Github user kiszk commented on a diff in the pull request:
https://github.com/apache/spark/pull/19811#discussion_r157410164
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/regexpExpressions.scala
---
@@ -112,15 +112,14 @@ case class Like(left: Expression, right: Expression)
extends StringRegexExpressi
override protected def doGenCode(ctx: CodegenContext, ev: ExprCode):
ExprCode = {
val patternClass = classOf[Pattern].getName
val escapeFunc = StringUtils.getClass.getName.stripSuffix("$") +
".escapeLikeRegex"
- val pattern = ctx.freshName("pattern")
if (right.foldable) {
val rVal = right.eval()
if (rVal != null) {
val regexStr =
StringEscapeUtils.escapeJava(escape(rVal.asInstanceOf[UTF8String].toString()))
- ctx.addMutableState(patternClass, pattern,
- s"""$pattern = ${patternClass}.compile("$regexStr");""")
+ val pattern = ctx.addMutableState(patternClass, "patternLike",
+ v => s"""$v = ${patternClass}.compile("$regexStr");""",
forceInline = true)
--- End diff --
Now, we have three rules to apply inlining
1. Have to use the original name
2. Frequently used in the hot spot
3. [Not expected to be frequently
generated](https://github.com/apache/spark/pull/19811#discussion_r157110933)
proposed by @viirya
Now, we have no rule for 2. I will try to run microbenchmark for 2. Is it
better to add these benchmarks into the benchmark directory?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
