diff --git a/hxt.cabal b/hxt.cabal --- a/hxt.cabal +++ b/hxt.cabal @@ -1,6 +1,6 @@ -- arch-tag: Haskell XML Toolbox main description file Name: hxt -Version: 9.3.1.0 +Version: 9.3.1.1 Synopsis: A collection of tools for processing XML with Haskell. Description: The Haskell XML Toolbox bases on the ideas of HaXml and HXML, but introduces a more general approach for processing XML with Haskell. @@ -16,6 +16,8 @@ hxt-unicode contains encoding and decoding functions, hxt-charproperties char properties for unicode and XML. . + Changes from 9.3.1.0: in readString all input decoding switched off + . Changes from 9.3.0.1: lower bound for network set to be >= 2.4 . Changes from 9.3.0: upper bound for network set to be < 2.4 (URI signatures changed in 2.4) @@ -34,7 +36,7 @@ Homepage: http://www.fh-wedel.de/~si/HXmlToolbox/index.html Copyright: Copyright (c) 2005-2012 Uwe Schmidt Build-type: Simple -Cabal-version: >=1.6 +Cabal-version: >=1.8 extra-source-files: examples/arrows/absurls/AbsURIs.hs @@ -189,3 +191,7 @@ hxt-charproperties >= 9.1 && < 10, hxt-unicode >= 9.0.1 && < 10, hxt-regex-xmlschema >= 9 && < 10 + +Source-Repository head + Type: git + Location: git://github.com/UweSchmidt/hxt.git diff --git a/src/Text/XML/HXT/Arrow/DocumentInput.hs b/src/Text/XML/HXT/Arrow/DocumentInput.hs --- a/src/Text/XML/HXT/Arrow/DocumentInput.hs +++ b/src/Text/XML/HXT/Arrow/DocumentInput.hs @@ -15,8 +15,7 @@ -- ------------------------------------------------------------ module Text.XML.HXT.Arrow.DocumentInput - ( getURIContents - , getXmlContents + ( getXmlContents , getXmlEntityContents , getEncoding , getTextEncoding @@ -171,35 +170,31 @@ getHttpContents = withoutUserState $ applyA $ getSysVar theHttpHandler -getURIContents :: IOStateArrow s XmlTree XmlTree -getURIContents - = getContentsFromString - `orElse` - getContentsFromDoc - where - getContentsFromString - = ( getAttrValue a_source - >>> - isA (isPrefixOf stringProtocol) - ) - `guards` - getStringContents +getContentsFromString :: IOStateArrow s XmlTree XmlTree +getContentsFromString + = ( getAttrValue a_source + >>> + isA (isPrefixOf stringProtocol) + ) + `guards` + getStringContents - getContentsFromDoc - = ( ( addTransferURI $< getBaseURI - >>> - getCont - ) - `when` - ( setAbsURI $< ( getAttrValue a_source - >>^ - ( \ src-> (if null src then "stdin:" else src) ) -- empty document name -> read from stdin - ) - ) - ) +getContentsFromDoc :: IOStateArrow s XmlTree XmlTree +getContentsFromDoc + = ( ( addTransferURI $< getBaseURI >>> - setDocumentStatusFromSystemState "getURIContents" - + getCont + ) + `when` + ( setAbsURI $< ( getAttrValue a_source + >>^ + ( \ src-> (if null src then "stdin:" else src) ) -- empty document name -> read from stdin + ) + ) + ) + >>> + setDocumentStatusFromSystemState "getContentsFromDoc" + where setAbsURI src = ifA ( constA src >>> changeBaseURI ) this @@ -211,7 +206,7 @@ getCont = applyA ( getBaseURI -- compute the handler and call it >>> - traceValue 2 (("getURIContents: reading " ++) . show) + traceValue 2 (("getContentsFromDoc: reading " ++) . show) >>> getSchemeFromURI >>> @@ -280,25 +275,28 @@ getXmlContents' :: IOStateArrow s XmlTree XmlTree -> IOStateArrow s XmlTree XmlTree getXmlContents' parseEncodingSpec - = ( getURIContents - >>> - choiceA - [ isXmlHtmlDoc :-> ( parseEncodingSpec - >>> - filterErrorMsg - >>> - decodeDocument - ) - , isTextDoc :-> decodeDocument - , this :-> this - ] - >>> - perform ( getAttrValue transferURI - >>> - traceValue 1 (("getXmlContents: content read and decoded for " ++) . show) - ) - >>> - traceDoc "getXmlContents'" + = ( getContentsFromString -- no decoding