maropu commented on a change in pull request #24851: [SPARK-27303][GRAPH] Add 
Spark Graph API
URL: https://github.com/apache/spark/pull/24851#discussion_r339282647
 
 

 ##########
 File path: 
graph/api/src/main/scala/org/apache/spark/graph/api/CypherSession.scala
 ##########
 @@ -0,0 +1,291 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *    http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.graph.api
+
+import scala.collection.JavaConverters._
+
+import org.slf4j.LoggerFactory
+
+import org.apache.spark.annotation.Evolving
+import org.apache.spark.sql.{Dataset, Row, SparkSession}
+import org.apache.spark.sql.types.{BooleanType, StructType}
+
+/**
+ * Contains constants used for convention based column naming.
+ */
+@Evolving
+object CypherSession {
+
+  /**
+   * Naming convention for identifier columns, both node and relationship 
identifiers.
+   */
+  val ID_COLUMN = "$ID"
+
+  /**
+   * Naming convention for relationship source identifier.
+   */
+  val SOURCE_ID_COLUMN = "$SOURCE_ID"
+
+  /**
+   * Naming convention for relationship target identifier.
+   */
+  val TARGET_ID_COLUMN = "$TARGET_ID"
+
+  /**
+   * Naming convention for node label prefixes.
+   */
+  val LABEL_COLUMN_PREFIX = ":"
+
+  /**
+   * Naming convention for relationship type prefixes.
+   */
+  val REL_TYPE_COLUMN_PREFIX = ":"
+
+  /**
+   * Extracts [[NodeDataset]]s from a [[Dataset]] using column name 
conventions.
+   *
+   * For information about naming conventions, see 
[[CypherSession.createGraph]].
+   *
+   * @param nodes node dataset
+   * @since 3.0.0
+   */
+  def extractNodeDatasets(nodes: Dataset[Row]): Array[NodeDataset] = {
+    val labelColumns = 
nodes.columns.filter(_.startsWith(LABEL_COLUMN_PREFIX)).toSet
+    validateLabelOrRelTypeColumns(nodes.schema, labelColumns, 
LABEL_COLUMN_PREFIX)
+
+    val nodeProperties = (nodes.columns.toSet - ID_COLUMN -- labelColumns)
+      .map(col => col -> col)
+      .toMap
+
+    val labelCount = labelColumns.size
+    if (labelCount > 5) {
+      LoggerFactory.getLogger(CypherSession.getClass).warn(
+        s"$labelCount label columns will result in ${Math.pow(labelCount, 2)} 
node datasets.")
+      if (labelCount > 10) {
+        throw new IllegalArgumentException(
+          s"Expected number of label columns to be less than or equal to 10, 
was $labelCount.")
+      }
+    }
+
+    val labelSets = labelColumns.subsets().toSet
+
+    labelSets.map { labelSet =>
+      val predicate = labelColumns
+        .map { labelColumn =>
+          if (labelSet.contains(labelColumn)) {
+            nodes.col(labelColumn)
+          } else {
+            !nodes.col(labelColumn)
+          }
+        }
+        .reduce(_ && _)
+
+      NodeDataset(nodes.filter(predicate), ID_COLUMN, 
labelSet.map(_.substring(1)), nodeProperties)
+    }.toArray
+  }
+
+  /**
+   * Extracts [[RelationshipDataset]]s from a [[Dataset]] using column name 
conventions.
+   *
+   * For information about naming conventions, see 
[[CypherSession.createGraph]].
+   *
+   * @param relationships relationship dataset
+   * @since 3.0.0
+   */
+  def extractRelationshipDatasets(relationships: Dataset[Row]): 
Array[RelationshipDataset] = {
+    val relColumns = relationships.columns.toSet
+    val relTypeColumns = 
relColumns.filter(_.startsWith(REL_TYPE_COLUMN_PREFIX))
+    validateLabelOrRelTypeColumns(relationships.schema, relTypeColumns, 
REL_TYPE_COLUMN_PREFIX)
+    val idColumns = Set(ID_COLUMN, SOURCE_ID_COLUMN, TARGET_ID_COLUMN)
+    val propertyColumns = relColumns -- idColumns -- relTypeColumns
+    val relProperties = propertyColumns.map(col => col -> col).toMap
+    relTypeColumns.map { relTypeColumn =>
+      val predicate = relationships.col(relTypeColumn)
+      // TODO: Make sure that each row represents a single relationship type
+      // see https://issues.apache.org/jira/browse/SPARK-29480
+      RelationshipDataset(
+        relationships.filter(predicate),
+        ID_COLUMN,
+        SOURCE_ID_COLUMN,
+        TARGET_ID_COLUMN,
+        relTypeColumn.substring(1),
+        relProperties)
+    }.toArray
+  }
+
+  /**
+   * Validates if the given columns fulfil specific constraints for
+   * representing node labels or relationship types.
+   *
+   * In particular, we check if the columns store boolean values and that
+   * the column name represents a single node label or relationship type.
+   *
+   * @param schema  Dataset schema
+   * @param columns columns to validate
+   * @param prefix  node label or relationship type prefix
+   */
+  private def validateLabelOrRelTypeColumns(
+      schema: StructType,
+      columns: Set[String],
+      prefix: String): Unit = {
+    schema.fields.filter(f => columns.contains(f.name)).foreach(field => {
+      if (field.dataType != BooleanType) {
+        throw new IllegalArgumentException(s"Column ${field.name} must be of 
type BooleanType.")
+      }
+    })
+    columns.foreach(typeColumn => {
+      if (typeColumn.sliding(prefix.length).count(_ == prefix) != 1) {
+        throw new IllegalArgumentException(s"Column $typeColumn must contain 
exactly one type.")
+      }
+    })
+  }
+
+}
+
+/**
+ * A CypherSession allows for creating, storing and loading [[PropertyGraph]] 
instances as well as
+ * executing Cypher queries on them.
+ *
+ * Wraps a [[org.apache.spark.sql.SparkSession]].
+ *
+ * @since 3.0.0
+ */
+@Evolving
 
 Review comment:
   Yea, I think the extensibility design is important, but I'm currently not 
100% sure that this trait leads to easy-to-extend interfaces in graph apis. IMO 
interfaces for extensions should be as limited as possible like 
`SparkSessionExtensions`. Do you have any design for that? 

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to