Make the UTF-8 decoding stricter and more verbose when
authorJarkko Hietaniemi <jhi@iki.fi>
Tue, 24 Oct 2000 02:55:33 +0000 (02:55 +0000)
committerJarkko Hietaniemi <jhi@iki.fi>
Tue, 24 Oct 2000 02:55:33 +0000 (02:55 +0000)
commitba210ebec161cde003bc967e8e460c72f71fb70c
tree7eefd78e8e365cbf64ddf49314681d17b83c3025
parent177b92d2814bfc842f28f277e0a2f353c652a5e3
Make the UTF-8 decoding stricter and more verbose when
malformation happens.  This involved adding an argument
to utf8_to_uv_chk(), which involved changing its prototype,
and prefer STRLEN over I32 for the UTF-8 length, which as
a domino effect necessitated changing the prototypes of
scan_bin(), scan_oct(), scan_hex(), and reg_uni().
The stricter UTF-8 decoding checking uses Markus Kuhn's
UTF-8 Decode Stress Tester from
http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt

p4raw-id: //depot/perl@7416
22 files changed:
doop.c
embed.h
embed.pl
handy.h
op.c
perl.c
perlapi.c
pod/perlapi.pod
pod/perldiag.pod
pod/perlunicode.pod
pp.c
pp_ctl.c
proto.h
regcomp.c
regexec.c
sv.c
t/pragma/utf8.t
t/pragma/warn/utf8
toke.c
utf8.c
utf8.h
util.c