Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Exclude all unicode line terminator characters from matching dot #5424

Merged
merged 2 commits into from
May 6, 2022
Merged

Exclude all unicode line terminator characters from matching dot #5424

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0ea8380
Exclude all unicode line terminator characters from matching dot
anthony-chang May 4, 2022
c28bff3
Use existing terminatorChars seq, add unit tests
anthony-chang May 5, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion sql-plugin/src/main/scala/com/nvidia/spark/rapids/RegexParser.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -565,7 +565,9 @@ class CudfRegexTranspiler(mode: RegexMode) {
case RegexChar(ch) => ch match {
case '.' =>
// workaround for https://github.com/rapidsai/cudf/issues/9619
RegexCharacterClass(negated = true, ListBuffer(RegexChar('\r'), RegexChar('\n')))
val terminatorChars = new ListBuffer[RegexCharacterClassComponent]()
terminatorChars ++= lineTerminatorChars.map(RegexChar)
RegexCharacterClass(negated = true, terminatorChars)
case '$' if mode == RegexSplitMode || mode == RegexReplaceMode =>
// see https://github.com/NVIDIA/spark-rapids/issues/4533
throw new RegexUnsupportedException("line anchor $ is not supported in split or replace")
Expand Down
21 changes: 14 additions & 7 deletions tests/src/test/scala/com/nvidia/spark/rapids/RegularExpressionTranspilerSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -277,10 +277,17 @@ class RegularExpressionTranspilerSuite extends FunSuite with Arm {
Seq("", "$", "$9.99"))
}

test("dot matches CR on GPU but not on CPU") {
// see https://github.com/rapidsai/cudf/issues/9619
val pattern = "1."
assertCpuGpuMatchesRegexpFind(Seq(pattern), Seq("1\r2", "1\n2", "1\r\n2"))
test("dot does not match all line terminators") {
// see https://github.com/NVIDIA/spark-rapids/issues/5415
val pattern = Seq("1.")
val inputs = Seq("123", "1\r2", "1\n2", "1\r\n2", "1\u00852", "1\u20282", "1\u20292")
assertCpuGpuMatchesRegexpFind(pattern, inputs)
}

test("dot does not match line terminator combinations") {
val pattern = Seq("a.")
val inputs = Seq("abc", "a\n\rb", "a\n\u0085b", "a\u2029\u0085b", "a\u2082\rb")
assertCpuGpuMatchesRegexpFind(pattern, inputs)
}

test("character class with ranges") {
Expand All @@ -306,11 +313,11 @@ class RegularExpressionTranspilerSuite extends FunSuite with Arm {
"[0-9]{2}:[0-9]{2}:[0-9]{2})" +
"(.[1-9]*(?:0)?[1-9]+)?(.0*[1-9]+)?(?:.0*)?\\z"

// input and output should be identical except for `.` being replaced with `[^\r\n]` and
// `\z` being replaced with `$`
// input and output should be identical except for `.` being replaced
// with `[^\n\r\u0085\u2028\u2029]` and `\z` being replaced with `$`
doTranspileTest(TIMESTAMP_TRUNCATE_REGEX,
TIMESTAMP_TRUNCATE_REGEX
.replaceAll("\\.", "[^\r\n]")
.replaceAll("\\.", "[^\n\r\u0085\u2028\u2029]")
.replaceAll("\\\\z", "\\$"))
}

Expand Down