check field length against remaining buffer in mysql Field::Parse

[ubeddulla]

What problem does this PR solve?

Issue Number: resolve N/A

Problem Summary:

The text-protocol MysqlReply::Field::Parse in src/brpc/policy/mysql/mysql_reply.cpp reads a length-encoded field length from the server and calls buf.cutn(&str, len) without checking len against the remaining buffer. IOBuf::cutn clamps to what is available, so a field whose length-encoded prefix claims more bytes than remain makes the string branch allocate len arena bytes, copy only the fewer bytes actually present, and then publish _data.str as a StringPiece of len bytes. The tail is uninitialized arena memory that gets read back by the caller, and the short read desyncs the packet stream. A malicious or compromised MySQL server can trigger it through any text result set. Every sibling length-encoded read in the same file (the binary Field::Parse, Column::Parse, the auth-plugin path, ParseBinaryTime/ParseBinaryDataTime) already rejects len > buf.size(); only this text overload was missing the check.

What is changed and the side effects?

Changed:

Added the same len > buf.size() guard before the cutn in the text Field::Parse, returning PARSE_ERROR_ABSOLUTELY_WRONG on an oversized length. Valid result sets are unaffected since a well-formed field always fits in the buffer.

Side effects:

• Performance effects: none, a single size comparison per field.

• Breaking backward compatibility: none.

---

Check List:

• Please make sure your changes are compilable.
• When providing us with a new feature, it is best to add related tests.
• Please follow Contributor Covenant Code of Conduct.

[Copilot]

Pull request overview

This PR hardens the MySQL text-protocol reply parser (MysqlReply::Field::Parse) by rejecting length-encoded field sizes that exceed the remaining packet buffer, preventing uninitialized memory exposure and stream desynchronization when parsing malformed/malicious server responses.

Changes:

• Add a len > buf.size() guard before buf.cutn(&str, len) in the text-protocol MysqlReply::Field::Parse.
• Log and return PARSE_ERROR_ABSOLUTELY_WRONG when an oversized length is encountered.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[ubeddulla]

Good call, added one in test/brpc_mysql_reply_parse_unittest.cpp. It crafts a text result set whose single row field uses a 0xFD lenenc prefix claiming 0xFFFFFF bytes with nothing left in the buffer, and asserts ConsumePartialIOBuf returns PARSE_ERROR_ABSOLUTELY_WRONG. There's also a well-formed-field case so the guard doesn't reject legitimate rows.