Created
December 8, 2012 16:03
-
-
Save alxgsv/4240842 to your computer and use it in GitHub Desktop.
FineReader HTML post processing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env ruby | |
# encoding: utf-8 | |
require "rubygems" | |
require "nokogiri" | |
require "typograf" | |
content = File.open(ARGV[0]).read.force_encoding("cp1251").encode('UTF-8') | |
doc = Nokogiri::HTML(content) | |
result = "" | |
spans = doc.xpath("//span")#.map(&:inner_html) | |
merged_spans = [] | |
previous_dashed = false | |
spans.each do |s| | |
if previous_dashed | |
merged_spans[-1] += s.inner_html | |
elsif s.content =~ /^\p{Lower}/ | |
puts s | |
# если параграф начинается с маленькой буквы, присоединяем к предыдущему через пробел | |
merged_spans[-1] += " " + s.inner_html | |
else | |
merged_spans << s.inner_html | |
end | |
last_span = merged_spans[-1] | |
if last_span[-1] == "-" | |
# если параграф кончается переносом, убираем его и запоминаем это | |
merged_spans[-1] = last_span[0..-2] | |
previous_dashed = true | |
else | |
previous_dashed = false | |
end | |
end | |
merged_spans.each do |s| | |
typografed = Typograf.process(s) | |
result += typografed | |
end | |
File.open(ARGV[1], "w").write(result) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment