check-ligatures.lua

-- check-ligatures.lua
--[[ 
    analyze Ligatures based on a ConTeXt log file
    enable Ligature tracking in the ConTeXt file with
    \enabletrackers[languages.goodies]
    then run this script with
    mtxrun --script check-ligatures.lua input_file whitelist_file
    for the input_file we assume .log, so no need to add this
    for the whitelist_file the file ending must be specified
    results will be saved to input_file-ligatures-new.log
	if no whitelist is specified as a command line option
	the script falls back to input_file-ligatures-old.log
]] 

-- this is used only for testing purposes
-- local pprint = require('pprint')

--------------------------------------------------------------------
--------------------------------------------------------------------

function main (input_file, whitelist_file)
    local lines = string.splitlines(io.loaddata(input_file .. ".log")or "") or { }
	local whitelist = {}
    if whitelist_file == nil then
		whitelist = string.splitlines(io.loaddata(input_file .. '-ligatures-old.log')or "") or { }
	else
		whitelist = string.splitlines(io.loaddata(whitelist_file)or "") or { }
	end
    local filteredWordlist = 
        filterLigaturesWordlist
            (cleanLines
                (getLigatureLines(lines)), 
                    whitelist)
    -- sort the table in-place
    table.sort(filteredWordlist)
    saveResultsToFile(filteredWordlist, input_file .. '-ligatures-new.log')
end

----------------------------------------------------------------
----------------------------------------------------------------

-- String testing
function starts_with(str, start)
    return str:sub(1, #start) == start
end

-- get relevant lines
function getLigatureLines(lines)
    local result = {}
    for k,v in pairs(lines) do
        if 
            (starts_with(v, "languages       > goodies") 
            and not string.find(v, "goodies > properties:")
            and not string.find(v, "goodies > tag")
            -- goodies > tag
        )
        then table.insert(result, v) end
    end
    return result
end

-- String cleaning
-- wrapper functions

function cleanLines (xs)
    local result = {}
    for k,v in pairs(xs) do
        table.insert(result, cleanLine(v))
    end
    return result
end

function cleanLine (x)
    --return removeTrailingPunctuation(getWord(x)) -- leave out for now
    return getWord(x)
end

-- 1. Start reading at colon 
function getWord(x)
    -- wir lesen das Wort nach dem Doppelpunkt
    return string.match(x, ".*: (%w+)")
end

-- 2. Remove trailing punctuation
function removeTrailingPunctuation (x)
    if string.find(x, ',') then
        return x:sub(1, -2)
    else
        return x
    end
end

-- test if word is in list
function inList (x, list)
    for k,v in ipairs(list) do
        if v == x then
            return true
        end
    end
    return nil
end

-- Filter words based on second list (whitelist)
function filterLigaturesWordlist (xs, list)
    local result = {}
    for k,v in ipairs(xs) do
        if (hasMaybeLigature(v) and not inList(v, list) and not inList(v, result)) then table.insert(result, v) end
    end
    return result
end

-- keep only words with potential ligatures
function hasMaybeLigature (x)
    return 
        string.find(x, "fi")
        or string.find(x, "fl")
        or string.find(x, "ff") 
        -- or string.find(x, "ffi")
        -- or string.find(x, "ffl")
        or string.find(x, "ft")
        -- or string.find(x, "fft")
        or string.find(x, "fb")
        -- or string.find(x, "ffb")
        or string.find(x, "fh")
        -- or string.find(x, "ffh")
        or string.find(x, "fj")
        -- or string.find(x, "ffj")
        or string.find(x, "fk")
        -- or string.find(x, "ffk")
end

-- 
function saveResultsToFile(results, output_file)
    -- Opens a file in write mode
    output_file = io.open(output_file, "w")
    -- sets the default output file as output_file
    io.output(output_file)
    -- iterate oiver 
    for k,v in ipairs(results) do
        io.write(v..'\n')
    end
    -- closes the open file
    io.close(output_file)
end

-- Run
main(arg[1], arg[2])