doc: Clarify that Unicode normalization or comparison algorithms are not used
authorPetri Lehtinen <petri@digip.org>
Sat, 14 May 2011 18:08:02 +0000 (21:08 +0300)
committerPetri Lehtinen <petri@digip.org>
Sun, 15 May 2011 09:39:10 +0000 (12:39 +0300)
doc/conformance.rst

index 6359f22..34d60bd 100644 (file)
@@ -30,6 +30,12 @@ error::
 
 All other Unicode codepoints U+0001 through U+10FFFF are allowed.
 
+Unicode normalization or any other transformation is never performed
+on any strings (string values or object keys). When checking for
+equivalence of strings or object keys, the comparison is performed
+byte by byte between the original UTF-8 representations of the
+strings.
+
 Numbers
 =======