By Rick James

2016-07-14 00:04:19 8 Comments

I tried to use UTF-8 and ran into trouble.

I have tried so many things; here are the results I have gotten:

  • ???? instead of Asian characters. Even for European text, I got Se?or for Señor.
  • Strange gibberish (Mojibake?) such as Señor or 新浪新闻 for 新浪新闻.
  • Black diamonds, such as Se�or.
  • Finally, I got into a situation where the data was lost, or at least truncated: Se for Señor.
  • Even when I got text to look right, it did not sort correctly.

What am I doing wrong? How can I fix the code? Can I recover the data, if so, how?


@castro_pereira 2019-03-24 19:22:27

I had similar issues with 2 of my projects, after a server migration. After searching and trying a lot of solutions i came across with this one:


After adding this line to my config file everything works fine!

I found this solution for mysqli when i was looking to solve a insert from html query

good luck!

@Rick James 2019-03-26 02:05:17

Yes, that is one of several things that can cause character set problems. Note: that syntax is valid for PHP, not other app languages, and only if using mysqli, not PDO.

@SIDU 2016-07-14 00:40:11

Funny how you answer your own question :)

  1. Set your code IDE language to UTF8

  2. Add to your webpage header where you collect data form.

  3. Check your MySQL table definition looks like this:

    CREATE TABLE your_table (
  4. If you are using PDO, make sure

    $options = array(PDO::MYSQL_ATTR_INIT_COMMAND=>'SET NAMES utf8'); 
    $dbL = new PDO($pdo, $user, $pass, $options);

If you already got a large database with above problem, you can try SIDU to export with correct charset, and import back with UTF8. Good luck

@Rick James 2016-07-14 03:03:43

(Answering one's own question is a feature of this forum.) I have been working for years trying to make the answer this concise, yet complete.

@Rick James 2016-07-14 03:04:36

The DEFAULT CHARSET for a table is just that, a default. It can, and sometimes should, be overridden on the column definition.

@Rick James 2016-07-14 03:06:45

PDO is better done with the charset option: $db = new PDO('dblib:host=host;dbname=db;charset=UTF8', $user, $pwd); (This is listed in the link to my 'charcoll' document.)

@SIDU 2016-07-14 03:09:25

You are 20K more expert than me :) Yes you can set charset for a column. Try not to over-use it. End up with more management time. Similarly you can grant access to a certain column of a MySQL table. However you do not have to use it unless you do not have better alternative.

@Rick James 2016-07-14 03:23:39

I advocate any hex string or truely fixed length string should, in most cases, be ascii. This is to override the 'new' default of utf8, which takes 3 times the space for CHAR(n). (This has bitten some people who did not notice the system-wide default changing to utf8.)

@ppmakeitcount 2017-10-08 02:16:22

If i alter the database to utf8, should the database be restarted to take the changes? And is there any chances of data loss.

@spencer7593 2017-11-09 16:01:31

@ppmakeitcount: no, the ALTER DATABASE statement does not require a restart of MySQL to take affect. However, changing the default characterset for a database does not affect any tables currently in the database; it only has an effect on new tables e.g. CREATE TABLE which do not specify a default characterset for the table; that's when the database default characterset comes into play. (Similarly, changing the default characterset of the table does not affect columns already in the table; it only has an effect on columns added to the table, when no column characterset is specified.

@Rick James 2018-05-10 04:00:12

@ppmakeitcount - There is no need to restart the database. However, be sure to use the appropriate ALTER, as discussed here:

@Rick James 2016-07-14 00:04:19

This problem plagues the participants of this site, and many others.

You have listed the five main cases of CHARACTER SET troubles.

Best Practice

Going forward, it is best to use CHARACTER SET utf8mb4 and COLLATION utf8mb4_unicode_520_ci. (There is a newer version of the Unicode collation in the pipeline.)

utf8mb4 is a superset of utf8 in that it handles 4-byte utf8 codes, which are needed by Emoji and some of Chinese.

Outside of MySQL, "UTF-8" refers to all size encodings, hence effectively the same as MySQL's utf8mb4, not utf8.

I will try to use those spellings and capitalizations to distinguish inside versus outside MySQL in the following.

Overview of what you should do

  • Have your editor, etc. set to UTF-8.
  • HTML forms should start like <form accept-charset="UTF-8">.
  • Have your bytes encoded as UTF-8.
  • Establish UTF-8 as the encoding being used in the client.
  • Have the column/table declared CHARACTER SET utf8mb4 (Check with SHOW CREATE TABLE.)
  • <meta charset=UTF-8> at the beginning of HTML

UTF-8 all the way through

More details for computer languages (and its following sections)

Test the data

Viewing the data with a tool or with SELECT cannot be trusted. Too many such clients, especially browsers, try to compensate for incorrect encodings, and show you correct text even if the database is mangled. So, pick a table and column that has some non-English text and do

SELECT col, HEX(col) FROM tbl WHERE ...

