Class: Ovec::Tier

Inherits:
TextManipulator show all
Defined in:
lib/ovec/tier.rb

Constant Summary collapse

REGEX =

The last character this regex matches is changed to a tilde.

/(
	((\p{Z}|[~\n(\[\{]|\A)[KkSsVvZzOoUu](\p{Z}|\n|\Z))|   # KSVZOU jako samostatne slovo
	((([\.\?\!](\p{Z}|\~)+)|(\A\p{Z}*))[KSVZOUAI](\p{Z}|\n))| # KSVZOUAI na zacatku vety
	(\p{Z}(?=--(\p{Z}|\n)))|              # mezera, za kterou je pomlcka
	(,(\p{Z}|\~|\n)+a(\p{Z}|\n))               # ... modulo 10, a~timto prvkem ...; TODO: plati tohle i pro "i"?
)/x
DATE_REGEX =

TODO: generally tie “5.~batalion”, … All changes within this regex are changed to a tilde.

/(
	(?<=\p{Z}|\A)\p{Nd}{1,2}\.\p{Z}
	(\p{Nd}{1,2}\.|leden|únor|březen|duben|květen|červen|červenec|srpen|září|říjen|listopad|prosinec| # TODO: plne sklonovani? nebo nejaky wildcard?
		ledna|února|března|dubna|května|června|července|srpna|září|října|listopadu|prosince)\p{Z}
	\p{Nd}{4}(?=(\p{Z}|[.,?!]|\Z)) # Datum jako "1. 5. 2013"
)/x

Instance Method Summary collapse

Methods inherited from TextManipulator

#bind

Instance Method Details

#run ⇒ Object 



22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
 
# File 'lib/ovec/tier.rb', line 22

def run
	# ~ neni mezi \p{Z}.
	# TODO: moznosti na dvojpismenne predlozky
	matches = @joined.to_enum(:scan, REGEX).map { Regexp.last_match }

	# Matches may overlap, find all matches with more scans.
	# TODO: optimize
	until matches.empty?
		for i in 0...matches.length
			# TODO: check ze za tim neni tilda nikde (napr. na dalsim radku, par mezer pozdeji, ...)
			match = matches[i]
			change = match.end(0) - 1
			chunk, offset = _find_chunk_and_offset(change)

			former_character = chunk[offset]

			chunk[offset] = '~'

			if former_character == '\n'
				# If we changed a newline to a tilde, change previous space to a
				# newline -- move the tied word to the other line.
				j = change - 1
				while j >= 0
					break if @joined[j] == '\n' # Don't cross newlines.

					if @joined[j] == ' '
						chunk, offset = _find_chunk_and_offset(j)
						chunk[offset] = '\n'
						break
					end
					j -= 1
				end
			end
		end
	
		_rejoin
		matches = @joined.to_enum(:scan, REGEX).map { Regexp.last_match }
	end

	# Dates can't overlap. 1 scan is enough.
	matches = @joined.to_enum(:scan, DATE_REGEX).map { Regexp.last_match }
	for match in matches
		for i in (match.begin(0))...(match.end(0))
			if @joined[i] == ' '
				chunk, offset = _find_chunk_and_offset(i)
				chunk[offset] = '~'
			end
		end
	end
end