TRUNK-6469: Add support for automatic initialization of Envers audit tables #5468

wikumChamith · 2025-11-09T12:27:21Z

Description of what I changed

This Implements a production-ready alternative to hbm2ddl.auto=update for automatically generating Hibernate Envers audit tables when auditing is enabled.

To test add hibernate.integration.envers.enabled=true to openmrs-runtime.properties file.

This need to also get backported to 2.7.x and 2.8.x. However because they are using Hibernate 5 we could use org.hibernate.envers.tools.hbm2ddl.EnversSchemaGenerator.

Issue I worked on

see https://openmrs.atlassian.net/browse/TRUNK-6469

Checklist: I completed these to help reviewers :)

My IDE is configured to follow the code style of this project.

No? Unsure? -> configure your IDE, format the code and add the changes with git add . && git commit --amend
I have added tests to cover my changes. (If you refactored
existing code that was well tested you do not have to add tests)

No? -> write tests and add them to this commit git add . && git commit --amend
I ran mvn clean package right before creating this pull request and
added all formatting changes to my commit.

No? -> execute above command
All new and existing tests passed.

No? -> figure out why and add the fix to your commit. It is your responsibility to make sure your code works.
My pull request is based on the latest changes of the master branch.

No? Unsure? -> execute command git pull --rebase upstream master

dkayiwa · 2025-11-09T23:09:18Z

api/src/main/java/org/openmrs/util/EnversAuditTableInitializer.java

+	 * @param hibernateProperties properties containing Envers configuration
+	 * @param serviceRegistry Hibernate service registry
+	 */
+	public static void initializeAuditTables(Metadata metadata, Properties hibernateProperties,


What do you think of removing parts of the method name that are already in the class name?

I changed the method name.

ibacher · 2025-11-10T20:56:40Z

api/src/main/java/org/openmrs/util/EnversAuditTableInitializer.java

+	private static void updateAuditTables(Metadata metadata, Properties hibernateProperties,
+	        ServiceRegistry serviceRegistry) throws Exception {
+		String auditTablePrefix = hibernateProperties.getProperty("org.hibernate.envers.audit_table_prefix", "");
+		String auditTableSuffix = hibernateProperties.getProperty("org.hibernate.envers.audit_table_suffix", "_aud");


The main thing I might ask here is that we name the audit tables _audit. It's more consistent with how we name other DB tables and this isn't MUMPS where tables names are (IIRC) something like 10 characters.

ibacher · 2025-11-10T20:58:28Z

api/src/main/java/org/openmrs/util/EnversAuditTableInitializer.java

+			updateAuditTables(metadata, hibernateProperties, serviceRegistry);
+			log.info("Envers audit table initialization completed successfully.");
+		}
+		catch (Exception e) {


Shouldn't this exception potentially stop the server if someone opted into using Envers but the tables didn't generate?

I tested this by maually throwing an exception from updateAuditTables and it did not stoped the server

Had you done something with the other try / catch here? I can't see where that exception would be swallowed.

I think we don't need try-catch here, so I removed it :)

wikumChamith · 2025-11-11T08:06:02Z

@ibacher I updated the PR.

nsalifu

I've tested this and works as expected. thank you @wikumChamith

wikumChamith · 2025-11-25T06:28:43Z

@dkayiwa, @ibacher is this ready to get merged?

ibacher · 2025-11-25T20:20:52Z

api/src/main/java/org/openmrs/api/db/hibernate/HibernateSessionFactoryBean.java

+	private void generateEnversAuditTables(Metadata metadata, SessionFactoryServiceRegistry serviceRegistry) {
+		try {
+			Properties hibernateProperties = getHibernateProperties();
+			EnversAuditTableInitializer.initialize(metadata, hibernateProperties, serviceRegistry);
+		} catch (Exception e) {
+			log.error("Failed to initialize Envers audit tables", e);
+		}
+	}


Why are we swallowing the exception here? If someone requested that we use envers, then we should fail the start-up if the tables can't be created, right?

I thought the application should still start even if the audit tables failed to initialize, but this approach also makes sense

ibacher · 2025-11-25T20:22:50Z

api/src/main/java/org/openmrs/util/EnversAuditTableInitializer.java

+		Map<String, Object> settings = new HashMap<>();
+		for (String key : hibernateProperties.stringPropertyNames()) {
+			settings.put(key, hibernateProperties.getProperty(key));
+		}


This seems extraneous when we could just cast the hibernateProperties to Map<String, Object>. Types are erased at runtime so it's kind of a distinction without a difference.

I changed this. However, is it safe to directly cast Properties to Map<String, Object>?

I mean technically no, but for all practical purposes, it will work.

…tables

wikumChamith force-pushed the TRUNK-6469 branch 4 times, most recently from 4e8f41a to 960a0af Compare November 9, 2025 13:55

wikumChamith requested review from ManojLL, dkayiwa and ibacher November 9, 2025 13:56

dkayiwa reviewed Nov 9, 2025

View reviewed changes

wikumChamith force-pushed the TRUNK-6469 branch from 960a0af to a4974af Compare November 10, 2025 04:47

wikumChamith requested a review from dkayiwa November 10, 2025 04:49

ibacher reviewed Nov 10, 2025

View reviewed changes

wikumChamith force-pushed the TRUNK-6469 branch 2 times, most recently from 07955a5 to 76ef78d Compare November 11, 2025 08:05

wikumChamith requested a review from ibacher November 11, 2025 08:06

wikumChamith force-pushed the TRUNK-6469 branch 4 times, most recently from 86f745f to aa101ef Compare November 12, 2025 16:59

nsalifu approved these changes Nov 13, 2025

View reviewed changes

ibacher reviewed Nov 25, 2025

View reviewed changes

wikumChamith force-pushed the TRUNK-6469 branch from aec293e to bd334fd Compare November 26, 2025 10:59

wikumChamith requested a review from ibacher November 26, 2025 11:02

wikumChamith force-pushed the TRUNK-6469 branch from bd334fd to dee3669 Compare November 26, 2025 11:04

TRUNK-6469: Add support for automatic initialization of Envers audit …

2b27b3c

…tables

wikumChamith force-pushed the TRUNK-6469 branch from dee3669 to 2b27b3c Compare November 26, 2025 11:06

Merge branch 'master' into TRUNK-6469

2b7797f

TRUNK-6469: Add support for automatic initialization of Envers audit tables #5468

Are you sure you want to change the base?

TRUNK-6469: Add support for automatic initialization of Envers audit tables #5468

Conversation

wikumChamith commented Nov 9, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description of what I changed

Issue I worked on

Checklist: I completed these to help reviewers :)

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

wikumChamith commented Nov 11, 2025

Uh oh!

nsalifu left a comment

Choose a reason for hiding this comment

Uh oh!

wikumChamith commented Nov 25, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

wikumChamith commented Nov 9, 2025 •

edited

Loading