Modul:Kleida-el

-- @brief
--  creates DEFAULTSORT for greek (and almost all ancient greek) lemmas
---- since greek lemmas in dictionaries are listed depending on one of 24 letters of alfabet 
---- independently from their accents
-- @param
--  none, uses PAGENAME from "frame"

-- ]

objkleida = {}
-- mapping dictionary
-- maps every greek letter (except the very archaic like digamma)
-- to one of the 24 anaccented letters of greek alphabet
objkleida._mappedletters = {
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "α",
          = "β",
          = "β",
          = "γ",
          = "γ",
          = "δ",
          = "δ",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ε",
          = "ζ",
          = "ζ",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "η",
          = "θ",
          = "θ",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "ι",
          = "κ",
          = "κ",
          = "λ",
          = "λ",
          = "μ",
          = "μ",
          = "ν",
          = "ν",
          = "ξ",
          = "ξ",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "ο",
          = "π",
          = "π",
          = "ρ",
          = "ρ",
          = "ρ",
          = "ρ",
          = "ρ",
          = "σ",
          = "σ",
          = "σ",
          = "τ",
          = "τ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "υ",
          = "φ",
          = "φ",
          = "χ",
          = "χ",
          = "ψ",
          = "ψ",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
          = "ω",
}

-- function _fmainkleida
--  (can also be used independently to sort lines in tables, etc)
-- Δεν ελέγχει την περίπτωση το πρώτο γράμμα να είναι παύλα και το δεύτερο κενό
-- Does not check if first letter is hyphent followed by space
-- LUA problem, if parameter (any_string) contains more than 32 characters
-- Except from the first character,
-- keeps only "mapped characters" 
-- (aka the 24 anaccented letters of greek alphabet)
function _fmainkleida ( any_string ) 
    local mystring = any_string
    local startingchar = mw.ustring.sub(mystring,1,1)
    local wordproduced = ''
    local mychar = ''
--[[    
-- if ending minus must be returned
--    local endingminus = false
--    if mw.ustring.sub(any_string, -1) == '-' then endingminus = true end
--]]
-- αν το πρώτο γράμμα είναι παύλα το αφαιρούμε
-- if first character is hyphen is removed
-- this way "endings" (like -λογία) will be sorted under their first character
    if startingchar == '-' then
        mystring = mw.ustring.sub(mystring,2)
    end
    -- αν υπήρχε μόνο η παύλα τότε επιστρέφει την αρχική παράμετρο
    --if only a hyphen existed then returns that hyphen
    if mystring == nil or mw.ustring.len(mystring) < 1 then
        return any_string 
    end
    startingchar = mw.ustring.sub(mystring,1,1)
--δοκίμασε να μετατρέψεις το πρώτο γράμμα
--"map" first letter
    convertedchar = objkleida._mappedletters    
    -- αν το πρώτο γράμμα δεν είναι ελληνικό κράτα το όπως ήταν
    -- if first letter was not a "greek" one
    -- keep it as is
    -- (why? maybe it is a punctuation mark or other mark)
    if convertedchar == nil or mw.ustring.len(convertedchar) < 1 then
         wordproduced = startingchar
    --αλλιώς βάλε στην αρχή αυτό που έφερε από τη μετατροπή
    --else replace it with the return value
    else
         wordproduced = convertedchar
    end
    -- αν έχει και άλλα
    -- if more characters exist...
    if mw.ustring.len(mystring) > 1 then
        --χρησιμοποίησε μόνο τα υπόλοιπα
        --use only the rest.
        mystring = mw.ustring.sub(mystring,2)        
        --για κάθε χαρακτήρα
        --For every character
        for codepoint in mw.ustring.gcodepoint( mystring ) do
            --αντικατάστησέ τον με βάση τον πίνακα
            --replace it according to mapping table
            mychar = mw.ustring.char(codepoint)
            convertedchar = objkleida._mappedletters            
            --αν βρήκε κάτι για αντικατάσταση
            --είναι δηλαδή ελληνικό γράμμα
            --if something returned
            --(aka was a greek letter, accented or not)
            if convertedchar ~= nil then
                --πρόσθεσέ το στη δημιουργούμενη κλείδα
                --add it to "defaultsort"
                wordproduced = wordproduced .. convertedchar
            end
            -- else is a non greek character a space, another hyphen etc
            -- do nothing with it
        end
    end
--    if endingminus then wordproduced = wordproduced .. '-' end
    return wordproduced
end

--function that will be used for DEFAULTSORT
--will use "frame" and current tile to invoke parameter
objkleida.kleida = function( frame ) 
    PAGENAME = mw.title.getCurrentTitle().text  
    --[[αφαίρεση των κενών πριν και μετά το λήμμα γιατί μάλλον τα στέλνει το πρότυπο
    και δεν χρειάζονται έτσι κι αλλιώς
    (συνάρτηση αντιγραμμένη "τυφλά", είναι σωστή;)]]
    local myarg1 = PAGENAME:match( "^%s*(.-)%s*$" )
    if mw.ustring.len(myarg1) > 32 then
        --δημιουργία κλείδας μόνο για τους πρώτους 32 χαρακτήρες        
        --LUE problems exist if more than 32 characters are used
        --But can be extended to support "chunks" of 32 characters
        --(which maybe is not so usefull in such big words)
        myarg1 =  mw.ustring.sub(myarg1,1,32)
    end
    --creates DEFAULTSORT entry for the first 32 characters
    return frame:preprocess(_fmainkleida(myarg1))
end
 
return objkleida