Merge pull request #461 from JohT/fix/execute-anomaly-graph-visualization-only-once

Fix anomaly Graph visualization executed twice and add DEPENDS_ON schema output

Author: JohT

cypher/Exploration/Explore_schema.cypher

@@ -0,0 +1,33 @@
+// Explore node labels and their relationships for a schema overview
+
+ MATCH (allNodes)
+  WITH COUNT(allNodes)              AS totalNodeCount
+MATCH (source)-[link]->(target)
+WITH  totalNodeCount 
+     ,labels(source)                AS sourceLabels
+     ,labels(target)                AS targetLabels
+     ,collect(DISTINCT type(link))  AS relationshipTypes
+     ,count(*)                      AS numberOfNodes
+UNWIND sourceLabels AS sourceLabel
+ WITH *
+ WHERE NOT sourceLabel STARTS WITH 'Mark4'
+ WITH totalNodeCount
+     ,collect(DISTINCT sourceLabel) AS sourceLabels
+     ,targetLabels
+     ,relationshipTypes
+     ,numberOfNodes
+UNWIND targetLabels AS targetLabel
+ WITH *
+ WHERE NOT targetLabel STARTS WITH 'Mark4'
+ WITH totalNodeCount
+     ,sourceLabels
+     ,collect(DISTINCT targetLabel) AS targetLabels
+     ,relationshipTypes
+     ,numberOfNodes
+RETURN apoc.text.join(sourceLabels, ',')      AS sourceLabels
+      ,apoc.text.join(relationshipTypes, ',') AS relationshipTypes
+      ,apoc.text.join(targetLabels, ',')      AS targetLabels
+      ,numberOfNodes
+      ,round(toFloat(numberOfNodes) / totalNodeCount * 100.0, 2) AS percentageOfTotalNodes
+ORDER BY numberOfNodes DESC
+LIMIT 200

cypher/Overview/Dependency_node_labels.cypher

@@ -0,0 +1,28 @@
+// Explore DEPENDS_ON relationship node labels 
+
+ MATCH (allNodes)
+  WITH COUNT(allNodes) AS totalNodeCount
+MATCH (source)-[:DEPENDS_ON]->(target)
+WITH  totalNodeCount 
+     ,labels(source)   AS sourceLabels
+     ,labels(target)   AS targetLabels
+     ,count(*)         AS numberOfNodes
+UNWIND sourceLabels AS sourceLabel
+ WITH *
+ WHERE NOT sourceLabel STARTS WITH 'Mark4'
+ WITH totalNodeCount
+     ,collect(DISTINCT sourceLabel) AS sourceLabels
+     ,targetLabels
+     ,numberOfNodes
+UNWIND targetLabels AS targetLabel
+ WITH *
+ WHERE NOT targetLabel STARTS WITH 'Mark4'
+ WITH totalNodeCount
+     ,sourceLabels
+     ,collect(DISTINCT targetLabel) AS targetLabels
+     ,numberOfNodes
+RETURN apoc.text.join(sourceLabels, ',') AS sourceLabels
+      ,apoc.text.join(targetLabels, ',') AS targetLabels
+      ,numberOfNodes
+      ,round(toFloat(numberOfNodes) / totalNodeCount * 100.0, 2) AS percentageOfTotalNodes
+ORDER BY numberOfNodes DESC, sourceLabels, targetLabels

domains/anomaly-detection/anomalyDetectionVisualization.sh

@@ -24,4 +24,4 @@ ANOMALY_DETECTION_SCRIPT_DIR=${ANOMALY_DETECTION_SCRIPT_DIR:-$(CDPATH=. cd -- "$
 ANOMALY_DETECTION_GRAPHS_DIR=${ANOMALY_DETECTION_GRAPHS_DIR:-"${ANOMALY_DETECTION_SCRIPT_DIR}/graphs"} # Contains everything (scripts, queries, templates) to create the Markdown summary report for anomaly detection
 
 # Delegate the execution to the responsible script.
-source "${ANOMALY_DETECTION_GRAPHS_DIR}/anomalyDetectionGraphVisualization.sh"
+source "${ANOMALY_DETECTION_GRAPHS_DIR}/anomalyDetectionGraphs.sh"

domains/anomaly-detection/graphs/anomalyDetectionGraphVisualization.sh renamed to domains/anomaly-detection/graphs/anomalyDetectionGraphs.sh

@@ -53,6 +53,7 @@ execute_cypher "${OVERVIEW_CYPHER_DIR}/Number_of_elements_per_module_for_Typescr
 execute_cypher "${OVERVIEW_CYPHER_DIR}/Node_label_count.cypher" > "${FULL_REPORT_DIRECTORY}/Node_label_count.csv"
 execute_cypher "${OVERVIEW_CYPHER_DIR}/Node_label_combination_count.cypher" > "${FULL_REPORT_DIRECTORY}/Node_label_combination_count.csv"
 execute_cypher "${OVERVIEW_CYPHER_DIR}/Relationship_type_count.cypher" > "${FULL_REPORT_DIRECTORY}/Relationship_type_count.csv"
+execute_cypher "${OVERVIEW_CYPHER_DIR}/Dependency_node_labels.cypher" > "${FULL_REPORT_DIRECTORY}/Dependency_node_labels.csv"
 
 # TODO Performance needs improvement. Included (limited) in OverviewGeneral Jupyter Notebook.
 # execute_cypher "${OVERVIEW_CYPHER_DIR}/Node_labels_and_their_relationships.cypher" > "${FULL_REPORT_DIRECTORY}/Node_labels_and_their_relationships.csv"