Skip to content

[SPARK-48281][SQL] Alter string search logic for UTF8_BINARY_LCASE collation (StringInStr, SubstringIndex) #46589

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
wants to merge 7 commits into from
Closed

[SPARK-48281][SQL] Alter string search logic for UTF8_BINARY_LCASE collation (StringInStr, SubstringIndex) #46589

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
f226c65
Fix and tests
uros-db May 15, 2024
a2e1307
Fix tests
uros-db May 15, 2024
4671b85
Merge branch 'apache:master' into alter-lcase-vol2
uros-db May 15, 2024
168dde3
Small fixes
uros-db May 17, 2024
a26124c
Centralize indexOf behaviour for empty substring
uros-db May 20, 2024
b571084
Update new method access
uros-db May 24, 2024
05d2ddb
Merge branch 'master' into alter-lcase-vol2
uros-db May 28, 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
48 changes: 14 additions & 34 deletions 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 @@ -345,14 +345,14 @@ public static int findInSet(final UTF8String match, final UTF8String set, int co
*/
public static int lowercaseIndexOf(final UTF8String target, final UTF8String pattern,
final int start) {
if (pattern.numChars() == 0) return 0;
if (pattern.numChars() == 0) return target.indexOfEmpty(start);
return lowercaseFind(target, pattern.toLowerCase(), start);
}

public static int indexOf(final UTF8String target, final UTF8String pattern,
final int start, final int collationId) {
if (pattern.numBytes() == 0) {
return 0;
return target.indexOfEmpty(start);
}

StringSearch stringSearch = CollationFactory.getStringSearch(target, pattern, collationId);
Expand Down Expand Up @@ -444,47 +444,27 @@ public static UTF8String lowercaseSubStringIndex(final UTF8String string,
return UTF8String.EMPTY_UTF8;
}

UTF8String lowercaseString = string.toLowerCase();
UTF8String lowercaseDelimiter = delimiter.toLowerCase();

if (count > 0) {
int idx = -1;
// Search left to right (note: the start code point is inclusive).
int matchLength = -1;
while (count > 0) {
idx = lowercaseString.find(lowercaseDelimiter, idx + 1);
if (idx >= 0) {
count--;
} else {
// can not find enough delim
return string;
}
}
if (idx == 0) {
return UTF8String.EMPTY_UTF8;
matchLength = lowercaseFind(string, lowercaseDelimiter, matchLength + 1);
if (matchLength > MATCH_NOT_FOUND) --count; // Found a delimiter.
else return string; // Cannot find enough delimiters in the string.
}
byte[] bytes = new byte[idx];
copyMemory(string.getBaseObject(), string.getBaseOffset(), bytes, BYTE_ARRAY_OFFSET, idx);
return UTF8String.fromBytes(bytes);

return string.substring(0, matchLength);
} else {
int idx = string.numBytes() - delimiter.numBytes() + 1;
// Search right to left (note: the end code point is exclusive).
int matchLength = string.numChars() + 1;
count = -count;
while (count > 0) {
idx = lowercaseString.rfind(lowercaseDelimiter, idx - 1);
if (idx >= 0) {
count--;
} else {
// can not find enough delim
return string;
}
matchLength = lowercaseRFind(string, lowercaseDelimiter, matchLength - 1);
if (matchLength > MATCH_NOT_FOUND) --count; // Found a delimiter.
else return string; // Cannot find enough delimiters in the string.
}
if (idx + delimiter.numBytes() == string.numBytes()) {
return UTF8String.EMPTY_UTF8;
}
int size = string.numBytes() - delimiter.numBytes() - idx;
byte[] bytes = new byte[size];
copyMemory(string.getBaseObject(), string.getBaseOffset() + idx + delimiter.numBytes(),
bytes, BYTE_ARRAY_OFFSET, size);
return UTF8String.fromBytes(bytes);
return string.substring(matchLength, string.numChars());
}
}

