]> mj.ucw.cz Git - libucw.git/commit
projects / libucw.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 5f099e3) | patch
Unicode: Reject denormalized UTF-8 sequences
authorMartin Mares <mj@ucw.cz>
Wed, 8 Jul 2015 21:58:21 +0000 (23:58 +0200)
committerMartin Mares <mj@ucw.cz>
Wed, 8 Jul 2015 21:58:21 +0000 (23:58 +0200)
commit4905e7908d5a7d37ac1b0d0ac18243b9ae6a381d
tree8c32954d93c7f89cd6edaceedf42d43961a67806tree | snapshot
parent5f099e3d6fb8ea3fea3e18aa93bda291bbdb894acommit | diff
Unicode: Reject denormalized UTF-8 sequences

When there are multiple possible encodings of a single Unicode
codepoint, the standard requires to use the shortest one. Not checking
this requirement on input has been observed to cause weird security
problems in some software, so better be careful.
ucw/ff-unicode.c diff | blob | history
ucw/ff-unicode.h diff | blob | history
ucw/ff-unicode.t diff | blob | history
ucw/unicode.h diff | blob | history
ucw/unicode.t diff | blob | history
UCW libraries