#  (C) 2006 by Remo Dentato (rdentato@users.sourceforge.net)
#
#  Permission to use, copy, modify and distribute this code and
# its documentation for any purpose is hereby granted without fee,
# provided that the above copyright notice, or equivalent attribution
# acknowledgement, appears in all copies and supporting documentation.
#  
#   Copyright holder makes no representations about the suitability
# of this software for any purpose. It is provided "as is" without
# express or implied warranty.

# * Introduction
#   This file is both a testsuite and part of the manual for yrx.
#   Every single matching operator is shown with examples. 
#   Group of patterns are defined with a number and the pattern enclosed
# in between two slashes ('/').
#   Following strings (enclosed in double quotes) are to be matched against
# the group of patterns. The number following the string indicates which
# pattern the string matches (0 if none).

# * Characters
# Usually characters in a pattern match themselves. 

1/plain string/
"plain string" 1 {plain string} {} {} {} {} {} {} {} {}
"other string" 0 {} {} {} {} {} {} {} {} {}

# * Character class
# A character set is way to specify a set of characters at once.
1/[aeiou]/                 # wovels
2/[bcdfghjklmnpqrstvwxyz]/ # consonants
"u" 1 {u} {} {} {} {} {} {} {} {}
"x" 2 {x} {} {} {} {} {} {} {} {}
"3" 0 {} {} {} {} {} {} {} {} {}

# A set may be specified with a range
1/[A-Ga-g]/    # Notes pitch
"A" 1 {A} {} {} {} {} {} {} {} {}
"b" 1 {b} {} {} {} {} {} {} {} {}
"K" 0 {} {} {} {} {} {} {} {} {}

# To specify characters that do not belong to a given set, the '^' symbol
# is used: 
1/[^A-Z]/        # any character that is not an uppercase letter
"u" 1 {u} {} {} {} {} {} {} {} {}
"X" 0 {} {} {} {} {} {} {} {} {}
"3" 1 {3} {} {} {} {} {} {} {} {}

# Also there are predefined classes:
#  \a    alphabetic character
#  \b    backspace character (ASCII 8)
#  \c    a "control" character (ASCII 1 to ASCII 7)
#  \d    decimal digit
#  \f    form feed character (ASCII 10)
#  \h    an hexadecimal digit ([0-9A-Za-z])
#  \l    lowercase  character
#  \n    newline character (ASCII 10)
#  \oxxx the character whose code is xxx (in octal)
#  \p    a punctuation character
#  \q    alphanumeric (a letter or a digit)
#  \r    carriage return character (ASCII 13)
#  \s    a spacing character (includes newlines, tabs, etc.) 
#  \t    tab character (ASCII 8)
#  \u    uppercase  character
#  \v    vertical tab character (ASCII 10)
#  \w    a "word" character ([\q_])
#  \xhh  the character whose code is hh (in hex)
#  \y    space or tab (ASCII 32 or ASCII 9)

1/\a\l\u\d\q\s\p\h\w\y/
"PaR1S :F_ "  1 {PaR1S :F_ } {} {} {} {} {} {} {} {}
"1Ag"         0 {} {} {} {} {} {} {} {} {}

# Note that the predefined sets are influenced by the current locale
# settings. This means that "\l" may be different from "[a-z]", for example
# the accented e "è" is considered a lowercase letter if locale is set to "It".

#  Predefined sets can be used within character set definitions.
1/[\d,]+/
2/[^\a\d]+/  # Anything but a letter or a digit
"543,21"   1 {543,21} {} {} {} {} {} {} {} {}
"100.31"   1 {100} {} {} {} {} {} {} {} {}
"abc@3"    2 {@} {} {} {} {} {} {} {} {}
"xyzwt"    0 {} {} {} {} {} {} {} {} {}


