Github user marmbrus commented on a diff in the pull request:
https://github.com/apache/spark/pull/4015#discussion_r28013127
--- Diff: sql/core/src/main/scala/org/apache/spark/sql/SQLContext.scala ---
@@ -129,19 +191,33 @@ class SQLContext(@transient val sparkContext:
SparkContext)
@transient
protected[sql] lazy val optimizer: Optimizer = DefaultOptimizer
- @transient
- protected[sql] val ddlParser = new DDLParser(sqlParser.apply(_))
-
- @transient
- protected[sql] val sqlParser = {
- val fallback = new catalyst.SqlParser
- new SparkSQLParser(fallback(_))
+ protected[sql] def dialectCache = tlSession.get().dialectCache
+ protected[sql] def setDialectCache(d: SQLDialect) = {
+ tlSession.get().dialectCache = d
}
- protected[sql] def parseSql(sql: String): LogicalPlan = {
- ddlParser(sql, false).getOrElse(sqlParser(sql))
+ protected[sql] def getSQLDialect(): SQLDialect = {
+ try {
+ val clazz = Utils.classForName(dialectClassName)
+ if (dialectCache == null || dialectCache.getClass != clazz) {
+
setDialectCache(clazz.newInstance().asInstanceOf[SQLDialect].initialize(this))
+ }
+
+ dialectCache
--- End diff --
This caching logic is pretty complicated and hard to follow. Its not even
caching the reflective part which is likely the most expensive part. I would
remote it and just construct dialects on demand unless you have a benchmark to
show that is expensive.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
