonetheless both of these issues should be addressed in the long run.