subtitleParser 0.3 → 0.4

raw patch · 9 files changed

+111/−11 lines, 9 filesPVP ok

version bump matches the API change (PVP)

API changes (from Hackage documentation)

Files

• CHANGELOG +14/−0
• Text/Subtitles/SRT.hs +21/−8
• Text/Subtitles/SRT/Datatypes.hs +7/−0
• Text/Subtitles/SUB/Datatypes.hs +2/−2
• subtitleParser.cabal +6/−1
• test/TestSRT.hs +24/−0
• test/TestSUB.hs +24/−0
• test/example.srt +9/−0
• test/example.sub +4/−0

@@ -0,0 +1,14 @@+0.3+  - Show instances for SRT now are simpler and more related on how is defined+    the ADT. this is a PARSER not a prettyPrinter nor a valid .srt producer+  - SUB files added support for optional flag as underline and bold+  - SRT's parseDialog now uses peekChar so attoparsec dependency is upgraded and+    the code is cleaner. As a bug we provide for now parseOnly' until is fixed+    upstream+0.2+  - Basic support for .sub files added+  - Separated Datatypes for .sub and .srt files+0.1.2+  - Now haddock documentation is build correctly+0.1+  - Initial release

@@ -50,7 +50,7 @@ -- "Text.Subtitles.SRT.Datatypes" --  -- >2--- >00:00:50,050 --> 00:00:52,217+-- >00:00:50,050 --> 00:00:52,217 X1:1 X2:2 Y1:1 Y2:2 -- >Drama here -- -- The whole Line is represented in the 'Line' ADT which constructors@@ -61,7 +61,10 @@ -- -- * The second one is called 'Range', which correspond to two separated 'Time'. ----- * The last one is the 'subs'. which is just Text and correspond to the third+-- * After the range is an optional field called Rectangle which says what+--   geometry should the text obey.+--+-- * The last one is the 'subs'. Which is just Text and correspond to the third --   constructor of 'Line'.  -- |Main Parser, gives you a list of all the Lines of the subtitle. It fails if@@ -73,7 +76,7 @@ -- corresponding Line representation parseSingleLine :: Parser Line parseSingleLine = -  Line <$> parseIndex <*> parseRange <*> parseDialog T.empty+  Line <$> parseIndex <*> parseRange <*> optionalGeometry <*> parseDialog T.empty  parseIndex :: Parser Int parseIndex = decimal <* eol@@ -93,6 +96,14 @@     arrowString :: Parser Text     arrowString = string (T.pack " --> ")  +{- the order X1 X2 Y1 Y2 seems to be enforced, so we can check only for order+ - instead of keywords -}+optionalGeometry :: Parser (Maybe Rectangle)+optionalGeometry = option Nothing (Just <$> rectangle <* eol)+  where rectangle = R <$> valueSpace <*> valueSpace <*> valueSpace <*> value+        value   = letter *> digit *> char ':' *> decimal +        valueSpace = value <* space+ parseTime :: Parser Time parseTime = Time <$> numDot <*> numDot <*> decimal <* char ',' <*> decimal   where@@ -102,14 +113,16 @@ {- return the dialog checking for newlines that could be in there. That why is  - written in a monad instead of applicative. More efficient version welcome -} parseDialog :: Text -> Parser Text-parseDialog t = do +parseDialog stateLine = do    line <- takeWhile1 (not . isEndOfLine)   endOfLine-  let lineState = T.append t (T.snoc line '\n') --takeWhile1 didn't consume \n+  let stateCurrent = T.append stateLine line+      lineState = T.snoc stateCurrent '\n' --takeWhile1 didn't consume \n   next <- peekChar   case next of-    Nothing     -> return lineState -    (Just '\n') -> eol >> return lineState -    (Just _)    -> parseDialog lineState +    Nothing     -> return stateCurrent  -- the end of the file+    (Just '\n') -> eol >> return stateCurrent -- End of this Line, new Line coming.+    (Just _)    -> parseDialog lineState {- in between lines, the next one belong to this Line+                                         explicit eol required -}  

