Skip to content

[SPARK-48403][SQL] Fix Lower & Upper expressions for UTF8_BINARY_LCASE & ICU collations #46720

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

Closed
uros-db wants to merge 10 commits into from
Closed

[SPARK-48403][SQL] Fix Lower & Upper expressions for UTF8_BINARY_LCASE & ICU collations #46720

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
949eb7c
Initial commit
uros-db May 23, 2024
182c2c5
Tests
uros-db May 23, 2024
d9a0b11
Update doc comments
uros-db May 24, 2024
df12951
Remove InitCap
uros-db May 24, 2024
3e70e6d
Undo unnecessary changes
uros-db May 24, 2024
f5a3939
Correct naming
uros-db May 24, 2024
aa74fff
Merge branch 'apache:master' into lower-upper-initcap
uros-db May 28, 2024
a2d16c0
Small fixes
uros-db Jun 5, 2024
3646785
Add tests
uros-db Jun 7, 2024
799a2a0
Merge branch 'apache:master' into lower-upper-initcap
uros-db Jun 7, 2024
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
70 changes: 69 additions & 1 deletion common/unsafe/src/main/java/org/apache/spark/sql/catalyst/util/CollationAwareUTF8String.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -135,22 +135,90 @@ public static UTF8String lowercaseReplace(final UTF8String src, final UTF8String
return buf.build();
}

/**
* Convert the input string to uppercase using the ICU root locale rules.
*
* @param target the input string
* @return the uppercase string
*/
public static UTF8String toUpperCase(final UTF8String target) {
return UTF8String.fromString(toUpperCase(target.toString()));
}
public static String toUpperCase(final String target) {
return UCharacter.toUpperCase(target);
}

/**
* Convert the input string to uppercase using the specified ICU collation rules.
*
* @param target the input string
* @return the uppercase string
*/
public static UTF8String toUpperCase(final UTF8String target, final int collationId) {
return UTF8String.fromString(toUpperCase(target.toString(), collationId));
}
public static String toUpperCase(final String target, final int collationId) {
ULocale locale = CollationFactory.fetchCollation(collationId)
.collator.getLocale(ULocale.ACTUAL_LOCALE);
return UCharacter.toUpperCase(locale, target);
}

/**
* Convert the input string to lowercase using the ICU root locale rules.
*
* @param target the input string
* @return the lowercase string
*/
public static UTF8String toLowerCase(final UTF8String target) {
return UTF8String.fromString(toLowerCase(target.toString()));
}
public static String toLowerCase(final String target) {
return UCharacter.toLowerCase(target);
}

/**
* Convert the input string to lowercase using the specified ICU collation rules.
*
* @param target the input string
* @return the lowercase string
*/
public static UTF8String toLowerCase(final UTF8String target, final int collationId) {
return UTF8String.fromString(toLowerCase(target.toString(), collationId));
}
public static String toLowerCase(final String target, final int collationId) {
ULocale locale = CollationFactory.fetchCollation(collationId)
.collator.getLocale(ULocale.ACTUAL_LOCALE);
return UCharacter.toLowerCase(locale, target);
}

/**
* Convert the input string to lowercase using the ICU root locale rules.
*
* @param target the input string
* @return the lowercase string
*/
public static UTF8String toTitleCase(final UTF8String target) {
return UTF8String.fromString(toTitleCase(target.toString()));
}
public static String toTitleCase(final String target) {
BreakIterator wordIterator = BreakIterator.getWordInstance();
return UCharacter.toTitleCase(target, wordIterator);
}

/**
* Convert the input string to lowercase using the specified ICU collation rules.
*
* @param target the input string
* @return the lowercase string
*/
public static UTF8String toTitleCase(final UTF8String target, final int collationId) {
return UTF8String.fromString(toTitleCase(target.toString(), collationId));
}
public static String toTitleCase(final String target, final int collationId) {
ULocale locale = CollationFactory.fetchCollation(collationId)
.collator.getLocale(ULocale.ACTUAL_LOCALE);
return UCharacter.toTitleCase(locale, target, BreakIterator.getWordInstance(locale));
BreakIterator wordIterator = BreakIterator.getWordInstance(locale);
return UCharacter.toTitleCase(locale, target, wordIterator);
}

public static int findInSet(final UTF8String match, final UTF8String set, int collationId) {
Expand Down
54 changes: 33 additions & 21 deletions common/unsafe/src/main/java/org/apache/spark/sql/catalyst/util/CollationSupport.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -208,87 +208,99 @@ public static boolean execICU(final UTF8String l, final UTF8String r,
public static class Upper {
public static UTF8String exec(final UTF8String v, final int collationId) {
CollationFactory.Collation collation = CollationFactory.fetchCollation(collationId);
if (collation.supportsBinaryEquality || collation.supportsLowercaseEquality) {
if (collation.supportsBinaryEquality) {
return execUTF8(v);
} else {
} else if (collation.supportsLowercaseEquality) {
return execLowercase(v);
} else {
return execICU(v, collationId);
}
}
public static String genCode(final String v, final int collationId) {
CollationFactory.Collation collation = CollationFactory.fetchCollation(collationId);
String expr = "CollationSupport.Upper.exec";
if (collation.supportsBinaryEquality || collation.supportsLowercaseEquality) {
if (collation.supportsBinaryEquality) {
return String.format(expr + "UTF8(%s)", v);
} else {
} else if (collation.supportsLowercaseEquality) {
return String.format(expr + "Lowercase(%s)", v);
} else {
return String.format(expr + "ICU(%s, %d)", v, collationId);
}
}
public static UTF8String execUTF8(final UTF8String v) {
return v.toUpperCase();
}
public static UTF8String execLowercase(final UTF8String v) {
return CollationAwareUTF8String.toUpperCase(v);
}
public static UTF8String execICU(final UTF8String v, final int collationId) {
return UTF8String.fromString(CollationAwareUTF8String.toUpperCase(v.toString(), collationId));
return CollationAwareUTF8String.toUpperCase(v, collationId);
}
}

