Wan Kun created SPARK-35379:
-------------------------------
Summary: Improve InferFiltersFromConstraints rule performance when
parsing spark sql
Key: SPARK-35379
URL: https://issues.apache.org/jira/browse/SPARK-35379
Project: Spark
Issue Type: Improvement
Components: SQL
Affects Versions: 3.1.1
Reporter: Wan Kun
*InferFiltersFromConstraints* rule will generates too many constraints when
there are many aliases in Project, . For example:
{code:java}
test("Expression explosion when analyze test") {
RuleExecutor.resetMetrics()
Seq((1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14))
.toDF("a", "b", "c", "d", "e", "f", "g", "h", "i", "j",
"k", "l", "m", "n")
.write.saveAsTable("test")
val df = spark.table("test")
val df2 = df.filter("a+b+c+d+e+f+g+h+i+j+k+l+m+n > 100")
val df3 = df2.select('a as 'a1, 'b as 'b1,
'c as 'c1, 'd as 'd1, 'e as 'e1, 'f as 'f1,
'g as 'g1, 'h as 'h1, 'i as 'i1, 'j as 'j1,
'k as 'k1, 'l as 'l1, 'm as 'm1, 'n as 'n1)
val df4 = df3.join(df2, df3("a1") === df2("a"))
df4.explain(true)
logWarning(RuleExecutor.dumpTimeSpent())
}
{code}
So we need to optimize *InferFiltersFromConstraints* rule to improve
performance.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
