[CONJ-805] >1Byte UTF8 character issue if maxFieldSize is set

[MarkusLutum]

Detected at one of my clients.
They use polish special characters like ść
The length in TextRowProtocol is set to 2 for each of those characters so 4 in our example.
With this the current code
return new String(buf, pos, Math.min(maxFieldSize * 3, length), StandardCharsets.UTF_8) .substring(0, Math.min(maxFieldSize, length));
results into an ArrayOutOfBoundsError at the substring call.

I added a test case and a quick fix for it.
Travis fails on environment issues. Nothing to do with my change.

[MarkusLutum]

Files a bug for this as well:
https://jira.mariadb.org/browse/CONJ-805

[harawata]

Both the current implementation and the proposed fix count (or try to count) the length of the string, however, the doc says...

Sets the limit for the maximum number of bytes that can be returned for character and binary column values in a ResultSet object produced by this Statement object.

So, it should look something like this, shouldn't it?

if (maxFieldSize > 0) {
  return new String(buf, pos, Math.min(maxFieldSize, length), StandardCharsets.UTF_8);
}

[rusher]

good point @harawata

In JDBC from spec :

8.3.1 Silent Truncation
The Statement.setMaxFieldSize method allows a maximum size (in bytes) to
be set. This limit applies only to the BINARY, VARBINARY, LONGVARBINARY, CHAR,
VARCHAR, LONGVARCHAR, NCHAR, NVARCHAR, and LONGNVARCHAR data types.
If a limit has been set using setMaxFieldSize and there is an attempt to read data
that exceeds the limit, any truncation that occurs as a result of exceeding the set limit
will not be reported.

CHAR and VARCHAR are explicitly listed.

This is then not just a bug correction, but a change of behavior compares to current driver implementation, so has then to be in the next 2.7 version, not in a corrective patch version.

[MarkusLutum]

@rusher: I change my pull request to @harawata s suggestion.