Skip to content

for neo4j integration please also add node labels and relationship-type to graphistry data #101

Closed
Closed
for neo4j integration please also add node labels and relationship-type to graphistry data#101

Description

@jexp
jexp
opened on Dec 5, 2018

currently it only copies id and properties:

def bolt_graph_to_edges_dataframe(graph):
    import pandas as pd
    return pd.DataFrame([
        util.merge_two_dicts(
            { key: value for (key, value) in relationship.items() },
            {
                relationship_id_key:    relationship.id,
                start_node_id_key:          relationship.start_node.id,
                end_node_id_key:     relationship.end_node.id
            }
        )
        for relationship in graph.relationships
    ])


def bolt_graph_to_nodes_dataframe(graph):
    import pandas as pd
    return pd.DataFrame([
        util.merge_two_dicts(
            { key: value for (key, value) in node.items() },
            {
                node_id_key: node.id
            }
        )
        for node in graph.nodes
])

Activity

  1. lmeyerov commented on Dec 13, 2018

    @lmeyerov
    lmeyerov
    on Dec 13, 2018
    Contributor

    Agreed, both for pygraphistry and the direct connector. The type mapping fits naturally, such as with the legend.

    It's a little unclear how we should do labels for the table -- maybe column label1=True, and once we support array-types, label=[label1, ...].

  2. jexp commented on Dec 13, 2018

    @jexp
    jexp
    on Dec 13, 2018
    Author

    Sounds good to me

  3. graphistry
    deleted a comment from jexp on Dec 15, 2018

  4. cwharris commented on Dec 15, 2018

    @cwharris
    cwharris
    on Dec 15, 2018
    Contributor

    Deleted a duplicate comment. Didn't realize it would leave a note in it's place, but it makes sense! Backfire! :)

  5. cwharris commented on Dec 15, 2018

    @cwharris
    cwharris
    on Dec 15, 2018
    Contributor

    @jexp As I work on pygraphisty v1 (which will be compatible with a future release), I'll be making attempting to make the neo4j/pygraphistry datasets isomorphic, including labels, relationship types, etc. What that looks like exactly is yet to be determined, but I'll refer back for community input as the design choices become more preference-based. :)

  6. lmeyerov commented on Dec 16, 2018

    @lmeyerov
    lmeyerov
    on Dec 16, 2018
    Contributor

    image

    Looks like the enterprise connector does the right thing out-of-the-box, we just need pygraphistry to do the same. Agreed, let's do it on the arrow branch for jan.

  7. lmeyerov commented on Feb 22, 2019

    @lmeyerov
    lmeyerov
    on Feb 22, 2019
    Contributor

    Already been in the enterprise studio, and now in notebooks as well with https://github.com/graphistry/pygraphistry/releases/tag/0.9.63 . Win!

  8. lmeyerov
    closed this as completedon Feb 22, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions