Module: LongRegexen

Defined in:

lib/coderunner/long_regexen.rb

Constant Summary

FLOAT =

/(?<float>\-?(?:(?>\d+\.\d*)|(?>\d*\.\d+))(?:[eE][+-]?\d+)?)/

FLOAT_WSP =

allows a space between the minus sign and the number

/(?<float>\-?\s*(?:(?>\d+\.\d*)|(?>\d*\.\d+))(?:[eE][+-]?\d+)?)/

INT =

INTEGER = /(?<int>\-?\d++)/

NUMBER =

Regexp.new("(?<number>#{FLOAT.verbatim}|#{INTEGER.verbatim})")

STRING =

/(?:"(?<string>(?:[^"]|\\")*)")|(?:'(?<string>(?:[^']|\\")*)')/

EQUALITY =

/\b	#a word boundary

(?<name>[A-Za-z_]\w*)  # the name, which must be a single word (not beginning 
				# with a digit) followed by

\s*=\s*    # an equals sign (possibly with whitespace either side), then

(?<default>(?>    # the default answer, which can be either:

	(?<float>\-?(?:(?>\d+\.\d*)|(?>\d*\.\d+))(?:[eE][+-]?\d+)?) # a floating point number
	
	|						#or

	(?<int>\-?\d++)	# an integer

	|						#or

	(?:(?<delim>["'])(?<string>.*?)\k<delim>) # a string in quotes: '' or ""

	|						# or

	(?:(?<word>\S+)(?=\s|\)|\]|[\n\r]+)) # a single word containing no spaces 
					# which must be followed by a space or ) or ] or a new line

))/x

KEYVALUE =

/<key>		#  <key> then

(?<name>[A-Za-z_]\w*)  # the name, which must be a single word (not beginning 
				# with a digit) followed by

<\/key>\s*    # <\/key> (possibly with following whitespace), then

<value>  	# <value>, then

(?<default>    # the default answer, which can be either:


	(?<float>\-?(?:(?>\d+\.\d*)|(?>\d*\.\d+))(?:[eE]\-?\d+)?) # a floating point number 
	
	|						#or

	(?<int>\-?\d+)		# an integer

	|						#or

	(?<string>[^<]*) 	# a string (containing no '<'s)

)

<value>  	# <\/value>

/x

TAG =

/<			# < 

[^\/\S>]+		#  some word (containing no '\/'s or '>'s)

\s+ 		#some whitespace

name=

(?<delim>["'])  # a double or single quote

(?<name>[A-Za-z_]\w*) # the name, which must be a single word (not beginning 
				# with a digit) followed by

\k<delim> 	#a closing quote  

\s+ 		#some whitespace

(?<delimnext>["'])  # a double or single quote

(?<default>    # the default answer, which can be either:


	(?<float>\-?(?:(?>\d+\.\d*)|(?>\d*\.\d+))(?:[eE]\-?\d+)?) # a floating point number 
	
	|						#or

	(?<int>\-?\d+)	# an integer

	|						#or

	(?<string>[^"']*) # a string (containing no quotation marks)

)

\k<delimnext> 	#a closing quote  

\s* 		#some whitespace

>		# >

/x

NAMRSQBQS =

<comp> : only complete sets of delimiters [], “”, ”, // or complete method calls: meth(params) or no delimiters at all: ensures that any following expressions are not inside open square brackets, quotation marks or regexen, or inside method parameter lists

NOT_IN_A_METHOD_OR_IN_A_REGEXP_OR_IN_SQUARE_BRACKETS_OR_IN_QUOTES_OR_A_SYMBOL = 
	
	    /(?<comp> 
		(?:   #a set of three options 
(?:  #option 1: complete sets of delimiters 
			(?<nest> #sub-expression denoting nested delimiters
 (?:   #  A: meth(stuff)
				(?:[A-Za-z_]\d*?) #method name ending in numbers or a letter
				\(  #begin brackets
  (?:
					[^()\[\]"\/'] #no delimiters
					|
					\g<nest> #complete nested sub-expression
  )*
				\)
 )
 |
 (?:   #  B: [stuff]
				\[  #begin square brackets
  (?:
					[^()\[\]"\/'] #no delimiters
					|
					\g<nest> #complete nested sub-expression
  )*
				\] 
 )
 |
 (?:   #  C: \/stuff\/  or "stuff" or 'stuff'
				(?<delim>[\/"'])  #begin symmetrical delimiters
  (?:
					[^()\[\]"\/'] #no delimiters
					|
					\g<nest> #complete nested sub-expression
  )*
				\k<delim> #end symmetrical delimiters
 )
			) # end sub-expression denoting correctly nested delimiters 
)
|
[^()\[\]"\/'] #option 2: not delimiters
|
(?:  # option 3: open brackets with no methods call before them
			(?:
 \W\d*  # possibly some kind of number but not a name
 |
 ^
			)
			\(
)
		)*? #as many of the above options as you like
	    )
	    (?: #Here we specify the immediately preceding letter (it cannot be a dot or a colon)
		[^.:()\[\]"\/'] #Not a dot, a colon or a delimiter
		| 
		(?: # not a method call 
\W\d*  
|
^
		)
		\( #followed by an open bracket
		| 
		^ #we can match the beginning of the string
)/x