Merge pull request #11898 from IQSS/11828-unacceptable-performance-deleting-datasets

adding indexes and other sql for performance improvements with deleting Datasets and DatasetVersions

Author: pdurbin

doc/release-notes/11828-unacceptable-performance-deleting-datasets.md

@@ -0,0 +1 @@
+This release adds database indexes on GuestbookResponse and DatasetMetrtics, speeding up Dataset deletes. It also adds a constraint preventing null VersionState, as a matter of good housekeeping practice. 

src/main/java/edu/harvard/iq/dataverse/DatasetVersion.java

@@ -159,7 +159,7 @@ public enum VersionState {
      * @todo versionState should never be null so when we are ready, uncomment
      * the `nullable = false` below.
      */
-//    @Column(nullable = false)
+    @Column(nullable = false)
     @Enumerated(EnumType.STRING)
     private VersionState versionState;
 

src/main/java/edu/harvard/iq/dataverse/FileAccessRequest.java

@@ -55,7 +55,7 @@ public class FileAccessRequest implements Serializable{
     private Long id;
 
     @ManyToOne
-    @JoinColumn(nullable=false)
+    @JoinColumn(name="datafile_id", nullable=false)
     private DataFile dataFile;
 
     @ManyToOne

src/main/java/edu/harvard/iq/dataverse/GuestbookResponse.java

@@ -37,6 +37,8 @@
 @Table(indexes = {
         @Index(columnList = "guestbook_id"),
         @Index(columnList = "datafile_id"),
+        @Index(columnList = "datasetversion_id"),
+        @Index(columnList = "authenticateduser_id"),
         @Index(columnList = "dataset_id")
 })
 

src/main/java/edu/harvard/iq/dataverse/makedatacount/DatasetMetrics.java

@@ -8,8 +8,10 @@
 import jakarta.persistence.GeneratedValue;
 import jakarta.persistence.GenerationType;
 import jakarta.persistence.Id;
+import jakarta.persistence.Index;
 import jakarta.persistence.JoinColumn;
 import jakarta.persistence.ManyToOne;
+import jakarta.persistence.Table;
 import jakarta.persistence.Transient;
 import jakarta.validation.constraints.NotNull;
 
@@ -27,6 +29,9 @@
  * downloads for a given month.
  */
 @Entity
+@Table(indexes = {
+    @Index(columnList = "dataset_id")
+})
 public class DatasetMetrics implements Serializable {
 
     @Id

src/main/resources/db/migration/V6.8.0.3.sql

@@ -0,0 +1,8 @@
+CREATE INDEX IF NOT EXISTS  INDEX_GUESTBOOKRESPONSE_datasetversion_id ON GUESTBOOKRESPONSE (datasetversion_id);
+
+CREATE INDEX IF NOT EXISTS INDEX_DATASETMETRICS_dataset_id ON DATASETMETRICS (dataset_id);
+
+CREATE INDEX IF NOT EXISTS INDEX_GUESTBOOKRESPONSE_authenticateduser_id ON GUESTBOOKRESPONSE (authenticateduser_id);
+
+UPDATE datasetversion SET versionstate = 'DRAFT' where versionstate IS NULL;
+ALTER TABLE datasetversion ALTER COLUMN versionstate SET NOT NULL;