html-conduit 1.2.0 → 1.2.1

raw patch · 4 files changed

+40/−8 lines, 4 filesPVP ok

version bump matches the API change (PVP)

API changes (from Hackage documentation)

+ Text.HTML.DOM: eventConduitText :: Monad m => Conduit Text m Event

+ Text.HTML.DOM: parseLT :: Text -> Document

+ Text.HTML.DOM: parseSTChunks :: [Text] -> Document

+ Text.HTML.DOM: sinkDocText :: MonadThrow m => Sink Text m Document

Files

ChangeLog.md +4/−0
Text/HTML/DOM.hs +29/−7
html-conduit.cabal +1/−1
test/main.hs +6/−0

ChangeLog.md view

@@ -1,3 +1,7 @@+## 1.2.1++* Add strict and lazy text parsing [#66](https://github.com/snoyberg/xml/pull/66)+ ## 1.2.0  * Drop system-filepath

Text/HTML/DOM.hs view

@@ -5,6 +5,10 @@     , readFile     , parseLBS     , parseBSChunks+    , eventConduitText+    , sinkDocText+    , parseLT+    , parseSTChunks     ) where  import Control.Monad.Trans.Resource@@ -15,10 +19,10 @@ import qualified Data.XML.Types as XT import Data.Conduit import Data.Conduit.Text (decodeUtf8Lenient)-import Data.Text (Text) import qualified Data.Text as T+import qualified Data.Text.Lazy as TL import qualified Data.Conduit.List as CL-import Control.Arrow ((***), second)+import Control.Arrow ((***)) import qualified Data.Set as Set import qualified Text.XML as X import Text.XML.Stream.Parse (decodeHtmlEntities)@@ -33,8 +37,14 @@ -- Note that there may be multiple (or not) root elements. @sinkDoc@ addresses -- that case. eventConduit :: Monad m => Conduit S.ByteString m XT.Event-eventConduit =-    decodeUtf8Lenient =$= TS.tokenStream =$= go []+eventConduit = decodeUtf8Lenient =$= eventConduit' ++eventConduitText :: Monad m => Conduit T.Text m XT.Event+eventConduitText = eventConduit'++eventConduit' :: Monad m => Conduit T.Text m XT.Event+eventConduit' = +    TS.tokenStream =$= go []   where     go stack = do         mx <- await@@ -92,8 +102,14 @@         ]  sinkDoc :: MonadThrow m => Sink S.ByteString m X.Document-sinkDoc =-    fmap stripDummy $ mapOutput ((,) Nothing) eventConduit =$ addDummyWrapper =$ X.fromEvents+sinkDoc = sinkDoc' eventConduit ++sinkDocText :: MonadThrow m => Sink T.Text m X.Document+sinkDocText = sinkDoc' eventConduitText++sinkDoc' :: (Monad m, MonadThrow m) => Conduit a m XT.Event -> Sink a m X.Document+sinkDoc' f = +    fmap stripDummy $ mapOutput ((,) Nothing) f =$ addDummyWrapper =$ X.fromEvents   where     addDummyWrapper = do         yield (Nothing, XT.EventBeginElement "html" [])@@ -115,5 +131,11 @@ parseLBS = parseBSChunks . L.toChunks  parseBSChunks :: [S.ByteString] -> X.Document-parseBSChunks bss = runIdentity $ runExceptionT_ $ CL.sourceList bss $$ sinkDoc+parseBSChunks tss = runIdentity $ runExceptionT_ $ CL.sourceList tss $$ sinkDoc++parseLT :: TL.Text -> X.Document+parseLT = parseSTChunks . TL.toChunks++parseSTChunks :: [T.Text] -> X.Document+parseSTChunks tss = runIdentity $ runExceptionT_ $ CL.sourceList tss $$ sinkDocText

html-conduit.cabal view

@@ -1,5 +1,5 @@ Name:                html-conduit-Version:             1.2.0+Version:             1.2.1 Synopsis:            Parse HTML documents using xml-conduit datatypes. Description:         This package uses tagstream-conduit for its parser. It automatically balances mismatched tags, so that there shouldn't be any parse failures. It does not handle a full HTML document rendering, such as adding missing html and head tags. Homepage:            https://github.com/snoyberg/xml

test/main.hs view

@@ -42,6 +42,12 @@         it "split code-points" $             X.parseLBS_ X.def "<foo>&#xa0;</foo>" @=?             H.parseBSChunks ["<foo>\xc2", "\xa0</foo>"]+        it "latin1 codes" $+            X.parseText_ X.def "<foo>\232</foo>" @=?+            H.parseSTChunks ["<foo>\232</foo>"]+        it "latin1 codes strict vs lazy" $+            H.parseLT "<foo>\232</foo>" @=?+            H.parseSTChunks ["<foo>\232</foo>"]     describe "HTML parsing" $ do         it "XHTML" $             let html = "<html><head><title>foo</title></head><body><p>Hello World</p></body></html>"