Support parquet load for Neptune Analytics

ChangeLog.md

@@ -3,6 +3,8 @@
 Starting with v1.31.6, this file will contain a record of major features and updates made in each release of graph-notebook.
 
 ## Upcoming
+- Support edgeOnlyLoad parameter for neptune database bulk load operation ([Link to PR](https://github.com/aws/graph-notebook/pull/750))
+- Support loading parquet format data for neptune analytics incremental load operation ([Link to PR](https://github.com/aws/graph-notebook/pull/752))
 
 
 ## Release 5.1.0 (Oct 26, 2025)

src/graph_notebook/magics/graph_magic.py

@@ -47,7 +47,7 @@
 from graph_notebook.magics.streams import StreamViewer
 from graph_notebook.neptune.client import (ClientBuilder, Client, PARALLELISM_OPTIONS, PARALLELISM_HIGH, \
     LOAD_JOB_MODES, MODE_AUTO, FINAL_LOAD_STATUSES, SPARQL_ACTION, FORMAT_CSV, FORMAT_OPENCYPHER, FORMAT_NTRIPLE, \
-    DB_LOAD_TYPES, ANALYTICS_LOAD_TYPES, VALID_BULK_FORMATS, VALID_INCREMENTAL_FORMATS, \
+    DB_LOAD_TYPES, ANALYTICS_LOAD_TYPES, VALID_BULK_FORMATS, VALID_INCREMENTAL_FORMATS, FORMAT_PARQUET, \
     FORMAT_NQUADS, FORMAT_RDFXML, FORMAT_TURTLE, FORMAT_NTRIPLE, STREAM_RDF, STREAM_PG, STREAM_ENDPOINTS, \
     NEPTUNE_CONFIG_HOST_IDENTIFIERS, is_allowed_neptune_host, \
     STATISTICS_LANGUAGE_INPUTS, STATISTICS_LANGUAGE_INPUTS_SPARQL, STATISTICS_MODES, SUMMARY_MODES, \
@@ -153,7 +153,7 @@
 
 DEFAULT_NAMEDGRAPH_URI = "http://aws.amazon.com/neptune/vocab/v01/DefaultNamedGraph"
 DEFAULT_BASE_URI = "http://aws.amazon.com/neptune/default"
-RDF_LOAD_FORMATS = [FORMAT_NTRIPLE, FORMAT_NQUADS, FORMAT_RDFXML, FORMAT_TURTLE]
+RDF_LOAD_FORMATS = [FORMAT_NTRIPLE, FORMAT_NQUADS, FORMAT_RDFXML, FORMAT_TURTLE, FORMAT_PARQUET]
 BASE_URI_FORMATS = [FORMAT_RDFXML, FORMAT_TURTLE]
 DEFAULT_LOAD_CONCURRENCY = 1
 
@@ -2496,7 +2496,7 @@ def on_button_clicked(b):
                         'parserConfiguration': {},
                     }
 
-                    if source_format.value.lower() == FORMAT_CSV:
+                    if source_format.value.lower() == FORMAT_CSV or source_format.value.lower() == FORMAT_PARQUET:
                         bulk_load_kwargs['edgeOnlyLoad'] = edge_only_load.value
 
                     if dependencies:

src/graph_notebook/neptune/client.py

@@ -53,6 +53,7 @@
 FORMAT_CSV = 'csv'
 FORMAT_OPENCYPHER = 'opencypher'
 FORMAT_NTRIPLE = 'ntriples'
+FORMAT_PARQUET = 'parquet'
 FORMAT_NQUADS = 'nquads'
 FORMAT_RDFXML = 'rdfxml'
 FORMAT_TURTLE = 'turtle'
@@ -69,8 +70,16 @@
 LOAD_JOB_MODES = [MODE_RESUME, MODE_NEW, MODE_AUTO]
 DB_LOAD_TYPES = ['bulk']
 ANALYTICS_LOAD_TYPES = ['incremental']
-VALID_INCREMENTAL_FORMATS = ['', FORMAT_CSV, FORMAT_OPENCYPHER, FORMAT_NTRIPLE]
-VALID_BULK_FORMATS = VALID_INCREMENTAL_FORMATS + [FORMAT_NQUADS, FORMAT_RDFXML, FORMAT_TURTLE]
+VALID_COMMON_FORMATS = ['', FORMAT_CSV, FORMAT_OPENCYPHER, FORMAT_NTRIPLE]
+
+# --------------
+# Currently, Parquet format is only supported for incremental loads, which are exclusively used with Neptune Analytics.
+# Bulk loads (used with Neptune DB) do not support the parquet format.
+# This distinction is handled in the load magic function when processing the format parameter.
+VALID_INCREMENTAL_FORMATS = VALID_COMMON_FORMATS + [FORMAT_PARQUET]
+VALID_BULK_FORMATS = VALID_COMMON_FORMATS + [FORMAT_NQUADS, FORMAT_RDFXML, FORMAT_TURTLE]
+# --------------
+
 PARALLELISM_OPTIONS = [PARALLELISM_LOW, PARALLELISM_MEDIUM, PARALLELISM_HIGH, PARALLELISM_OVERSUBSCRIBE]
 LOADER_ACTION = 'loader'