diff --git a/CHANGES.md b/CHANGES.md --- a/CHANGES.md +++ b/CHANGES.md @@ -1,6 +1,17 @@ Seonbi changelog ================ +Version 0.2.2 +------------- + +Released on September 25, 2021. + + - Fixed stops normalizer's bug where unnecessary trailing spaces following + stops had been left even after normalized. + - Fixed stops normalizer's bug where commas followed by tag boundaries had + been not normalized. + + Version 0.2.1 ------------- diff --git a/README.md b/README.md --- a/README.md +++ b/README.md @@ -48,7 +48,7 @@ [hackage]: https://hackage.haskell.org/package/seonbi [hackage-badge]: https://img.shields.io/hackage/v/seonbi [dockerhub]: https://hub.docker.com/r/dahlia/seonbi -[dockerhub-badge]: https://img.shields.io/microbadger/image-size/dahlia/seonbi +[dockerhub-badge]: https://img.shields.io/docker/v/dahlia/seonbi?label=docker%20image&sort=semver [ci]: https://github.com/dahlia/seonbi/actions [ci-status-badge]: https://github.com/dahlia/seonbi/workflows/build/badge.svg [demo web app]: https://dahlia.github.io/seonbi/ @@ -128,7 +128,7 @@ HTTP/1.1 200 OK Content-Type: application/json - Server: Seonbi/0.2.1 + Server: Seonbi/0.2.2 { "success": true, diff --git a/seonbi.cabal b/seonbi.cabal --- a/seonbi.cabal +++ b/seonbi.cabal @@ -5,7 +5,7 @@ -- see: https://github.com/sol/hpack name: seonbi -version: 0.2.1 +version: 0.2.2 synopsis: SmartyPants for Korean language description: Please see the README.md on GitHub at . category: Text @@ -234,7 +234,8 @@ test ghc-options: -Wno-incomplete-uni-patterns -threaded -rtsopts -with-rtsopts=-N build-depends: - aeson >=1.3.1 && <1.5 + Diff >=0.3.4 && <0.4 + , aeson >=1.3.1 && <1.5 , base >=4.7 && <5 , bytestring , containers diff --git a/src/Text/Seonbi/Punctuation.hs b/src/Text/Seonbi/Punctuation.hs --- a/src/Text/Seonbi/Punctuation.hs +++ b/src/Text/Seonbi/Punctuation.hs @@ -36,6 +36,7 @@ import Prelude hiding (takeWhile) +import Control.Applicative ((<|>)) import Control.Monad import Data.Char (isSpace) import Data.Either @@ -315,7 +316,7 @@ , Data.Text.singleton <$> anyChar ] endOfInput - return $ Data.Text.concat chunks + return $ Data.Text.dropWhileEnd (' ' ==) $ Data.Text.concat chunks stops' :: Parser Text stops' = choice [ period' >> return (toEntity $ period stops) @@ -340,7 +341,7 @@ comma' :: Parser () comma' = void $ choice [ char '、' >> return "" - , string ", " + , string "," >> (takeWhile isSpace <|> (endOfInput >> return "")) , string ", " , string ", " , string "、" diff --git a/test/Text/Seonbi/FacadeSpec.hs b/test/Text/Seonbi/FacadeSpec.hs --- a/test/Text/Seonbi/FacadeSpec.hs +++ b/test/Text/Seonbi/FacadeSpec.hs @@ -4,6 +4,8 @@ import Control.Monad +import Data.Algorithm.Diff +import Data.Text.Lazy import Data.Text.Lazy.IO import System.Directory import System.FilePath @@ -23,6 +25,27 @@ , (".ko-KP.html", ko_KP) ] +shouldHaveSameText :: (HasCallStack) => Text -> Text -> Expectation +actual `shouldHaveSameText` expected = + unless (actual == expected) (expectationFailure msg) + where + expectedLines :: [Text] + expectedLines = Data.Text.Lazy.lines expected + actualLines :: [Text] + actualLines = Data.Text.Lazy.lines actual + diffLines :: [Diff Text] + diffLines = getDiff expectedLines actualLines + diff :: Text + diff = Data.Text.Lazy.unlines + [ case d of + First line -> "- " <> line + Second line -> "+ " <> line + Both line _ -> " " <> line + | d <- diffLines + ] + msg :: String + msg = "Two values are not equal:\n\n--- expected\n+++ actual\n\n" ++ + unpack diff spec :: Spec spec = do @@ -39,10 +62,13 @@ inputData <- Data.Text.Lazy.IO.readFile (dataDirPath input) outputData <- Data.Text.Lazy.IO.readFile (dataDirPath output) return (input, output, inputData, outputData, cfg) - forM_ testData $ \ (iname, oname, input, output, cfg) -> - specify ("transformHtmlLazyText: " ++ iname ++ " -> " ++ oname) $ do - transformHtmlLazyText noOp input `shouldBe` Just input - transformHtmlLazyText cfg input `shouldBe` Just output + describe "transformHtmlLazyText" $ + forM_ testData $ \ (iname, oname, input, output, cfg) -> + specify (iname ++ " -> " ++ oname) $ do + let Just noOpResult = transformHtmlLazyText noOp input + noOpResult `shouldHaveSameText` input + let Just cfgResult = transformHtmlLazyText cfg input + cfgResult `shouldHaveSameText` output where noOp :: Monad m => Configuration m a noOp = Configuration diff --git a/test/Text/Seonbi/PunctuationSpec.hs b/test/Text/Seonbi/PunctuationSpec.hs --- a/test/Text/Seonbi/PunctuationSpec.hs +++ b/test/Text/Seonbi/PunctuationSpec.hs @@ -498,7 +498,7 @@ normalizeStops horizontalStops input `shouldBe` [ HtmlStartTag [] P "" , HtmlText [P] - "봄·여름·가을·겨울. 어제, 오늘. " + "봄·여름·가을·겨울. 어제, 오늘." , HtmlEndTag [] P ] normalizeStops verticalStops input `shouldBe` @@ -509,6 +509,27 @@ ] normalizeStops horizontalStopsWithSlashes input `shouldBe` [ HtmlStartTag [] P "" - , HtmlText [P] "봄/여름/가을/겨울. 어제, 오늘. " + , HtmlText [P] "봄/여름/가을/겨울. 어제, 오늘." + , HtmlEndTag [] P + ] + it "normalizes stops followed by boundaries as well" $ do + let input = + [ HtmlStartTag [] P "" + , HtmlStartTag [P] Span "" + , HtmlText [P, Span] "쉼표," + , HtmlEndTag [P] Span + , HtmlStartTag [P] Span "" + , HtmlText [P, Span] "마침표." + , HtmlEndTag [P] Span + , HtmlEndTag [] P + ] :: [HtmlEntity] + normalizeStops verticalStops input `shouldBe` + [ HtmlStartTag [] P "" + , HtmlStartTag [P] Span "" + , HtmlText [P, Span] "쉼표、" + , HtmlEndTag [P] Span + , HtmlStartTag [P] Span "" + , HtmlText [P, Span] "마침표。" + , HtmlEndTag [P] Span , HtmlEndTag [] P ]