diff --git a/CHANGES.md b/CHANGES.md --- a/CHANGES.md +++ b/CHANGES.md @@ -1,6 +1,18 @@ Seonbi changelog ================ +Version 0.1.1 +------------- + +Released on October 7, 2019. + + - Added the `embed-dictionary` flag to the Cabal package. + - Fixed a bug that *Standard Korean Language Dictionary* had not been + included in executables if `flag(static)` is turned on. [[#1]] + +[#1]: https://github.com/dahlia/seonbi/issues/1 + + Version 0.1.0 ------------- diff --git a/README.md b/README.md --- a/README.md +++ b/README.md @@ -3,6 +3,8 @@ [![][releases-badge]][releases] [![][hackage-badge]][hackage] [![][dockerhub-badge]][dockerhub] [![][ci-status-badge]][ci] +(TL;DR: See the [demo web app].) + Seonbi (선비) is is an HTML preprocessor that makes typographic adjustments to an HTML so that the result uses accurate punctuations according to the modern Korean orthography. @@ -45,6 +47,7 @@ [dockerhub-badge]: https://img.shields.io/microbadger/image-size/dahlia/seonbi [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/ [SmartyPants]: https://daringfireball.net/projects/smartypants/ [Korean mixed script]: https://en.wikipedia.org/wiki/Korean_mixed_script @@ -121,7 +124,7 @@ HTTP/1.1 200 OK Content-Type: application/json - Server: Seonbi/0.1.0 + Server: Seonbi/0.1.1 { "success": true, @@ -133,7 +136,18 @@ seonbi-api -o https://example.com +To learn more parameters interactively, try the [demo web app]. + [CORS]: https://developer.mozilla.org/en-US/docs/Web/HTTP/CORS + + +Haskell API +----------- + +All functions and types lie inside `Text.Seonbi` module and its submodules. +The highest-level API is `Text.Seonbi.Facade` module. + +See also the API docs on [Hackage]. License diff --git a/seonbi.cabal b/seonbi.cabal --- a/seonbi.cabal +++ b/seonbi.cabal @@ -4,10 +4,10 @@ -- -- see: https://github.com/sol/hpack -- --- hash: d4a46f2e66dbb5471c1056a03c8d90108abbc383e058d391e679715309b07a2c +-- hash: 8339f3a5d2f79ad22fe631feda924f87bd347414e314cf9e312d062f4c481d92 name: seonbi -version: 0.1.0 +version: 0.1.1 synopsis: SmartyPants for Korean language description: Please see the README.md on GitHub at . category: Text @@ -43,6 +43,11 @@ , text , zip >=1.1 && <1.3 +flag embed-dictionary + description: Embed dictionary rather than load from file + manual: True + default: False + flag iconv description: Use iconv manual: True @@ -92,6 +97,10 @@ ghc-options: -Wall -fprint-explicit-kinds -optP-Wno-nonportable-include-path else ghc-options: -Wall -fprint-explicit-kinds + if flag(static) || flag(embed-dictionary) + cpp-options: -DEMBED_DICTIONARY + else + cpp-options: -DNO_EMBED_DICTIONARY default-language: Haskell2010 executable seonbi diff --git a/src/Text/Seonbi/Facade.hs b/src/Text/Seonbi/Facade.hs --- a/src/Text/Seonbi/Facade.hs +++ b/src/Text/Seonbi/Facade.hs @@ -1,7 +1,11 @@ +{-# LANGUAGE CPP #-} {-# LANGUAGE DeriveGeneric #-} {-# LANGUAGE NamedFieldPuns #-} {-# LANGUAGE OverloadedLists #-} {-# LANGUAGE OverloadedStrings #-} +#ifdef EMBED_DICTIONARY +{-# LANGUAGE TemplateHaskell #-} +#endif -- | Provides higher-level APIs. Read 'transformHtmlText' function first, -- and then see also 'Configuration' type. module Text.Seonbi.Facade @@ -32,13 +36,23 @@ import Data.ByteString.Lazy import Data.Csv +#ifdef EMBED_DICTIONARY +import Data.FileEmbed +#endif import Data.Map.Strict import Data.Set import Data.Text import qualified Data.Text.Lazy as LT -import System.FilePath (()) +import System.FilePath + ( () +#ifdef EMBED_DICTIONARY + , takeDirectory +#endif + ) +#ifndef EMBED_DICTIONARY import Paths_seonbi (getDataDir) +#endif import Text.Seonbi.Hanja import Text.Seonbi.Html import Text.Seonbi.Punctuation @@ -49,7 +63,7 @@ -- -- - 'ko_KR' -- - 'ko_KP' -data Monad m => Configuration m a = Configuration +data Monad m => Configuration m a = Configuration { -- | An optional debugging logger to print its internal AST. debugLogger :: Maybe (HtmlEntity -> m a) -- | Whether to take and result in XHTML instead of HTML. @@ -91,10 +105,10 @@ = CurvedQuotes -- | East Asian guillemets (@〈@: U+3008, @〉@: U+3009, @《@: U+300A, @》@: -- U+300B), which are used by North Korean orthography. - | Guillemets + | Guillemets -- | Use English-style curved quotes (@‘@: U+2018, @’@: U+2019) for single -- quotes, and HTML @\@ tags for double quotes. - | CurvedSingleQuotesWithQ + | CurvedSingleQuotesWithQ deriving (Enum, Eq, Generic, Read, Show) -- | An option to transform folk-citing quotes (e.g., @\<\<한겨레\>\>@) into @@ -300,12 +314,20 @@ readDictionaryFile :: FilePath -> IO HanjaDictionary readDictionaryFile path = do byteString <- Data.ByteString.Lazy.readFile path + case readDictionaryByteString byteString of + Right dic -> return dic + Left err -> fail err + +-- | Reads a dictionary from TSV bytes. +readDictionaryByteString :: Data.ByteString.Lazy.ByteString + -> Either String HanjaDictionary +readDictionaryByteString byteString = case decodeWith tsvDecodeOptions NoHeader byteString of - Right vector -> return $ Prelude.foldl + Right vector -> Right $ Prelude.foldl (\ d (DictionaryPair k v) -> Trie.insert k v d) Trie.empty (GHC.Exts.toList vector) - Left err -> fail err + Left err -> Left err where tsvDecodeOptions :: DecodeOptions tsvDecodeOptions = defaultDecodeOptions @@ -324,9 +346,22 @@ -- | Loads [Standard Korean Language Dictionary](https://stdict.korean.go.kr/) -- (標準國語大辭典) data. southKoreanDictionary :: IO HanjaDictionary +#ifdef EMBED_DICTIONARY +southKoreanDictionary = + case readDictionaryByteString bytes of + Right dic -> return dic + Left err -> fail err + where + bytes :: Data.ByteString.Lazy.ByteString + bytes = Data.ByteString.Lazy.fromStrict $ $(embedFile $ + takeDirectory __FILE__ ".." ".." ".." "data" + "ko-kr-stdict.tsv") + +#else southKoreanDictionary = do dataDir <- getDataDir readDictionaryFile (dataDir "ko-kr-stdict.tsv") +#endif data DictionaryPair = DictionaryPair !Text !Text deriving (Generic, Show)