# * Recognizers
#   Some commonly used patterns have been predefined:
#      .    any character (except '\0'). Depending on the switch \E it
#           can also match an escaped character.
#     \Q    quoted string. Equivalent to:
#            ("(\\.|[^"])*"|'(\\.|[^'])*')
#     \N    decimal integer number. Equivalent to:
#            [+-]?\d+
#     \H    hex integer number. Equivalent to:
#            (0x)?\h+
#     \F    floating point number. Equivalent to:
#            [+-]?\d*(.\d*)?([eE][+-]?\d*(.\d*))?
#           but doesn't match the empty string
#     \L    generalized newline. Equivalent to:
#            (\r\n?|\n) 
#     \I    identifier. Equivalent to:
#            [\a_]\w+
#     \Bxy  balanced braces. The character x and y must be
#           different. For example %B() matches "(f(g(x))".
#     \W    white space. Equivalent to:
#            [ \t]*
#
#     \A
#     \Z
            
      
# * Grouping
#   Expressions can be grouped by parenthesis. If "X" is an expression, "(X)"
# is a pattern that defines the same set of strings as "X".
1/(a[bc])/
2/ab/       # this will never match!
3/\l\l/
"ab" 1 {ab} {} {} {} {} {} {} {} {}
"ac" 1 {ac} {} {} {} {} {} {} {} {}
"bc" 3 {bc} {} {} {} {} {} {} {} {}
"00" 0 {} {} {} {} {} {} {} {} {}
 
#   Within a group it is possible to define alternative patterns. 
# If "X" and "Y" are patterns, "(X|Y)" is a pattern that defines
# the set of strings that match "X" or "Y".
1/(ab|ac)/
2/\l\l/
"ab" 1 {ab} {} {} {} {} {} {} {} {}
"ac" 1 {ac} {} {} {} {} {} {} {} {}
"bc" 2 {bc} {} {} {} {} {} {} {} {}

# * Operators
#   The following operators can be applied to a pattern X:
#   ^X       matches X at the beginning of the input string
#    X$      matches X at the end of the input string
#    X*      matches 0 or more repetition of X
#    X+      matches 1 or more repetition of X
#    X?      matches X or the empty string
#    X<m,n>  matches m to n repetition of X
#    X<m>    matches m repetition of X
#    X<m,>   matches m or more repetition of X
#    X<,n>   matches 0 to n repetition of X
#    X!      matches the empty string if X doesn't match
#    X&      matches the empty string if X matches
#    X#      matches any string whose end matches X

1/^\l\l/   # two lowercase letters at the beginning of the string
2/\l\l$/   # two lowercase letters at the end of the string
3/\l\l/    # two lowercase letters anywhere in the string 
"xx<<" 1 {xx} {} {} {} {} {} {} {} {}
">>zy" 2 {zy} {} {} {} {} {} {} {} {}
">wy<" 3 {wy} {} {} {} {} {} {} {} {}
"q11p" 0 {} {} {} {} {} {} {} {} {}
"db"   1 {db} {} {} {} {} {} {} {} {}

1/.<2,3>/
"a"     0 {} {} {} {} {} {} {} {} {}
"ab"    1 {ab} {} {} {} {} {} {} {} {}
"abc"   1 {abc} {} {} {} {} {} {} {} {}
"abcd"  1 {abc} {} {} {} {} {} {} {} {}

1/.<0,3>/
"a"     1 {a} {} {} {} {} {} {} {} {}
"ab"    1 {ab} {} {} {} {} {} {} {} {}
"abc"   1 {abc} {} {} {} {} {} {} {} {}
"abcd"  1 {abc} {} {} {} {} {} {} {} {}

1/.<2,0>/
"a"     0 {} {} {} {} {} {} {} {} {}
"ab"    1 {ab} {} {} {} {} {} {} {} {}
"abc"   1 {abc} {} {} {} {} {} {} {} {}
"abcd"  1 {abcd} {} {} {} {} {} {} {} {}

1/,\d!/   # a comma not followed by a digit
2/\a:&/   # a letter followed by a colon
",x" 1 {,} {} {} {} {} {} {} {} {}
",3" 0 {} {} {} {} {} {} {} {} {}
".3" 0 {} {} {} {} {} {} {} {} {}
"x:" 2 {x} {} {} {} {} {} {} {} {}
"9:" 0 {} {} {} {} {} {} {} {} {}
"x." 0 {} {} {} {} {} {} {} {} {}

