swordjs/libs/bcv/src/template/regexps.coffee

swordjs - access modules from crosswire.org/sword in JS

bcv_parser::regexps.space = "[\\s\\xa0]"
bcv_parser::regexps.escaped_passage = ///
	(?:^ | $PRE_PASSAGE_ALLOWED_CHARACTERS )	# Beginning of string or not in the middle of a word or immediately following another book. Only count a book if it's part of a sequence: `Matt5John3` is OK, but not `1Matt5John3`
		(
			# Start inverted book/chapter (cb)
			(?:
				  (?: ch (?: apters? | a?pts?\.? | a?p?s?\.? )? \s*
					\d+ \s* (?: [\u2013\u2014\-] | through | thru | to) \s* \d+ \s*
					(?: from | of | in ) (?: \s+ the \s+ book \s+ of )?\s* )
				| (?: ch (?: apters? | a?pts?\.? | a?p?s?\.? )? \s*
					\d+ \s*
					(?: from | of | in ) (?: \s+ the \s+ book \s+ of )?\s* )
				| (?: \d+ (?: th | nd | st ) \s*
					ch (?: apter | a?pt\.? | a?p?\.? )? \s* #no plurals here since it's a single chapter
					(?: from | of | in ) (?: \s+ the \s+ book \s+ of )? \s* )
			)? # End inverted book/chapter (cb)
			\x1f(\d+)(?:/\d+)?\x1f		#book
				(?:
				    /\d+\x1f				#special Psalm chapters
				  | $VALID_CHARACTERS
				  | $TITLE (?! [a-z] )		#could be followed by a number
				  | $PASSAGE_COMPONENTS
				  | $AB (?! \w )			#a-e allows 1:1a
				  | $						#or the end of the string
				 )+
		)
	///gi
# These are the only valid ways to end a potential passage match. The closing parenthesis allows for fully capturing parentheses surrounding translations (ESV**)**. The last one, `[\d\x1f]` needs not to be +; otherwise `Gen5ff` becomes `\x1f0\x1f5ff`, and `adjust_regexp_end` matches the `\x1f5` and incorrectly dangles the ff.
bcv_parser::regexps.match_end_split = ///
	  \d \W* $TITLE
	| \d \W* $FF (?: [\s\xa0*]* \.)?
	| \d [\s\xa0*]* $AB (?! \w )
	| \x1e (?: [\s\xa0*]* [)\]\uff09] )? #ff09 is a full-width closing parenthesis
	| [\d\x1f]
	///gi
bcv_parser::regexps.control = /[\x1e\x1f]/g
bcv_parser::regexps.pre_book = "$PRE_BOOK_ALLOWED_CHARACTERS"

bcv_parser::regexps.first = "$FIRST\\.?#{bcv_parser::regexps.space}*"
bcv_parser::regexps.second = "$SECOND\\.?#{bcv_parser::regexps.space}*"
bcv_parser::regexps.third = "$THIRD\\.?#{bcv_parser::regexps.space}*"
bcv_parser::regexps.range_and = "(?:[&\u2013\u2014-]|$AND|$TO)"
bcv_parser::regexps.range_only = "(?:[\u2013\u2014-]|$TO)"
# Each book regexp should return two parenthesized objects: an optional preliminary character and the book itself.
bcv_parser::regexps.get_books = (include_apocrypha, case_sensitive) ->
	books = [
		osis: ["Ps"]
		apocrypha: true
		extra: "2"
		regexp: ///(\b)( # Don't match a preceding \d like usual because we only want to match a valid OSIS, which will never have a preceding digit.
			Ps151
			# Always follwed by ".1"; the regular Psalms parser can handle `Ps151` on its own.
			)(?=\.1)///g # Case-sensitive because we only want to match a valid OSIS.
	,
$BOOK_REGEXPS
	]
	# Short-circuit the look if we know we want all the books.
	return books if include_apocrypha is true and case_sensitive is "none"
	# Filter out books in the Apocrypha if we don't want them. `Array.map` isn't supported below IE9.
	out = []
	for book in books
		continue if include_apocrypha is false and book.apocrypha? and book.apocrypha is true
		if case_sensitive is "books"
			book.regexp = new RegExp book.regexp.source, "g"
		out.push book
	out

# Default to not using the Apocrypha
bcv_parser::regexps.books = bcv_parser::regexps.get_books false, "none"