#!/usr/bin/env ruby
# Usage: getheadings <pdffile> <delimited list of headings>

delimiter = ","

pdf         = ARGV.shift
headings    = ARGV.join("").split(delimiter)
raise "#{pdf} does not exist"    unless File.exist?(pdf)
raise "Failed to read #{pdf}."   unless File.stat(pdf).readable? && File.stat(pdf).file?
raise "#{pdf} is not a PDF file" unless /^\%PDF-[\d\.]+\s*$/i === File.new(pdf).readline

cmd         = "pdftohtml -q -stdout -noframes #{pdf}"
pages       = []
@returns    = {}
manual      = IO.popen(cmd).read.split("<hr>")[0..-3]

manual.each do |page|
	/^<A name=(\w+)><\/a><b>\w+\<\/b><br>\n(\w+)<br>$/.match(page)
	pages << [$1,$2]
end

headings.each do |heading|
	manual.each do |page|
		pnum  = pages[manual.index(page)]
		regex = /^<b>#{Regexp.escape(heading)}<\/b><br>$/
		@returns[heading] = "#{pnum[0]},#{pnum[1]}" if regex =~ page
	end
end

@returns.each { |r| puts r[1] }