done for string: protocol + `orElse` + ( getContentsFromDoc + >>> + choiceA + [ isXmlHtmlDoc :-> ( parseEncodingSpec + >>> + filterErrorMsg + >>> + decodeDocument + ) + , isTextDoc :-> decodeDocument + , this :-> this + ] + >>> + perform ( getAttrValue transferURI + >>> + traceValue 1 (("getXmlContents: content read and decoded for " ++) . show) + ) + >>> + traceDoc "getXmlContents'" + ) ) `when` isRoot diff --git a/src/Text/XML/HXT/Arrow/ReadDocument.hs b/src/Text/XML/HXT/Arrow/ReadDocument.hs --- a/src/Text/XML/HXT/Arrow/ReadDocument.hs +++ b/src/Text/XML/HXT/Arrow/ReadDocument.hs @@ -58,7 +58,7 @@ this filter can be configured by a list of configuration options, a value of type 'Text.XML.HXT.XmlState.TypeDefs.SysConfig' -for all available options see module 'Text.XML.HXT.XmlState.SystemConfig' +for all available options see module 'Text.XML.HXT.Arrow.XmlState.SystemConfig' - @withValidate yes\/no@ : switch on\/off DTD validation. Only for XML parsed documents, not for HTML parsing. @@ -88,27 +88,27 @@ - @withCheckNamespaces yes\/no@ : Switch on\/off namespace propagation and checking -- @withInputEncoding @ : +- @withInputEncoding \@ : Set default encoding. - @withTagSoup@ : use light weight and lazy parser based on tagsoup lib. This is only available when package hxt-tagsoup is installed and - 'Text.XML.HXT.TagSoup' is imported + the source contains an @import Text.XML.HXT.TagSoup@. -- @withRelaxNG @ : +- @withRelaxNG \@ : validate document with Relax NG, the parameter is for the schema URI. This implies using XML parser, no validation against DTD, and canonicalisation. -- @withCurl [...]@ : +- @withCurl [\...]@ : Use the libCurl binding for HTTP access. This is only available when package hxt-curl is installed and - 'Text.XML.HXT.Curl' is imported + the source contains an @import Text.XML.HXT.Curl@. -- @withHTTP [...]@ : +- @withHTTP [\...]@ : Use the Haskell HTTP package for HTTP access. This is only available when package hxt-http is installed and - 'Text.XML.HXT.HTTP' is imported + the source contains an @import Text.XML.HXT.HTTP@. examples: @@ -445,15 +445,16 @@ -- read a document that is stored in a normal Haskell String -- -- the same function as readDocument, but the parameter forms the input. --- All options available for 'readDocument' are applicable for readString. +-- All options available for 'readDocument' are applicable for readString, +-- except input encoding options. -- --- Default encoding: No encoding is done, the String argument is taken as Unicode string +-- Encoding: No decoding is done, the String argument is taken as Unicode string +-- All decoding must be done before calling readString, even if the +-- XML document contains an encoding spec. readString :: SysConfigList -> String -> IOStateArrow s b XmlTree readString config content - = readDocument - (withInputEncoding unicodeString : config) - (stringProtocol ++ content) + = readDocument config (stringProtocol ++ content) -- ------------------------------------------------------------ diff --git a/src/Text/XML/HXT/DOM/QualifiedName.hs b/src/Text/XML/HXT/DOM/QualifiedName.hs --- a/src/Text/XML/HXT/DOM/QualifiedName.hs +++ b/src/Text/XML/HXT/DOM/QualifiedName.hs @@ -182,7 +182,11 @@ px1 == px2 instance Ord QName where - compare x y = compare (qualifiedName x) (qualifiedName y) + compare (QN lp1 px1 ns1) (QN lp2 px2 ns2) + | isNullXName ns1 && isNullXName ns2 -- no namespaces set: px is significant + = compare (px1, lp1) (px2, lp2) + | otherwise -- namespace aware cmp: ns is significant, px is irrelevant + = compare (lp1, ns1) (lp2, ns2) instance NFData QName instance WNFData QName