Expand Down
2 changes: 1 addition & 1 deletion 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 @@ -354,7 +354,7 @@ public static int execBinary(final UTF8String string, final UTF8String substring
return string.indexOf(substring, 0);
}
public static int execLowercase(final UTF8String string, final UTF8String substring) {
return string.toLowerCase().indexOf(substring.toLowerCase(), 0);
return CollationAwareUTF8String.lowercaseIndexOf(string, substring, 0);
}
public static int execICU(final UTF8String string, final UTF8String substring,
final int collationId) {
Expand Down
13 changes: 12 additions & 1 deletion common/unsafe/src/main/java/org/apache/spark/unsafe/types/UTF8String.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -773,6 +773,17 @@ public UTF8String repeat(int times) {
return UTF8String.fromBytes(newBytes);
}

/**
* Returns the (default) position of the first occurrence of an empty substr in the current
* string from the specified position (0-based index).
*
* @param start the start position of the current string for searching
* @return the position of the first occurrence of the empty substr (now, always 0)
*/
public int indexOfEmpty(int start) {
return 0; // TODO: Fix this behaviour (SPARK-48284)
}

/**
* Returns the position of the first occurrence of substr in
* current string from the specified position (0-based index).
Expand All @@ -783,7 +794,7 @@ public UTF8String repeat(int times) {
*/
public int indexOf(UTF8String v, int start) {
if (v.numBytes() == 0) {
return 0;
return indexOfEmpty(start);
}

// locate to the start position.
Expand Down
50 changes: 48 additions & 2 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 @@ -635,8 +635,28 @@ public void testStringInstr() throws SparkException {
assertStringInstr("aaads", "dS", "UNICODE_CI", 4);
assertStringInstr("test大千世界X大千世界", "界y", "UNICODE_CI", 0);
assertStringInstr("test大千世界X大千世界", "界x", "UNICODE_CI", 8);
assertStringInstr("abİo12", "i̇o", "UNICODE_CI", 3);
assertStringInstr("abi̇o12", "İo", "UNICODE_CI", 3);
assertStringInstr("i̇", "i", "UNICODE_CI", 0);
assertStringInstr("i̇", "\u0307", "UNICODE_CI", 0);
assertStringInstr("i̇", "İ", "UNICODE_CI", 1);
assertStringInstr("İ", "i", "UNICODE_CI", 0);
assertStringInstr("İoi̇o12", "i̇o", "UNICODE_CI", 1);
assertStringInstr("i̇oİo12", "İo", "UNICODE_CI", 1);
assertStringInstr("abİoi̇o", "i̇o", "UNICODE_CI", 3);
assertStringInstr("abi̇oİo", "İo", "UNICODE_CI", 3);
assertStringInstr("ai̇oxXİo", "Xx", "UNICODE_CI", 5);
assertStringInstr("aİoi̇oxx", "XX", "UNICODE_CI", 7);
assertStringInstr("i̇", "i", "UTF8_BINARY_LCASE", 1); // != UNICODE_CI
assertStringInstr("i̇", "\u0307", "UTF8_BINARY_LCASE", 2); // != UNICODE_CI
assertStringInstr("i̇", "İ", "UTF8_BINARY_LCASE", 1);
assertStringInstr("İ", "i", "UTF8_BINARY_LCASE", 0);
assertStringInstr("İoi̇o12", "i̇o", "UTF8_BINARY_LCASE", 1);
assertStringInstr("i̇oİo12", "İo", "UTF8_BINARY_LCASE", 1);
assertStringInstr("abİoi̇o", "i̇o", "UTF8_BINARY_LCASE", 3);
assertStringInstr("abi̇oİo", "İo", "UTF8_BINARY_LCASE", 3);
assertStringInstr("abI\u0307oi̇o", "İo", "UTF8_BINARY_LCASE", 3);
assertStringInstr("ai̇oxXİo", "Xx", "UTF8_BINARY_LCASE", 5);
assertStringInstr("abİoi̇o", "\u0307o", "UTF8_BINARY_LCASE", 6);
assertStringInstr("aİoi̇oxx", "XX", "UTF8_BINARY_LCASE", 7);
}