@@ -17,6 +17,7 @@    -- * Datatypes   Subtitles,+  Rectangle(..),   Line(..),   Range(..),   Time(..)@@ -26,6 +27,11 @@ import Data.Text (Text, unpack) import Data.Attoparsec.Text (Parser) +-- |This represent the position on screen of the Line. Is usually optional in+-- the file.+data Rectangle = R { x1 :: Int, x2 :: Int, y1 :: Int, y2 :: Int}+  deriving (Eq, Ord, Show)+ data Time = Time   { hour    :: Int   , minutes :: Int@@ -43,6 +49,7 @@ data Line = Line   { index  :: Int   , range  :: Range+  , geometry :: Maybe Rectangle    , dialog   :: Text   } deriving (Eq, Ord, Show) 

@@ -38,11 +38,11 @@ -- |Optional property of text, the constructor should be clear. data TextProperty = Italic | Bold | UnderLine | Stroked                    | C Color -- ^ This is just text because is non-standard RGB-  deriving (Show)+  deriving (Show, Eq)  data Line = Line { startFrame :: Frame,                    finalFrame :: Frame,                    property   :: Maybe TextProperty,                     dialog     :: Text }-  deriving (Show)+  deriving (Show, Eq) 

@@ -1,5 +1,5 @@ name:            subtitleParser-version:         0.3+version:         0.4 license:         BSD3 license-file:    LICENSE category:        Text, Parsing@@ -17,6 +17,11 @@ extra-source-files:     LICENSE     THANKS+    CHANGELOG+    test/TestSUB.hs+    test/example.srt+    test/example.sub+    test/TestSRT.hs  Flag developer   Description: Whether to build the library in development mode

@@ -0,0 +1,24 @@+module Main where++import Test.HUnit+import Test.HUnit.Text+import Data.Text.IO as TI +import Data.Text (pack)+import Text.Subtitles.SRT++main :: IO ()+main = runTestTT (TestList [test srtAssert]) >> return ()++srtAssert :: Assertion+srtAssert = do+  srtContents <- TI.readFile "./test/example.srt"+  case parseOnly' parseSRT srtContents of+    Left _ -> assertFailure "Error while parsing the example .srt"+    Right r -> assertEqual "Parser didn't produce the expected value" expectedValue r++{- I wanted the parser to produce this output, when it accomplish it, it will be+ - complete -}+expectedValue :: Subtitles+expectedValue = [ Line 1 (Range (Time 0 2 26 407) (Time 0 2 31 356)) (Just (R 100 100 100 100)) (pack "<font color=\"#00ff00\">Detta handlar om min storebrors</font> \n<b><i><u>kriminella beteende och foersvinnade.</u></i></b>"),+  Line 2 (Range (Time 0 2 31 567) (Time 0 2 37 164)) Nothing (pack "Vi talar inte laengre om Wade. Det aer \nsom om han aldrig hade existerat.")]+

@@ -0,0 +1,24 @@+module Main where++import Test.HUnit+import Test.HUnit.Text+import Text.Subtitles.SUB+import Data.Text (pack)+import Data.Text.IO as TI (readFile)+import Data.Attoparsec.Text (parseOnly)++main :: IO ()+main = runTestTT (test assertSUB) >> return ()++expectedSUB :: Subtitles+expectedSUB = [Line 13187 13215 (Just Italic) (pack "Áno!"),+  Line 13430 13456 Nothing (pack "Áno!"),+  Line 13935 14020 (Just (C (pack "$0000ff"))) (pack "Nezdie¾am vašu rados..."),+  Line 14034 14147 Nothing (pack "- Dobrá práca, agentka Marcusová.|- Ïakujem.")]++assertSUB :: Assertion+assertSUB = do+  example <- TI.readFile "./test/example.sub"+  case parseOnly parseSUB example of+    Left _ -> assertFailure "parseSUB failed on example"+    Right r -> assertEqual "parseSUB result didn't match expected" expectedSUB r

@@ -0,0 +1,9 @@+1+00:02:26,407 --> 00:02:31,356  X1:100 X2:100 Y1:100 Y2:100+<font color="#00ff00">Detta handlar om min storebrors</font> +<b><i><u>kriminella beteende och foersvinnade.</u></i></b>++2+00:02:31,567 --> 00:02:37,164 +Vi talar inte laengre om Wade. Det aer +som om han aldrig hade existerat.

@@ -0,0 +1,4 @@+{13187}{13215}{y:i}Áno!+{13430}{13456}Áno!+{13935}{14020}{C:$0000ff}Nezdie¾am vašu rados...+{14034}{14147}- Dobrá práca, agentka Marcusová.|- Ïakujem.