Stephen Mallette created TINKERPOP-2935:
-------------------------------------------
Summary: Better extensibility around user supplied ids with
certain test patterns for gherkin
Key: TINKERPOP-2935
URL: https://issues.apache.org/jira/browse/TINKERPOP-2935
Project: TinkerPop
Issue Type: Improvement
Components: test-suite
Affects Versions: 3.6.2
Reporter: Stephen Mallette
Using this test in {{MergedEdge.feature}} as an example:
{code}
@UserSuppliedVertexIds
Scenario: g_mergeEXlabel_knows_out_marko_in_vadasX
Given the empty graph
And the graph initializer of
"""
g.addV("person").property(T.id, 100).property("name", "marko").
addV("person").property(T.id, 101).property("name", "vadas")
"""
And using the parameter xx1 defined as "m[{\"t[label]\": \"knows\",
\"D[OUT]\":\"v[100]\", \"D[IN]\":\"v[101]\"}]"
And the traversal of
"""
g.mergeE(xx1)
"""
When iterated to list
Then the result should have a count of 1
And the graph should return 1 for count of
"g.V().has(\"person\",\"name\",\"marko\").out(\"knows\").has(\"person\",\"name\",\"vadas\")"
{code}
it correctly tags with {{UserSuppliedVertexIds}} but if the graph doesn't
support {{Numeric}} ids then it's going to be in trouble. Ultimately, these
sorts of tests are important for ensuring the complex semantics of {{mergeV()}}
and {{mergeE()}} are supported properly so it's bad that graphs will have to
line-item ignore them.
A rough idea would be to let parameters be applied to complex parameters like
{{Map}} and {{List}} and to apply parameters to "the graph initializer" the
same way we do to "the traversal of". A quick test (see attached diff file)
shows that this works for Java and this revision to the test structure:
{code}
@UserSuppliedVertexIds
Scenario: g_mergeEXlabel_knows_out_marko_in_vadasX
Given the empty graph
And using the parameter vid100 defined as "v[100].id"
And using the parameter vid101 defined as "v[101].id"
And using the parameter xx1 defined as "m[{\"t[label]\": \"knows\",
\"D[OUT]\":\"v[vid100]\", \"D[IN]\":\"v[vid101]\"}]"
And the graph initializer of
"""
g.addV("person").property(T.id, vid100).property("name", "marko").
addV("person").property(T.id, vid101).property("name", "vadas")
"""
And the traversal of
"""
g.mergeE(xx1)
"""
When iterated to list
Then the result should have a count of 1
And the graph should return 1 for count of
"g.V().has(\"person\",\"name\",\"marko\").out(\"knows\").has(\"person\",\"name\",\"vadas\")"
{code}
Not sure if this is ideal but it works. Open to other solutions that might be
better if anyone can suggest any.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
