Hacker News Reader

Possibly a difference in implementation of unicode between validation and persistence/use. Validates successfully due to one or more "valid" unicode characters -> translated to zero characters due to stripping of invalid unicode character at some point after validation.

View on HN

Maybe related to MySQL's 3-byte "utf8" charset which doesn't actually support unicode properly? (you are supposed to use "utf8mb4" nowadays though a lot of material still refers to "utf8")

Found this out when someone pushed a git commit message with a 4byte emoji and it broke our CI server -_-

Reply on news.ycombinator.com