private void assertFindInSet(String word, String set, String collationName,
Expand Down Expand Up @@ -878,6 +898,32 @@ public void testSubstringIndex() throws SparkException {
assertSubstringIndex("ai̇bi̇oİo12İoi̇o", "i̇o", -4, "UNICODE_CI", "İo12İoi̇o");
assertSubstringIndex("ai̇bİoi̇o12i̇oİo", "İo", -4, "UNICODE_CI", "i̇o12i̇oİo");
assertSubstringIndex("ai̇bİoi̇o12i̇oİo", "i̇o", -4, "UNICODE_CI", "i̇o12i̇oİo");
assertSubstringIndex("abi̇12", "i", 1, "UNICODE_CI", "abi̇12");
assertSubstringIndex("abi̇12", "\u0307", 1, "UNICODE_CI", "abi̇12");
assertSubstringIndex("abi̇12", "İ", 1, "UNICODE_CI", "ab");
assertSubstringIndex("abİ12", "i", 1, "UNICODE_CI", "abİ12");
assertSubstringIndex("ai̇bi̇oİo12İoi̇o", "İo", -4, "UNICODE_CI", "İo12İoi̇o");
assertSubstringIndex("ai̇bi̇oİo12İoi̇o", "i̇o", -4, "UNICODE_CI", "İo12İoi̇o");
assertSubstringIndex("ai̇bİoi̇o12i̇oİo", "İo", -4, "UNICODE_CI", "i̇o12i̇oİo");
assertSubstringIndex("ai̇bİoi̇o12i̇oİo", "i̇o", -4, "UNICODE_CI", "i̇o12i̇oİo");
assertSubstringIndex("ai̇bi̇oİo12İoi̇o", "İo", 3, "UNICODE_CI", "ai̇bi̇oİo12");
assertSubstringIndex("ai̇bi̇oİo12İoi̇o", "i̇o", 3, "UNICODE_CI", "ai̇bi̇oİo12");
assertSubstringIndex("ai̇bİoi̇o12i̇oİo", "İo", 3, "UNICODE_CI", "ai̇bİoi̇o12");
assertSubstringIndex("ai̇bİoi̇o12i̇oİo", "i̇o", 3, "UNICODE_CI", "ai̇bİoi̇o12");
assertSubstringIndex("abi̇12", "i", 1, "UTF8_BINARY_LCASE", "ab"); // != UNICODE_CI
assertSubstringIndex("abi̇12", "\u0307", 1, "UTF8_BINARY_LCASE", "abi"); // != UNICODE_CI
assertSubstringIndex("abi̇12", "İ", 1, "UTF8_BINARY_LCASE", "ab");
assertSubstringIndex("abİ12", "i", 1, "UTF8_BINARY_LCASE", "abİ12");
assertSubstringIndex("ai̇bi̇oİo12İoi̇o", "İo", -4, "UTF8_BINARY_LCASE", "İo12İoi̇o");
assertSubstringIndex("ai̇bi̇oİo12İoi̇o", "i̇o", -4, "UTF8_BINARY_LCASE", "İo12İoi̇o");
assertSubstringIndex("ai̇bİoi̇o12i̇oİo", "İo", -4, "UTF8_BINARY_LCASE", "i̇o12i̇oİo");
assertSubstringIndex("ai̇bİoi̇o12i̇oİo", "i̇o", -4, "UTF8_BINARY_LCASE", "i̇o12i̇oİo");
assertSubstringIndex("bİoi̇o12i̇o", "\u0307oi", 1, "UTF8_BINARY_LCASE", "bİoi̇o12i̇o");
assertSubstringIndex("ai̇bi̇oİo12İoi̇o", "İo", 3, "UTF8_BINARY_LCASE", "ai̇bi̇oİo12");
assertSubstringIndex("ai̇bi̇oİo12İoi̇o", "i̇o", 3, "UTF8_BINARY_LCASE", "ai̇bi̇oİo12");
assertSubstringIndex("ai̇bİoi̇o12i̇oİo", "İo", 3, "UTF8_BINARY_LCASE", "ai̇bİoi̇o12");
assertSubstringIndex("ai̇bİoi̇o12i̇oİo", "i̇o", 3, "UTF8_BINARY_LCASE", "ai̇bİoi̇o12");
assertSubstringIndex("bİoi̇o12i̇o", "\u0307oi", 1, "UTF8_BINARY_LCASE", "bİoi̇o12i̇o");
}

private void assertStringTrim(
Expand Down