uros-db commented on code in PR #45643:
URL: https://github.com/apache/spark/pull/45643#discussion_r1570050123
##########
common/unsafe/src/test/java/org/apache/spark/unsafe/types/CollationSupportSuite.java:
##########
@@ -261,6 +261,84 @@ public void testEndsWith() throws SparkException {
assertEndsWith("The i̇o", "İo", "UNICODE_CI", true);
}
+ private void assertStringInstr(String string, String substring, String
collationName,
+ Integer expected) throws SparkException {
+ UTF8String str = UTF8String.fromString(string);
+ UTF8String substr = UTF8String.fromString(substring);
+ int collationId = CollationFactory.collationNameToId(collationName);
+ assertEquals(expected, CollationSupport.StringInstr.exec(str, substr,
collationId) + 1);
+ }
+
+ @Test
+ public void testStringInstr() throws SparkException {
+ assertStringInstr("aaads", "Aa", "UTF8_BINARY", 0);
+ assertStringInstr("aaaDs", "de", "UTF8_BINARY", 0);
+ assertStringInstr("aaads", "ds", "UTF8_BINARY", 4);
+ assertStringInstr("xxxx", "", "UTF8_BINARY", 1);
+ assertStringInstr("", "xxxx", "UTF8_BINARY", 0);
+ assertStringInstr("test大千世界X大千世界", "大千", "UTF8_BINARY", 5);
+ assertStringInstr("test大千世界X大千世界", "界X", "UTF8_BINARY", 8);
+ assertStringInstr("aaads", "Aa", "UTF8_BINARY_LCASE", 1);
+ assertStringInstr("aaaDs", "de", "UTF8_BINARY_LCASE", 0);
+ assertStringInstr("aaaDs", "ds", "UTF8_BINARY_LCASE", 4);
+ assertStringInstr("xxxx", "", "UTF8_BINARY_LCASE", 1);
+ assertStringInstr("", "xxxx", "UTF8_BINARY_LCASE", 0);
+ assertStringInstr("test大千世界X大千世界", "大千", "UTF8_BINARY_LCASE", 5);
+ assertStringInstr("test大千世界X大千世界", "界x", "UTF8_BINARY_LCASE", 8);
+ assertStringInstr("aaads", "Aa", "UNICODE", 0);
+ assertStringInstr("aaads", "aa", "UNICODE", 1);
+ assertStringInstr("aaads", "de", "UNICODE", 0);
+ assertStringInstr("xxxx", "", "UNICODE", 1);
+ assertStringInstr("", "xxxx", "UNICODE", 0);
+ assertStringInstr("test大千世界X大千世界", "界x", "UNICODE", 0);
+ assertStringInstr("test大千世界X大千世界", "界X", "UNICODE", 8);
+ assertStringInstr("aaads", "AD", "UNICODE_CI", 3);
+ 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);
+ }
Review Comment:
```suggestion
assertStringInstr("abİo12", "i̇o", "UNICODE_CI", 3);
assertStringInstr("abi̇o12", "İo", "UNICODE_CI", 3);
}
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
