removing the fix for invalid input in FuzzCSVReadTest fixes the broken support for valid input in #497

[pjfanning]

jackson-dataformats-text/csv/src/main/java/com/fasterxml/jackson/dataformat/csv/impl/UTF8Reader.java

Lines 406 to 418 in d43289d

	if (!canModifyBuffer()) {
	// 15-Aug-2022, tatu: Occurs (only) if we have half-decoded UTF-8
	// characters; uncovered by:
	//
	// https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=50036
	//
	// and need to be reported as IOException
	if (_inputSource == null) {
	throw new IOException(String.format(
	"End-of-input after first %d byte(s) of a UTF-8 character: needed at least one more",
	available));
	}
	}

invalid input that we might accidentally treat as valid if we remove the code
https://github.com/FasterXML/jackson-dataformats-text/blob/d43289df0be12aa0d8ee9cea97405cca337049bf/csv/src/test/java/com/fasterxml/jackson/dataformat/csv/deser/FuzzCSVReadTest.java

valid input that we will now properly support if we remove the code

jackson-dataformats-text/csv/src/test/java/com/fasterxml/jackson/dataformat/csv/tofix/UnicodeCSVRead497Test.java

Line 23 in d43289d

public void testUnicodeAtEnd() throws Exception

Would it be possible to consider removing the code and trying later to come up with another fix for the fuzz case?
I think it is better to fix #497 and later come back to the fuzz case.

[pjfanning]

@cowtowncoder #546 seems to indicate the removal of the code is fine - the fuzz case still behaves correctly, just the exception message changes a little

[cowtowncoder]

Merged #546 with all the rework; I think we can close this.