1/''#/  # string delimited by a single quote
"''"          1 {''} {} {} {} {} {} {} {} {}
"'aaa'"       1 {'aaa'} {} {} {} {} {} {} {} {}
"'aaa'bbb'"   1 {'aaa'} {} {} {} {} {} {} {} {}

# BEWARE! Since the YRX patterns are always greedy, you can't have  
#         alternates pattern with one of the patterns matching the empty
#         string.
1/(a*|b)+/
"aaaa"        1 {aaaa} {} {} {} {} {} {} {} {}
"aaaababab"   1 {aaaa} {} {} {} {} {} {} {} {}

1/(a*b?)+/
"aaaa"        1 {aaaa} {} {} {} {} {} {} {} {}
"aaaababab"   1 {aaaababab} {} {} {} {} {} {} {} {}

# * Special Characters
# If a special character is to be matched literally, it must be 
# escaped with a backslash. Except if they are inside a character set.
1/\[.[|*#^$&!()?+{}<>\\]+\]/
"[*))((*]" 1 {[*))((*]} {} {} {} {} {} {} {} {}

# * Captures
#   Part of the matching string can be captured and returned among results.
# There can be up to 8 captures.
#    {   begin capture
#    }   end capture
#    \1  reference 1st capture
#    \2  reference 2nd capture
#    ...  
#    \8  reference 8th capture

1/{\a}\1/    # twice the same letter
2/{\a\a}\1/  # twice the same couple of letters
"xaax" 1 {aa} {a} {} {} {} {} {} {} {}
"xaxa" 2 {xaxa} {xa} {} {} {} {} {} {} {}
"abcd" 0 {} {} {} {} {} {} {} {} {}

1/{.}\1#/   # a string delimited by its first character
"'abcd'" 1 {'abcd'} {'} {} {} {} {} {} {} {}
"<abcd>" 0 {} {} {} {} {} {} {} {} {}
"$$"     1 {$$} {$} {} {} {} {} {} {} {}

1/:{\l{\d+}}:/ # nested captures
":a123:" 1 {:a123:} {a123} {123} {} {} {} {} {} {}

1/a{({bc}|{bd})}/
"abc"  1 {abc} {bc} {bc} {} {} {} {} {} {}
"abd"  1 {abd} {bd} {} {bd} {} {} {} {} {}
"xbd"  0 {} {} {} {} {} {} {} {} {}
"abe"  0 {} {} {} {} {} {} {} {} {}

1/\y*{\l+}\({[?\)]*}\)/
"xx"0 {} {} {} {} {} {} {} {} {}

# * Switch
#   A switch turns on and off a specific feature of the pattern matchin
# engine.
#    \C    case sensitive match on/off
#    \Ex   turns on escaped match for the "." recognizer   
#    \e    turns off escaped match

1/abcd/
2/ab\Ccd/
"abcd" 1 {abcd} {} {} {} {} {} {} {} {}
"abCd" 2 {abCd} {} {} {} {} {} {} {} {}
"ABcd" 0 {} {} {} {} {} {} {} {} {}

1/\E\\''#/     # Remember that "X#"" is equivalente to "(X!.)*X"
2/\E`{.}\1#/
"'abcde'"     1 {'abcde'} {} {} {} {} {} {} {} {}
"$xyz$"       2 {$xyz$} {$} {} {} {} {} {} {} {}
"$xyz`$abc$"  2 {$xyz`$abc$} {$} {} {} {} {} {} {} {}
"abcd"        0 {} {} {} {} {} {} {} {} {}

1/\E$'{[^']*}'\e{[^']*}'/
"'ab'cd'"         1 {'ab'cd'} {ab} {cd} {} {} {} {} {} {}
"'ab$'cd'ef'"     1 {'ab$'cd'ef'} {ab$'cd} {ef} {} {} {} {} {} {}
"'ab$'cd'ef$'gh'" 1 {'ab$'cd'ef$'} {ab$'cd} {ef$} {} {} {} {} {} {}