diff --git a/CHANGELOG.md b/CHANGELOG.md --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -37,3 +37,8 @@ * Fourth version. Fixed issue with the documentation for the 'show7s5' function. Added some new library functions. Added the possibility to get some statistic metrics and to obtain a list of lists of Strings -- the information about how many and what actually Ukrainian sounds representations are needed to be created if every sound representation is unique. + +## 0.5.0.0 -- 2020-02-26 + +* Fifth version. Added a function 'show7s9' and some descriptive statistics functions. For the first command line argument "-s" the program now prints some +descriptive statistics metrices for the given text. diff --git a/MMSyn7s.hs b/MMSyn7s.hs --- a/MMSyn7s.hs +++ b/MMSyn7s.hs @@ -26,13 +26,20 @@ , show7s6 , show7s7 , show7s8 + , show7s9 -- *** Inner predicate (auxiliary) , eqSnds -- *** Inner backward conversion function , listToString - -- *** Some statistic metrics + -- * Some descriptive statistics metrics , countSnds , sndsDensity + , uniquenessPeriods + , uniqMax + , uniqMin + , uniqPeriodsMean + , uniqPeriodsDispersion + , uniqStdQDeviation ) where import Data.Char (isSpace) @@ -50,8 +57,7 @@ -- -- \"-v\" -- prints version number and exits; -- --- \"-s\" -- prints a ratio of the total number of the Ukrainian sounds representations needed if every one is used just once --- (except silent) to the total number of the represenations if they are the same no matter where the corresponding sound is located; +-- \"-s\" -- prints some general descriptive statistics metrices for the given text; -- -- \"1\" -- prints the list of String being unique (without silence) and then the rest of the text with whitespaces and some phonetical conversions; -- @@ -75,17 +81,16 @@ ; putStrLn "SYNOPSYS: " ; putStrLn "mmsyn7s -h OR: " ; putStrLn "mmsyn7s -v OR: " - ; putStrLn "mmsyn7s -s OR: " + ; putStrLn "mmsyn7s -s {Ukrainian text} OR: " ; putStrLn "mmsyn7s 1 {Ukrainian text} OR: " ; putStrLn "mmsyn7s -1 {Ukrainian text} OR: " ; putStrLn "mmsyn7s 0 {Ukrainian text} OR: " ; putStrLn "mmsyn7s {Ukrainian text}" ; putStrLn "where the first one prints this help message; " ; putStrLn " the second one prints a version number; " - ; putStr " the third one prints a ratio of the total number of the Ukrainian sounds representations needed if every one is used just once " - ; putStrLn "(except silent) to the total number of the represenations if they are the same no matter where the corresponding sound is located; " - ; putStrLn " the \"1\" option prints the list of String being unique (without silence)" - ; putStr " the \"-1\" option prints the rest of the text after the first duplicated sound representation (except silent ones) " + ; putStrLn " the third one prints some general descriptive statistics metrices for the given text; " + ; putStrLn " the \"1\" option prints the list of String being unique (without silence);" + ; putStr " the \"-1\" option prints the rest of the text after the first duplicated sound representation (except silent ones); " ; putStrLn "including it with whitespaces and some phonetical conversions; " ; putStrLn " the \"0\" option prints the list of String for the whole text; " ; putStrLn " the \"2\" option prints the list of String for the whole text where every sound representation is unique; " @@ -93,8 +98,18 @@ ; putStrLn "created if every sound is unique; " ; putStrLn " the other beginning is equivalent to the previous one behaviour." } - "-v" -> putStrLn "mmsyn7s: version 0.4.0.0" - "-s" -> putStrLn . show . sndsDensity . unwords . drop 1 $ texts + "-v" -> putStrLn "mmsyn7s: version 0.5.0.0" + "-s" -> let ys = unwords . drop 1 $ texts in + do { putStrLn $ "Possibly unique sounds representations density: " ++ (show . sndsDensity $ ys) + ; putStrLn $ "Number of sounds representation that are enough to cover the text: " ++ (show . countSnds . show7s6 $ ys) + ; putStrLn $ "Possibly unique sounds representations density: " ++ (show . sndsDensity $ ys) + ; putStrLn $ "Uniqueness periods: " ++ (show . uniquenessPeriods $ ys) + ; putStrLn $ "Mean for uniqueness periods list: " ++ (show . uniqPeriodsMean $ ys) + ; putStrLn $ "Dispersion for uniqueness periods list: " ++ (show . uniqPeriodsDispersion $ ys) + ; putStrLn $ "Standard quadratic deviation for uniqueness periods list: " ++ (show . uniqStdQDeviation $ ys) + ; putStrLn $ "The maximum element in the uniqueness periods list: " ++ (show . uniqMax $ ys) + ; putStrLn $ "The minimum element in the uniqueness periods list: " ++ (show . uniqMin $ ys) + } "1" -> putStrLn . show . fst . show7s5 . unwords . drop 1 $ texts "-1" -> putStrLn . snd . show7s5 . unwords . drop 1 $ texts "0" -> putStrLn . show7s2 . unwords . drop 1 $ texts @@ -214,3 +229,54 @@ show7s8 :: String -> [String] show7s8 t@(_:_) = (fst . show7s7 $ t):(show7s8 . snd . show7s7 $ t) show7s8 _ = [] + +-- | Function 'show7s9' takes a Ukrainian text being a @String@ and returns a list of Strings. Each String is a Ukrainian sound representation of the duplicated +-- non-silent sounds, it begins a new second list of Strings in the 'show7s4' function. This information can be helpful e. g. in music and composing. +show7s9 :: String -> [String] +show7s9 xs@(_:_) = (if not . null . snd . show7s4 $ xs then head . snd . show7s4 $ xs else []):(show7s9 . snd . show7s5 $ xs) +show7s9 _ = [] + +-- | Function 'uniquenessPeriods' takes a Ukrainian text being a @String@ and returns a list of Ints. Each Int value is a number of +-- the Ukrainian sounds representations (non-silent ones) being unique and not duplicated alongside the given text starting from the beginning to the end. +-- This function provides some important information about the phonetical and in some cases semantical structures of the text. +uniquenessPeriods :: String -> [Int] +uniquenessPeriods xs | not . null . filter (not . isSpace) $ xs = map length . show7s6 $ xs + | otherwise = [0::Int] + +-- | Function 'uniqPeriodsMean' is a mathematical expectation for the list obtained by 'uniquenessPeriods' function. It is a statistic metrics. +-- It is a mean for the quantities of the unique (not duplicated, not repeated) Ukrainian sounds in the given Ukrainian text as a @String@. +-- If there are no Ukrainian letters in the text, it is equal to 0.0. The greater it is, the more diverse (phonetically) is the text. +uniqPeriodsMean :: String -> Double +uniqPeriodsMean xs | not . null . filter (not . isSpace) $ xs = let ys = uniquenessPeriods xs in (fromIntegral . sum $ ys) / (fromIntegral . length $ ys) + | otherwise = 0.0 + +-- | Function 'uniqPeriodsDispersion' is a dispersion for the list obtained by 'uniquenessPeriods' function. It is a statistic metrics. +-- It is a dispersion for the quantities of the unique (not duplicated, not repeated) Ukrainian sounds in the given Ukrainian text as a @String@. +-- If there are no Ukrainian letters in the text, it is equal to 0.0. The greater it is, the more suitable for changing pronunciation for the sounds +-- (and may be for intonation changes, too) the text is. +uniqPeriodsDispersion :: String -> Double +uniqPeriodsDispersion xs | not . null . filter (not . isSpace) $ xs = + let ys = uniquenessPeriods xs + z = uniqPeriodsMean xs + l = length ys in + (sum . map (\t -> (fromIntegral t - z) * (fromIntegral t - z)) $ ys) / fromIntegral l + | otherwise = 0.0 + +-- | Function 'uniqStdQDeviation' is a standard quadratic deviation for the list obtained by 'uniquenessPeriods' function. It is a statistic metrics. +-- It is a standard quadratic deviation for the quantities of the unique (not duplicated, not repeated) Ukrainian sounds in the given Ukrainian text as a @String@. +-- If there are no Ukrainian letters in the text, it is equal to 0.0. The greater it is, the more suitable for changing pronunciation for the sounds +-- (and may be for intonation changes, too) the text is. +uniqStdQDeviation :: String -> Double +uniqStdQDeviation xs = sqrt . uniqPeriodsDispersion $ xs + +-- | Function 'uniqMax' is a maximum element in the 'uniquenessPeriods' function for the same argument. It is provided as a standard element for the +-- descriptive statistics. +uniqMax :: String -> Int +uniqMax xs | not . null . filter (not . isSpace) $ xs = maximum . uniquenessPeriods $ xs + | otherwise = 0::Int + +-- | Function 'uniqMin' is a minimum element in the 'uniquenessPeriods' function for the same argument. It is provided as a standard element for the +-- descriptive statistics. +uniqMin :: String -> Int +uniqMin xs | not . null . filter (not . isSpace) $ xs = minimum . uniquenessPeriods $ xs + | otherwise = 0::Int diff --git a/README.markdown b/README.markdown --- a/README.markdown +++ b/README.markdown @@ -11,10 +11,7 @@ "-h" -- prints help and exits; -"-s" -- prints a ratio of the total number of the Ukrainian sounds -representations needed if every one is used just once (except silent) -to the total number of the represenations if they are the same no matter -where the corresponding sound is located; +"-s" -- prints some general descriptive statistics metrices for the given text; "-v" -- prints version number and exits; diff --git a/mmsyn7s.cabal b/mmsyn7s.cabal --- a/mmsyn7s.cabal +++ b/mmsyn7s.cabal @@ -2,7 +2,7 @@ -- documentation, see http://haskell.org/cabal/users-guide/ name: mmsyn7s -version: 0.4.0.0 +version: 0.5.0.0 synopsis: Shows a sorted list of the Ukrainian sounds representations that can be used by mmsyn7 series of programs description: A program and a library that show a sorted list of the Ukrainian sounds representations that can be used by mmsyn7 series of programs homepage: https://hackage.haskell.org/package/mmsyn7s