8 no utf8; # this test contains raw 8-bit data on purpose; don't switch to \x{}
14 # This table is based on Markus Kuhn's UTF-8 Decode Stress Tester,
15 # http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt,
16 # version dated 2000-09-02.
18 # Note the \0 instead of a raw zero byte in 2.1.1: for example
19 # GNU patch v2.1 has "issues" with raw zero bytes.
21 my @MK = split(/\n/, <<__EOMK__);
23 1.1.1 y "κόσμε" - 11 ce:ba:e1:bd:b9:cf:83:ce:bc:ce:b5 5
25 2.1 First possible sequence of certain length
27 2.1.2 y "
\80" 80 2 c2:80 1
28 2.1.3 y "ࠀ" 800 3 e0:a0:80 1
29 2.1.4 y "𐀀" 10000 4 f0:90:80:80 1
30 2.1.5 y "" 200000 5 f8:88:80:80:80 1
31 2.1.6 y "" 4000000 6 fc:84:80:80:80:80 1
32 2.2 Last possible sequence of certain length
33 2.2.1 y "
\7f" 7f 1 7f 1
34 2.2.2 y "߿" 7ff 2 df:bf 1
35 # The ffff is illegal unless UTF8_ALLOW_FFFF
Software error:
Malformed UTF-8 character (fatal) at /var/www/git.shadowcat.co.uk/docroot/gitweb/gitweb.cgi line 1024, <$fd> line 36.
For help, please send mail to the webmaster (chrisj@shadowcatsystems.co.uk), giving this error message
and the time and date of the error.