public static class Lower {
public static UTF8String exec(final UTF8String v, final int collationId) {
CollationFactory.Collation collation = CollationFactory.fetchCollation(collationId);
if (collation.supportsBinaryEquality || collation.supportsLowercaseEquality) {
if (collation.supportsBinaryEquality) {
return execUTF8(v);
} else if (collation.supportsLowercaseEquality) {
return execLowercase(v);
} else {
return execICU(v, collationId);
}
}
public static String genCode(final String v, final int collationId) {
CollationFactory.Collation collation = CollationFactory.fetchCollation(collationId);
String expr = "CollationSupport.Lower.exec";
if (collation.supportsBinaryEquality || collation.supportsLowercaseEquality) {
if (collation.supportsBinaryEquality) {
return String.format(expr + "UTF8(%s)", v);
} else {
} else if (collation.supportsLowercaseEquality) {
return String.format(expr + "Lowercase(%s)", v);
} else {
return String.format(expr + "ICU(%s, %d)", v, collationId);
}
}
public static UTF8String execUTF8(final UTF8String v) {
return v.toLowerCase();
}
public static UTF8String execLowercase(final UTF8String v) {
return CollationAwareUTF8String.toLowerCase(v);
}
public static UTF8String execICU(final UTF8String v, final int collationId) {
return UTF8String.fromString(CollationAwareUTF8String.toLowerCase(v.toString(), collationId));
return CollationAwareUTF8String.toLowerCase(v, collationId);
}
}

public static class InitCap {
public static UTF8String exec(final UTF8String v, final int collationId) {
CollationFactory.Collation collation = CollationFactory.fetchCollation(collationId);
if (collation.supportsBinaryEquality || collation.supportsLowercaseEquality) {
if (collation.supportsBinaryEquality) {
return execUTF8(v);
} else if (collation.supportsLowercaseEquality) {
return execLowercase(v);
} else {
return execICU(v, collationId);
}
}

public static String genCode(final String v, final int collationId) {
CollationFactory.Collation collation = CollationFactory.fetchCollation(collationId);
String expr = "CollationSupport.InitCap.exec";
if (collation.supportsBinaryEquality || collation.supportsLowercaseEquality) {
if (collation.supportsBinaryEquality) {
return String.format(expr + "UTF8(%s)", v);
} else if (collation.supportsLowercaseEquality) {
return String.format(expr + "Lowercase(%s)", v);
} else {
return String.format(expr + "ICU(%s, %d)", v, collationId);
}
}

public static UTF8String execUTF8(final UTF8String v) {
return v.toLowerCase().toTitleCase();
}

public static UTF8String execLowercase(final UTF8String v) {
return CollationAwareUTF8String.toTitleCase(v);
}
public static UTF8String execICU(final UTF8String v, final int collationId) {
return UTF8String.fromString(
CollationAwareUTF8String.toTitleCase(
CollationAwareUTF8String.toLowerCase(
v.toString(),
collationId
),
collationId));
return CollationAwareUTF8String.toTitleCase(v, collationId);
}
}

Expand Down
11 changes: 6 additions & 5 deletions common/unsafe/src/test/java/org/apache/spark/unsafe/types/CollationSupportSuite.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,7 @@

import static org.junit.jupiter.api.Assertions.*;


// checkstyle.off: AvoidEscapedUnicodeCharacters
public class CollationSupportSuite {

/**
Expand Down Expand Up @@ -526,10 +526,10 @@ public void testInitCap() throws SparkException {
assertInitCap("aB 世 de", "UNICODE_CI", "Ab 世 De");
assertInitCap("ÄBĆΔE", "UNICODE_CI", "Äbćδe");
// Case-variable character length
assertInitCap("İo", "UTF8_BINARY", "İo");
assertInitCap("İo", "UTF8_BINARY_LCASE", "İo");
assertInitCap("İo", "UNICODE", "İo");
assertInitCap("İo", "UNICODE_CI", "İo");
assertInitCap("İo", "UTF8_BINARY", "I\u0307o");
assertInitCap("İo", "UTF8_BINARY_LCASE", "İo");
assertInitCap("İo", "UNICODE", "I\u0307o");
assertInitCap("İo", "UNICODE_CI", "İo");
}

private void assertStringInstr(String string, String substring, String collationName,
Expand Down Expand Up @@ -1008,3 +1008,4 @@ public void testStringTrim() throws SparkException {
// TODO: Test other collation-aware expressions.

}
// checkstyle.on: AvoidEscapedUnicodeCharacters