diff --git a/src/Text/XML/Light/Extractors.hs b/src/Text/XML/Light/Extractors.hs --- a/src/Text/XML/Light/Extractors.hs +++ b/src/Text/XML/Light/Extractors.hs @@ -21,17 +21,19 @@ -- -- And a data type for a book: -- --- > data Book = Book { bookdId :: Int +-- > data Book = Book { bookId :: Int -- > , isbn :: Maybe String -- > , author, title :: String -- > } -- -- You can parse the xml file into a generic tree structure using --- 'Text.XML.Light.Input.parseXMLDoc', then extract information from --- the tree using this library. +-- 'Text.XML.Light.Input.parseXMLDoc' from the `xml` package. -- +-- Using this library one can define extractors to extract data from +-- the generic tree. +-- -- @ --- library = 'element' "library" $ 'children' $ 'many' book +-- library = 'element' "library" $ 'children' $ 'only' $ 'many' book -- -- book = 'element' "book" $ do -- i <- 'attribAs' "id" 'Text.XML.Light.Extractors.Extra.integer' @@ -45,8 +47,16 @@ -- extractLibrary = 'extractDocContents' library -- @ -- --- /Note:/ The 'Control.Applicative' module contains some useful --- combinators like 'optional', 'many' and '<|>'. +-- = Notes +-- +-- * The "Control.Applicative" module contains some useful +-- combinators like 'optional', 'many' and '<|>'. +-- +-- * The "Text.XML.Light.Extractors.ShowErr" contains some +-- predefined functions to convert error values to strings. +-- +-- * The "Text.XML.Light.Extractors.Extra" module provides some +-- functions to read numeric data. -- module Text.XML.Light.Extractors ( diff --git a/src/Text/XML/Light/Extractors/Extra.hs b/src/Text/XML/Light/Extractors/Extra.hs --- a/src/Text/XML/Light/Extractors/Extra.hs +++ b/src/Text/XML/Light/Extractors/Extra.hs @@ -1,6 +1,5 @@ module Text.XML.Light.Extractors.Extra where - import Safe (readMay) import Text.XML.Light.Extractors diff --git a/src/Text/XML/Light/Extractors/Internal.hs b/src/Text/XML/Light/Extractors/Internal.hs --- a/src/Text/XML/Light/Extractors/Internal.hs +++ b/src/Text/XML/Light/Extractors/Internal.hs @@ -23,14 +23,10 @@ ) where -import Control.Applicative - import Control.Monad.Identity import Control.Monad.Trans.Reader import Control.Monad.Trans.State -import Data.Monoid - import Text.XML.Light.Types as XML import qualified Text.XML.Light.Proc as XML @@ -39,8 +35,10 @@ -------------------------------------------------------------------------------- +elemName :: Element -> String elemName = qName . elName +qname :: String -> QName qname name = QName name Nothing Nothing -------------------------------------------------------------------------------- @@ -53,8 +51,6 @@ addIdx i p = show i : p addElem :: XML.Element -> Path -> Path - - addElem e p = elemName e : p addAttrib :: String -> Path -> Path @@ -99,7 +95,7 @@ type ElementExtractor a = ReaderT (Path, XML.Element) (ResultT ExtractionErr Identity) a runElementExtractor :: ElementExtractor a -> XML.Element -> Path -> Result ExtractionErr a -runElementExtractor p elem path = runIdentity $ runResultT $ runReaderT p (path,elem) +runElementExtractor p elem path = runIdentity $ runResultT $ runReaderT p (path, elem) makeElementExtractor :: Result ExtractionErr a -> ElementExtractor a makeElementExtractor (Fatal e) = throwFatal e @@ -142,7 +138,7 @@ -- | Lift a string function to an element extractor. liftToElement :: (String -> Either Err a) -> String -> ElementExtractor a liftToElement f s = do - (path,x) <- ask + (path,_) <- ask case f s of Left e -> throwError (ExtractionErr e path) Right a -> return a @@ -184,9 +180,9 @@ element :: String -> ElementExtractor a -> ContentsExtractor a element name p = first expect go where - go c@(Elem x) path + go (Elem x) path | elemName x == name = escalate $ runElementExtractor p x (addElem x path) - go c path = Fail (ExtractionErr (ErrExpect expect c) path) + go c path = Fail (ExtractionErr (ErrExpect expect c) path) expect = "element " ++ show name diff --git a/src/Text/XML/Light/Extractors/Internal/Result.hs b/src/Text/XML/Light/Extractors/Internal/Result.hs --- a/src/Text/XML/Light/Extractors/Internal/Result.hs +++ b/src/Text/XML/Light/Extractors/Internal/Result.hs @@ -69,6 +69,7 @@ -- | Maps 'Fail' and 'Fatal' to 'Left'. +toEither :: Result a b -> Either a b toEither (Fatal e) = Left e toEither (Fail e) = Left e toEither (Ok a) = Right a @@ -145,13 +146,13 @@ -------------------------------------------------------------------------------- -testX :: ResultT String IO Int -testX = lift (print "x") >> return 1 +-- testX :: ResultT String IO Int +-- testX = lift (print "x") >> return 1 -testY :: ResultT String IO Int -testY = lift (print "error") >> throwError "error" +-- testY :: ResultT String IO Int +-- testY = lift (print "error") >> throwError "error" -testZ :: ResultT String IO Int -testZ = lift (print "fatal") >> throwFatal "fatal" +-- testZ :: ResultT String IO Int +-- testZ = lift (print "fatal") >> throwFatal "fatal" -------------------------------------------------------------------------------- diff --git a/src/Text/XML/Light/Extractors/ShowErr.hs b/src/Text/XML/Light/Extractors/ShowErr.hs --- a/src/Text/XML/Light/Extractors/ShowErr.hs +++ b/src/Text/XML/Light/Extractors/ShowErr.hs @@ -1,4 +1,10 @@ -module Text.XML.Light.Extractors.ShowErr where +-- | This module provide default functions to translate errors into strings. +module Text.XML.Light.Extractors.ShowErr + ( showExtractionErr + , showErr + , showPath + ) +where import Data.List @@ -8,22 +14,38 @@ showExtractionErr :: ExtractionErr -> String -showExtractionErr (ExtractionErr e path) = showErr e ++ "\nin " ++ showPath path +showExtractionErr (ExtractionErr e path) = showErr e ++ "in path: " ++ showPath path showPath :: Path -> String showPath = intercalate "/" . reverse +showErr :: Err -> String showErr (ErrExpect expect found) = - unwords ["Expected", expect, "found", take 60 $ XML.showContent found] - + unlines + [ unwords ["Expected", expect, "found"] + , take 60 $ XML.showContent found + , unwords ["at line:", showLine found] + ] showErr (ErrAttr expect parent) = - unwords ["Missing attribute", show expect, "of", qName $ elName parent] - -showErr (ErrMsg msg) = msg + unlines + [ unwords ["Missing attribute", show expect, "of", show $ qName $ elName parent] + , unwords ["at line:", showLine $ Elem parent] + ] +showErr (ErrMsg msg) = + unlines [msg] showErr (ErrNull expected) = - unwords ["Expected", expected] + unlines [unwords ["Expected", expected]] showErr (ErrEnd found) = - unwords ["Unexpected:", take 60 $ XML.showContent found] + unlines [unwords ["Unexpected:", take 60 $ XML.showContent found]] + + +contentLine :: Content -> Maybe Integer +contentLine (Elem e) = elLine e +contentLine (Text t) = cdLine t +contentLine _ = Nothing + +showLine :: Content -> String +showLine = maybe "??" show . contentLine diff --git a/xml-extractors.cabal b/xml-extractors.cabal --- a/xml-extractors.cabal +++ b/xml-extractors.cabal @@ -1,26 +1,27 @@ name: xml-extractors -version: 0.2.0.0 -synopsis: Simple wrapper over xml (Text.XML.Light) to extract data from parsed xml +version: 0.2.1.0 +synopsis: Simple wrapper over xml to extract data from parsed xml description: - This is a library to make it easier to extract data from parsed xml. + This is a library to simplify extraction of data from parsed xml. . See the 'Text.XML.Light.Extractors' module for an example. . = Motivation . - The xml package provides functions to parse and get information - from xml data. You can parse an xml string into a generic xml tree - representation. Then to extract information from that tree and into - you own data types you can use this library. + The `xml` package provides functions to parse and get information from + xml data. It will parse an xml string into a generic tree + representation. Extracting information from such a tree while + keeping track of location to handle errors is tricky. This library + helps with that. . If there is an error during extraction (expected information is - absent or wrong), you will get an error value with position information. + absent or wrong), it will return an error value with position information. . = Some limitations . - * Only handles unqualified names, so far. + * Only handles unqualified names. . - * No column or line number reference in error values. + * No column number and sometimes no line number reference in error values. homepage: https://github.com/holmisen/xml-extractors license: BSD3 @@ -28,7 +29,8 @@ author: Johan Holmquist maintainer: holmisen@gmail.com -- copyright: -category: Text, XML +category: XML +tested-with: GHC == 7.8.3 build-type: Simple extra-source-files: README.md cabal-version: >=1.10 @@ -39,7 +41,7 @@ Text.XML.Light.Extractors.Internal.Result, Text.XML.Light.Extractors.Extra, Text.XML.Light.Extractors.ShowErr - -- other-modules: + -- other-modules: other-extensions: NoMonomorphismRestriction build-depends: base >=4.6 && <4.8, xml >=1.3 && <1.4,