Removing metadata

[jjkoehorst]

I am trying to update the metadata of relatively newly created data_objects.

def set_metadata(element: DataObject, metadata=METADATA):
    for key in metadata:
        if key in element.metadata.keys():
            for remote_metadata in element.metadata.get_all(key):
                element.metadata.remove(remote_metadata)

This however returns

element.metadata.remove(remote_metadata)
  File "/venv/lib/python3.6/site-packages/irods/meta.py", line 63, in remove
    self._manager.remove(self._model_cls, self._path, meta)
  File "/venv/lib/python3.6/site-packages/irods/manager/metadata_manager.py", line 87, in remove
    response = conn.recv()
  File "/venv/lib/python3.6/site-packages/irods/connection.py", line 91, in recv
    raise get_exception_by_code(msg.int_info)
irods.exception.CAT_SUCCESS_BUT_WITH_NO_INFO: None

Process finished with exit code 1

I am trying to figure out what goes wrong here but the error message says its a success but not really?

[jjkoehorst]

Also when try / except this exception it will later on tell me the catalog already has the item but for some reason it does not want to update (delete/add) the item. Or should I use a different approach there?

[d-w-moore]

@jjkoehorst I modified the code example above, just a little, to read:

def remove_metadata(element : DataObject, metadata):
    for key in metadata.keys():
        for item in element.metadata.get_all(key):
            element.metadata.remove(item)
# ...
elem = sess.data_objects.get('/tempZone/home/rods/dataobj')
remove_metadata(elem, elem.metadata)

and I get:

$ imeta ls -d dataobj
AVUs defined for dataObj dataobj:
attribute: z
value: b
units: z
----
attribute: a
value: b
units: f
----
attribute: a
value: b
units: c
$ 
$ python3 removemeta.py 
$ imeta ls -d dataobj
AVUs defined for dataObj dataobj:
None

See if this modified form works for you. If not, I'd be interested in the particular combination of AVUs present on the object when you see the error.

[trel]

@jjkoehorst can you confirm this is behaving with @d-w-moore edits above? if so, please close.