[BUG] Unexpected Type Inference in PathExpand

[BingqingLyu]

Describe the bug

On LDBC Dataset (sf=0.1), we can run the following two queries:

# gremlin tests
gremlin> g.V().hasLabel("PERSON").has("id",933).out("5..6").endV().hasLabel("PLACE").has("id",999).count()
==>392

while a simiilar query in cypher is:

# cypher tests
@neo4j> Match (a:PERSON{id:933})-[c*5..6]->(b:PLACE{id:999})
        return a.id, b.id
        Limit 10;
+----------+
| id | id0 |
+----------+
+----------+

The results in the Cypher is empty, which is not as expected. The reason is that, in this case, the type inference tells that the type of c should be isLocatedIn (from the edge type that Person-isLocatedIn-Place). However, in PathExpand, this is not as expected. The path could be like Person-Knows[4..5]-Person-IsLocatedIn-Place, etc.

[github-actions]

/cc @shirly121, this issus/pr has had no activity for for a long time, could you folks help to review the status ?
To suppress further notifications,

• for issues,
  • if it is waiting for further response from the reporter/author, please help to add the label requires-further-info,
  • if you have already started working on it, please add the label work-in-progress to the issue,
  • if this issue requires further designing discussion and not in current plan, or won't be fixed, please add the label requires-further-discussion or wontfix to the issue,
• for pull requests,
  • if you are still working on it and it is not ready for reviewing, please convert this pull request as draft PR,
  • if you have decided to hold this development on, please add the requires-further-discussion label to the pull request.
    Thanks!

[BingqingLyu]

Fixed in #3837

[BingqingLyu]

There leaves a bug in the type infer in PathExpand.

For example, running the following query on LDBC dataset,

MATCH(a)-[*1..2]->(b) RETURN COUNT(a) AS pathCnt;

The inferred type of PathExpandBase within PathExpand is:
ExpandEdge(with direction IN) type ids: {0,1,3,5,6,7,10,11,12,13,15,16,17,21,22};
GetV type ids: {0,1,2,3,5,6,7}

While here the GetV types missed 4, i.e., Forum, while Forum is also a valid type for the expanded vertices (i.e., of a).

I guess, the reason is that the direction is not correctly considered in path, since forum only has out-neighbors in schema, but in this case, the expand direction is IN.