Removing an entire tree in Neptune using Gremlin

Question

I'm new to using Gremlin + Neptune. I'm writing a small python program to cleanup some edges + nodes in a Neptune DB. To start, I want to delete all nodes + edges rooted at a particular node.

After looking around, I came across How to delete all child nodes when parent node deleted using the gremlin query?, and I tried the following:

g.V().has("id", <insert id>).union(__(), __.repeat(__.out()).emit()).drop()

However, the node with "id" still exists in the DB. When I try running the above command straight on the python3 console, I get the following output:

>>> g.V().has("id", <insert id>).union(__(), __.repeat(__.out()).emit()).drop()
[['V'], ['has', 'id', '5c4266a3a44649ddb24ce6cf4f831300'], ['union', <gremlin_python.process.graph_traversal.__ object at 0x7fa51c967fd0>, [['repeat', [['out']]], ['emit']]], ['drop']]

What's going on here? It almost looks like it's listing out the steps it plans to take, but not executing them.

Edit -- From answer below, I added .iterate() to the end. I get this error now:

Traceback (most recent call last):
  ...
    tg.g.V().has("id", <id>).union(__(), __.repeat(__.out()).emit()).drop().iterate()
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/process/traversal.py", line 66, in iterate
    try: self.nextTraverser()
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/process/traversal.py", line 71, in nextTraverser
    self.traversal_strategies.apply_strategies(self)
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/process/traversal.py", line 573, in apply_strategies
    traversal_strategy.apply(traversal)
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/driver/remote_connection.py", line 149, in apply
    remote_traversal = self.remote_connection.submit(traversal.bytecode)
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/driver/driver_remote_connection.py", line 55, in submit
    result_set = self._client.submit(bytecode)
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/driver/client.py", line 111, in submit
    return self.submitAsync(message, bindings=bindings).result()
  File "/usr/lib/python3.7/concurrent/futures/_base.py", line 428, in result
    return self.__get_result()
  File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
    raise self._exception
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/driver/connection.py", line 66, in cb
    f.result()
  File "/usr/lib/python3.7/concurrent/futures/_base.py", line 428, in result
    return self.__get_result()
  File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
    raise self._exception
  File "/usr/lib/python3.7/concurrent/futures/thread.py", line 57, in run
    result = self.fn(*self.args, **self.kwargs)
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/driver/protocol.py", line 74, in write
    request_id, request_message)
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/driver/serializer.py", line 132, in serialize_message
    message = self.build_message(request_id, processor, op, args)
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/driver/serializer.py", line 142, in build_message
    return self.finalize_message(message, b"\x21", self.version)
  File "/home/vagrant/.cache/bazel/_bazel_vagrant/f9910d98673307a31f928c448bd4acd0/execroot/project/bazel-out/k8-fastbuild/bin/path/to/directory/python_code.runfiles/pip_deps/pypi__gremlinpython/gremlin_python/driver/serializer.py", line 145, in finalize_message
    message = json.dumps(message)
  File "/usr/lib/python3.7/json/__init__.py", line 231, in dumps
    return _default_encoder.encode(obj)
  File "/usr/lib/python3.7/json/encoder.py", line 199, in encode
    chunks = self.iterencode(o, _one_shot=True)
  File "/usr/lib/python3.7/json/encoder.py", line 257, in iterencode
    return _iterencode(o, 0)
  File "/usr/lib/python3.7/json/encoder.py", line 179, in default
    raise TypeError(f'Object of type {o.__class__.__name__} '
TypeError: Object of type __ is not JSON serializable

Answer 1

Unlike the Gremlin Console, the Python Console doesn't automatically iterate your traversals for you. As in your application code you must iterate them yourself. When you do

g.V().has("id", <insert id>).union(__.identity(), __.repeat(__.out()).emit()).drop()

that simply spawns a Traversal object but does not execute it. You must therefore iterate it in some way to exhaust the elements within it - in your case, the appropriate terminator to use would be iterate():

g.V().has("id", <insert id>).union(__.identity(), __.repeat(__.out()).emit()).drop().iterate()

The semantics around drops in TinkerPop aren't always consistent unfortunately. TinkerPop tried to preserve flexibility for providers in how they implement that but it causes confusion sometimes because a query will work fine in TinkerGraph but then behave slightly differently when executed on a different provider. If the above approach only drops the root, you could try to realize the results before the drop:

g.V().has("id", <insert id>).
  union(__.identity(), 
        __.repeat(__.out()).emit()).
  fold().unfold().
  drop()

Looks a bit dumb but will force all the vertices you wish to drop to be traversed into a list before they are dropped. In that way, you won't kill the repeat() by dropping the parent and its edges first and leaving the child paths untravelled.