Gremlin-Python and Janusgraph Gremlin-Server Full-Text Predicates (textContains*)

[Nick Meves]

Hey everyone,

Does anyone have any experience successfully extending (or monkeypatching or using Lambdas) the gremlin-python library to add the additional full-text search capabilities Janusgraph provides on top of its elasticsearch index?

Particularly:

import static org.janusgraph.core.attribute.Text.*
g.V().has('booksummary', textContains('unicorns'))
g.V().has('booksummary', textContainsPrefix('uni'))
g.V().has('booksummary', textContainsRegex('.*corn.*'))
g.V().has('booksummary', textContainsFuzzy('unicorn'))

I imagine a lot of monkeypatching to __, GraphTraversal & statics in this module:
https://github.com/apache/tinkerpop/blob/master/gremlin-python/src/main/jython/gremlin_python/process/graph_traversal.py

Doesn't anything else need to happen on the gremlin-server python/jython engine side to register those predicates as well?

Any help or peer confirmation that someone else has successfully done this would be greatly appreciated!

Thanks all,
Nick

[HadoopMarc]

Hi Nick,

You may want to check out this pre-alpha work (not my own work, but from janusgraph committers):

https://github.com/JanusGraph/janusgraph-python/tree/initial

In particular, look for a unit test that does what you want.

Cheers,    Marc

Op maandag 8 april 2019 20:01:58 UTC+2 schreef Nick Meves:

[Nick Meves]

Thanks Marc.  Not quite seeing what I need, but gives me ideas.  I'm going to give this a shot -- fingers crossed it works:

from gremlin_python import statics
from gremlin_python.process.traversal import P

class FullTextP(P):
    """Extend P with JanusGraph Full-Text search predicates"""
    def __init__(self, operator, value, other=None):
        P.__init__(self, operator, value, other)

    @staticmethod
    def textContains(*args):
        return FullTextP("textContains", *args)

    @staticmethod
    def textContainsFuzzy(*args):
        return FullTextP("textContainsFuzzy", *args)

    @staticmethod
    def textContainsPrefix(*args):
        return FullTextP("textContainsPrefix", *args)

    @staticmethod
    def textContainsRegex(*args):
        return FullTextP("textContainsRegex", *args)

    @staticmethod
    def textFuzzy(*args):
        return FullTextP("textFuzzy", *args)

    @staticmethod
    def textPrefix(*args):
        return FullTextP("textPrefix", *args)

    @staticmethod
    def textRegex(*args):
        return FullTextP("textRegex", *args)

def textContains(*args):
    return FullTextP.textContains(*args)
statics.add_static('textContains', textContains)

def textContainsFuzzy(*args):
    return FullTextP.textContainsFuzzy(*args)
statics.add_static('textContainsFuzzy', textContainsFuzzy)

def textContainsPrefix(*args):
    return FullTextP.textContainsPrefix(*args)
statics.add_static('textContainsPrefix', textContainsPrefix)

def textContainsRegex(*args):
    return FullTextP.textContainsRegex(*args)
statics.add_static('textContainsRegex', textContainsRegex)

def textFuzzy(*args):
    return FullTextP.textFuzzy(*args)
statics.add_static('textFuzzy', textFuzzy)

def textPrefix(*args):
    return FullTextP.textPrefix(*args)
statics.add_static('textPrefix', textPrefix)

def textRegex(*args):
    return FullTextP.textRegex(*args)
statics.add_static('textRegex', textRegex)

[Florian Hockmann]

What's missing in that library for text predicates? What you're looking for should be in this class:

https://github.com/JanusGraph/janusgraph-python/blob/initial/src/main/python/janusgraph_python/core/attribute/Text.py