MySQL Bugs: #77513: Some hungarian characters fail to save correctly

Bug #77513	Some hungarian characters fail to save correctly
Submitted:	26 Jun 2015 21:42	Modified:	28 Aug 2015 3:47
Reporter:	Rob Dylan	Email Updates:
Status:	Can't repeat	Impact on me:	None
Category:	MySQL for Windows: MySQL for Excel	Severity:	S3 (Non-critical)
Version:	1.3.4	OS:	Windows (8.1 x64)
Assigned to:		CPU Architecture:	Any

Description:
When i try to edit the mysql database(utf8_bin collation) with "mysql for excel" (in Excel 2013 x64) some hungarian characters fail to save correctly in the database.

Affected characters: ő (o with double acute accent) and ű (u with double acute accent). ő is saved as o, while ű is saved as u in mysql.

Other hungarian characters like í,é,á,ó,ü,ö work as expected.

How to repeat:
Try to save ő,ű,Ő,Ű in any database with any collation (tried utf8_bin, utf8_general_ci, utf8_hungarian_ci, utf8_unicode_ci and even utf8mb). These characters just will not get saved correctly.

Tested on version 1.3.4, I could save the following text using double acute accent:

A megoldás egyszerű
Egyszerű TEST

The u with double acute accent was saved successfully, verified with MySQL Workbench the data was correctly saved.

The test I performed was creating a new schema with utf8_general_ci collation, then exporting some data to feed the new table with it. Then edited the new table to insert more data with double acute accent.

Since I can't reproduce this, this sounds more like a configuration problem.
Although the schema is using utf8, the server variables need to be set as well to UTF8, you can use MySQL Workbench to inspect the value of other system variables like character_set_client, character_set_connection, character_set_server and character_set_system.

Have you tried saving data with double acute accent directly from the MySQL Server console??

The real test for whether it is saved correctly is to do SELECT HEX(col)... -- It should come back with hex C5B1.  If, instead, you get C385C2B1, you are a victim of "double encoding".