How to tame a :unicorn_face:
summary
Explains why characters break, how Unicode grows (initial 65,536 code points, later >1,048,576 across 17 planes), and why emojis in 2010 force use of characters beyond the first plane. Shows practical fixes: convert incoming text with iconv or mb_detect_encoding then iconv, avoid utf8_encode/utf8_decode, use mbstring functions, always set connection/table charset to utf8mb4 (MySQL’s utf8 is 3‑byte only). Includes the UTF‑8/16/32 byte layouts and the Java surrogate problem.
AI-written summary · about
chapters (7)
- 0:00Start
- 0:42any character out there
- 1:44Definitions
- 4:48Characterset Encoding
- 28:34Single byte
- 31:58Always use UTF-8!
- 44:38Questions?
related talks
15mCharacter set and collation basics
01h 02mThere's No Such Thing as Plain Text
59mBoosting MySQL Performance
45mThe wonderful world of collations, encodings and text storage
19mThe Invisible Charset That Silently Breaks MySQL Replication
34mMySQL 8 -- A New Beginning
52mCollations from A to Z
55mMySQL 8.0: Not Only Good, it's GREAT