Decoding old mysql UTF8 storage... how do?

Question

The following query, executed against an old MySQL database, should reveal a single UTF-8 character 'yama' for mountain.

select convert(sc_cardname using binary) as cn
  from mtg.mtg_cdb_set_cards where setcardid = 214400

Instead it yields the following 15 byte array:

[195, 165, 194, 177, 194, 177, 195, 168, 226, 128, 158, 226, 128, 176, 32]

What are these values and how do I get from there to a character identity?

For reference, the expected binary aray would be the following:

[229, 177, 177]

Update: the following code fixes the yama problem, but I don't know why:

var iconv = new Iconv('utf8','ISO-8859-1');
shortBuffer = buffer.slice(0,-9);
result = iconv.convert(shortBuffer).toString('utf8');

it's obviously not a single "character" (combining characters perhaps?). have you checked what that byte array looks like as utf8? — Karoly Horvath, Mar 02 '15 at 19:43
somehow php displays it properly, so I know that it is possible to decode it appropriately — lostinplace, Mar 02 '15 at 19:49
possible duplicate of [UTF-8 all the way through](http://stackoverflow.com/questions/279170/utf-8-all-the-way-through) — Marcus Adams, Mar 02 '15 at 20:02
this is a similar problem, but really the issue is that mysql is storing an ISO8859 string as utf8, and it's adding some weird padding to the end — lostinplace, Mar 02 '15 at 20:17

score 0 · Accepted Answer · answered Mar 27 '15 at 19:10

0

The answer was this, everything was actually encoded in LATIN1... changing the connection properties to reflect that solved the problem

answered Mar 27 '15 at 19:10

lostinplace

1 Answers1