--- testing environment for sorters
dofile "polishsort.lua"

      document.whatever = { }
document.whatever.words = { }

local my = {}
function my.gsub (s, patt, repl)
  patt = lpeg.S(patt)
  patt = lpeg.Cs((patt / repl + 1)^0)
  return lpeg.match(patt, s)
end

--- based on http://www.mail-archive.com/ntg-context@ntg.nl/msg47525.html

function document.whatever.sorttext()
    local dwtext = document.whatever.text
    --local split = sorters.splitters.utf
    local split = sorters.splitters.utflower
    dwtext = my.gsub(dwtext, '\n\t\v"', " ")
    dwtext = string.explode(dwtext, " +")
    local dwwords = document.whatever.words
    for i=1, #dwtext do
        local current = string.strip(dwtext[i])
        if current ~= "" then
            table.insert(dwwords, { word = current })
        end
    end

    for i=1, #dwwords do
        local word = dwwords[i]
        word.split = split(word.word) 
    end
    --sorters.sort(dwwords, sorters.comparers.basic)
    sorters.sort(dwwords, sorters.comparers.polish)
end

function document.whatever.flushtext()
    local words = document.whatever.words
    local previous = false
    local p_word = false
    for i=1, #words do
        local word = words[i]
        local letter, current = sorters.firstofsplit(word)
        local letter = utf.lower(letter)
        if previous ~= current then
            previous = current
            context.section(letter)
        end

        local c_word = word.word

        if p_word ~= c_word then
            context(tostring(i) .. ": " .. c_word) context.par()
            p_word = c_word
        end
    end
end

function testrun (lang)
    --f = assert(io.open("anna-utf.txt", "r"))
    --f = assert(io.open("sltext.txt", "r"))
    document.whatever.text = [[
polskie słowa dziwnie się szereguje
Polskie Słowa Dziwnie Się Szereguje
]]

    sorters.setlanguage(lang)
    context.starttext()
    document.whatever.sorttext()
    document.whatever.flushtext()
    context.stoptext()
end

testrun("pl")