The HEX for correctly stored UTF-8 will be

  • For a blank space (in any language): 20
  • For English: 4x, 5x, 6x, or 7x
  • For most of Western Europe, accented letters should be Cxyy
  • Cyrillic, Hebrew, and Farsi/Arabic: Dxyy
  • Most of Asia: Exyyzz
  • Emoji and some of Chinese: F0yyzzww
  • More details

Specific causes and fixes of the problems seen

Truncated text (Se for Señor):

  • The bytes to be stored are not encoded as utf8mb4. Fix this.
  • Also, check that the connection during reading is UTF-8.

Black Diamonds with question marks (Se�or for Señor); one of these cases exists:

Case 1 (original bytes were not UTF-8):

  • The bytes to be stored are not encoded as utf8. Fix this.
  • The connection (or SET NAMES) for the INSERT and the SELECT was not utf8/utf8mb4. Fix this.
  • Also, check that the column in the database is CHARACTER SET utf8 (or utf8mb4).

Case 2 (original bytes were UTF-8):

  • The connection (or SET NAMES) for the SELECT was not utf8/utf8mb4. Fix this.
  • Also, check that the column in the database is CHARACTER SET utf8 (or utf8mb4).

Black diamonds occur only when the browser is set to <meta charset=UTF-8>.

Question Marks (regular ones, not black diamonds) (Se?or for Señor):

  • The bytes to be stored are not encoded as utf8/utf8mb4. Fix this.
  • The column in the database is not CHARACTER SET utf8 (or utf8mb4). Fix this. (Use SHOW CREATE TABLE.)
  • Also, check that the connection during reading is UTF-8.

Mojibake (Señor for Señor): (This discussion also applies to Double Encoding, which is not necessarily visible.)

  • The bytes to be stored need to be UTF-8-encoded. Fix this.
  • The connection when INSERTing and SELECTing text needs to specify utf8 or utf8mb4. Fix this.
  • The column needs to be declared CHARACTER SET utf8 (or utf8mb4). Fix this.
  • HTML should start with <meta charset=UTF-8>.

If the data looks correct, but won't sort correctly, then either you have picked the wrong collation, or there is no collation that suits your need, or you have Double Encoding.

Double Encoding can be confirmed by doing the SELECT .. HEX .. described above.

é should come back C3A9, but instead shows C383C2A9
The Emoji 👽 should come back F09F91BD, but comes back C3B0C5B8E28098C2BD

That is, the hex is about twice as long as it should be. This is caused by converting from latin1 (or whatever) to utf8, then treating those bytes as if they were latin1 and repeating the conversion. The sorting (and comparing) does not work correctly because it is, for example, sorting as if the string were Señor.

Fixing the Data, where possible

For Truncation and Question Marks, the data is lost.

For Mojibake / Double Encoding, ...

For Black Diamonds, ...

(I'll have to continue this in another question/answer.)

@Rick James 2016-10-06 01:19:28

@david_adler 2018-01-15 16:50:51

If the client, database and tables are in utf8mb4 I seem to be able to store emojis fine. Some blogs suggest also setting collation-server and character-set-server in mysqld. Do I really need to change mysqld what difference does server setting make?

@Rick James 2018-01-15 18:47:28

@david_adler - There are multiple ways to get the effect of those settings. The best is to use the clients connection parameters. Second best by executing SET NAMES utf8mb4 right after connecting. After all, this is declaring the encoding in the client.

@Rick James 2018-05-10 03:56:55

With MySQL 8.0 (now released), the default is utf8mb4 and utf8mb4_0900_ai_ci. Most users should use them without considering the other charsets and collations.

@Rick James 2018-06-29 17:06:06

Tips on configuring Python, PHP, and about 40 other languages

@Rick James 2019-01-03 22:35:32

Another note: If a FUNCTION or STORED PROCEDURE is involved, you may have not been using the desired charset when you created it. DROP it, SET NAMES; re-CREATE it.

@paul 2016-08-03 16:13:56

Depending on how the server is setup you have to change the encode accordingly. utf8 from what you said should work the best however if your getting weird characters it might help if you change the webpage Encode to Ansi. This helped me when I was setting up a PHP MYSQLI this might help you understand more

@Rick James 2016-08-03 16:27:07

Notepad's ANSI is probably closest to MySQL's latin1. The 0x93 in that link is and probably came from some place like Word. You could either convert to utf8 (hex E2809C) or tell MySQL that the data is latin1 and hope that you don't trip up somewhere else.

Related Questions

Sponsored Content

9 Answered Questions

[SOLVED] What is Unicode, UTF-8, UTF-16?

14 Answered Questions

[SOLVED] UTF-8 all the way through

26 Answered Questions

13 Answered Questions

[SOLVED] What is the difference between UTF-8 and Unicode

20 Answered Questions

16 Answered Questions

8 Answered Questions

13 Answered Questions

7 Answered Questions

[SOLVED] Why does modern Perl avoid UTF-8 by default?

  • 2011-05-28 15:12:36
  • w.k
  • 96028 View
  • 551 Score
  • 7 Answer
  • Tags:   perl unicode utf-8

12 Answered Questions

[SOLVED] UTF-8, UTF-16, and UTF-32

Sponsored Content