Suppose you need to truncate an arbitrary-length unicode string to fit into a fixed byte-length field.
In this talk I introduce the Unicode::Truncate module, but not after we go down a deep rabbit-hole of unicode topics including encodings, surrogate pairs, over-long UTF-8, combining characters, normalisation forms, extended grapheme clusters, and unicode consortium test-suites.
Are there security implications of unicode? How many bytes can a single unicode character take? Which writing system is special-cased in the unicode segmentation standard?
In addition, we'll go over state-machine parsing with Ragel and the Inline::Filters::Ragel module, distributing Inline modules with Inline::Module::LeanDist, perl's "utf8 flag", and zero-copy string truncation.