text.mod/regex.mod/pcre/testdata/testinput14

# These test special UTF and UCP features of DFA matching. The output is
# different for the different widths.

#subject dfa

# ---------------------------------------------------- 
# These are a selection of the more comprehensive tests that are run for
# non-DFA matching.

/X/utf
    XX\x{d800}
    XX\x{d800}\=offset=3
    XX\x{d800}\=no_utf_check
    XX\x{da00}
    XX\x{da00}\=no_utf_check
    XX\x{dc00}
    XX\x{dc00}\=no_utf_check
    XX\x{de00}
    XX\x{de00}\=no_utf_check
    XX\x{dfff}
    XX\x{dfff}\=no_utf_check
    XX\x{110000}
    XX\x{d800}\x{1234}
          
/badutf/utf
    X\xdf
    XX\xef
    XXX\xef\x80
    X\xf7
    XX\xf7\x80
    XXX\xf7\x80\x80

/shortutf/utf
    XX\xdf\=ph
    XX\xef\=ph
    XX\xef\x80\=ph
    \xf7\=ph
    \xf7\x80\=ph
    
# ---------------------------------------------------- 
# UCP and casing tests - except for the first two, these will all fail in 8-bit
# mode because they are testing UCP without UTF and use characters > 255.

/\x{c1}/i,no_start_optimize
\= Expect no match
    \x{e1}

/\x{c1}+\x{e1}/iB,ucp
    \x{c1}\x{c1}\x{c1}
    \x{e1}\x{e1}\x{e1} 

/\x{120}\x{c1}/i,ucp,no_start_optimize
    \x{121}\x{e1}

/\x{120}\x{c1}/i,ucp
    \x{121}\x{e1}

/[^\x{120}]/i,no_start_optimize
    \x{121}

/[^\x{120}]/i,ucp,no_start_optimize
\= Expect no match
    \x{121}

/[^\x{120}]/i
    \x{121}

/[^\x{120}]/i,ucp
\= Expect no match
    \x{121}
    
/\x{120}{2}/i,ucp
    \x{121}\x{121}

/[^\x{120}]{2}/i,ucp
\= Expect no match
    \x{121}\x{121}

# ---------------------------------------------------- 

# End of testinput14