dsmiley commented on code in PR #1302:
URL: https://github.com/apache/solr/pull/1302#discussion_r1100587425
##########
solr/core/src/java/org/apache/solr/schema/ManagedIndexSchemaFactory.java:
##########
@@ -197,89 +200,99 @@ public ManagedIndexSchema create(
this.config = config;
this.loader = config.getResourceLoader();
InputStream schemaInputStream = null;
+ String loadedResource = null;
- if (null == resourceName) {
- resourceName = IndexSchema.DEFAULT_SCHEMA_FILE;
- }
-
- int schemaZkVersion = -1;
- if (!(loader instanceof ZkSolrResourceLoader)) {
- schemaInputStream = readSchemaLocally();
- } else { // ZooKeeper
- final ZkSolrResourceLoader zkLoader = (ZkSolrResourceLoader) loader;
- final SolrZkClient zkClient = zkLoader.getZkController().getZkClient();
- final String managedSchemaPath = lookupZKManagedSchemaPath();
- managedSchemaResourceName =
- managedSchemaPath.substring(managedSchemaPath.lastIndexOf("/") + 1);
// not loving this
- Stat stat = new Stat();
- try {
- // Attempt to load the managed schema
- byte[] data = zkClient.getData(managedSchemaPath, null, stat, true);
- schemaZkVersion = stat.getVersion();
- schemaInputStream =
- new ZkSolrResourceLoader.ZkByteArrayInputStream(data,
managedSchemaPath, stat);
- loadedResource = managedSchemaResourceName;
- warnIfNonManagedSchemaExists();
- } catch (InterruptedException e) {
- // Restore the interrupted status
- Thread.currentThread().interrupt();
- log.warn("", e);
- } catch (KeeperException.NoNodeException e) {
- log.info(
- "The schema is configured as managed, but managed schema resource
{} not found - loading non-managed schema {} instead",
- managedSchemaResourceName,
- resourceName);
- } catch (KeeperException e) {
- String msg = "Error attempting to access " + managedSchemaPath;
- log.error(msg, e);
- throw new SolrException(ErrorCode.SERVER_ERROR, msg, e);
+ try {
+ if (null == resourceName) {
+ resourceName = IndexSchema.DEFAULT_SCHEMA_FILE;
}
- if (null == schemaInputStream) {
- // The managed schema file could not be found - load the non-managed
schema
+
+ int schemaZkVersion = -1;
+ if (!(loader instanceof ZkSolrResourceLoader)) {
+ Entry<String, InputStream> localSchemaInput = readSchemaLocally();
Review Comment:
It sucks to still see this because we have a NotThreadSafe annotation, thus
I wasn't anticipating the checks. From reading the [Infer/RacerD docs in more
detail](https://fbinfer.com/docs/1.0.0/all-issue-types/#thread-safety-conditions-checked-before-reporting),
it appears this is because this class, despite **not** being ThreadSafe, and
advertising itself as such nonetheless contains the "synchronized" keyword in
one spot at the end of create(). The presence of the keyword in the code takes
precedence over NotThreadSafe annotation. It seems to me we don't need that
synchronized there because the schema is not published yet. Any way, such
changes / explorations don't belong here. I'll create a PR to toy with this.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
