state datatype consists of a system state and a user state the user state is not fixed

predefined system state data type with all components for the system functions, like trace, error handling, ...

The arrow type for stateful arrows

The arrow for stateful arrows with no user defined state

read the user defined part of the state

set the user defined part of the state

change the user defined part of the state

extend user state

Run an arrow with an extended user state component, The old component is stored together with a new one in a pair, the arrow is executed with this extended state, and the augmented state component is removed form the state when the arrow has finished its execution

change the type of user state

This conversion is useful, when running a state arrow with another structure of the user state, e.g. with () when executing some IO arrows

apply an IOSArrow to an empty root node with initialState () as initial state

the main entry point for running a state arrow with IO

when running runX f an empty XML root node is applied to f. usually f will start with a constant arrow (ignoring the input), e.g. a readDocument arrow.

for usage see examples with writeDocument

if input has to be feed into the arrow use runIOSLA like in runIOSLA f emptyX inputDoc

store a string in global state under a given attribute name

remove an entry in global state, arrow input remains unchanged

read an attribute value from global state

read all attributes from global state

store an int value in global state

read an int value from global state

getSysAttrInt 0 myIntAttr

reset global error variable

set global error variable

read current global error status

raise the global error status level to that of the input tree

set the error message handler and the flag for collecting the errors

the default error message handler: error output to stderr

error message handler for collecting errors

error message handler for output to stderr and collecting

error message handler for ignoring errors

if error messages are collected by the error handler for processing these messages by the calling application, this arrow reads the stored messages and clears the error message store

filter error messages from input trees and issue errors

generate a warnig message

generate an error message

generate a fatal error message, e.g. document not found

Default exception handler: issue a fatal error message and fail.

The parameter can be used to specify where the error occured

add the error level and the module where the error occured to the attributes of a document root node and remove the children when level is greater or equal to c_err. called by setDocumentStatusFromSystemState when the system state indicates an error

check whether the error level attribute in the system state is set to error, in this case the children of the document root are removed and the module name where the error occured and the error level are added as attributes with setDocumentStatus else nothing is changed

check whether tree is a document root and the status attribute has a value less than c_err

set the global trace level

read the global trace level

run an arrow with a given trace level, the old trace level is restored after the arrow execution

set the global trace command. This command does the trace output

acces the command for trace output

apply a trace arrow and issue message to stderr

issue a string message as trace

trace the current value transfered in a sequence of arrows.

The value is formated by a string conversion function. This is a substitute for the old and less general traceString function

an old alias for traceValue

issue the source representation of a document if trace level >= 3

for better readability the source is formated with indentDoc

issue the tree representation of a document if trace level >= 4

trace a main computation step issue a message when trace level >= 1, issue document source if level >= 3, issue tree when level is >= 4

set the base URI of a document, used e.g. for reading includes, e.g. external entities, the input must be an absolute URI

read the base URI from the globale state

change the base URI with a possibly relative URI, can be used for evaluating the xml:base attribute. Returns the new absolute base URI. Fails, if input is not parsable with parseURIReference

see also: setBaseURI, mkAbsURI

set the default base URI, if parameter is null, the system base ( file:///<cwd>/ ) is used, else the parameter, must be called before any document is read

get the default base URI

remember base uri, run an arrow and restore the base URI, used with external entity substitution

compute the absolut URI for a given URI and a base URI

arrow variant of expandURIString, fails if expandURIString returns Nothing

arrow for expanding an input URI into an absolute URI using global base URI, fails if input is not a legal URI

arrow for computing the fragment component of an URI, fails if input is not a legal URI

arrow for computing the path component of an URI, fails if input is not a legal URI

arrow for selecting the port number of the URI without leading ':', fails if input is not a legal URI

arrow for computing the query component of an URI, fails if input is not a legal URI

arrow for selecting the registered name (host) of the URI, fails if input is not a legal URI

arrow for selecting the scheme (protocol) of the URI, fails if input is not a legal URI.

See Network.URI for URI components

arrow for selecting the user info of the URI without trailing '@', fails if input is not a legal URI

read the system mimetype table

set the table mapping of file extensions to mime types in the system state

Default table is defined in MimeTypeDefaults. This table is used when reading loacl files, (file: protocol) to determine the mime type

set the table mapping of file extensions to mime types by an external config file

The config file must follow the conventions of etcmime.types on a debian linux system, that means all empty lines and all lines starting with a # are ignored. The other lines must consist of a mime type followed by a possible empty list of extensions. The list of extenstions and mime types overwrites the default list in the system state of the IOStateArrow

Specify the set of accepted mime types.

All contents of documents for which the mime type is not found in this list are discarded.

withSysAttr key value : store an arbitarty key value pair in system state

withCanonicalize yes/no : read option, canonicalize document, default is yes

Configure compression and decompression for binary serialization/deserialization. First component is the compression function applied after serialization, second the decompression applied before deserialization.

withCheckNamespaces yes/no: read option, check namespaces, default is no

withDefaultBaseURI URI , input option, set the default base URI

This option can be useful when parsing documents from stdin or contained in a string, and interpreting relative URIs within the document

Strict input for deserialization of binary data

withEncodingErrors yes/no : input option, ignore all encoding errors, default is no

withErrors yes/no : system option for suppressing error messages, default is no

Force a given mime type for all file contents.

The mime type for file access will then not be computed by looking into a mime.types file

withIgnoreNoneXmlContents yes/no : input option, ignore document contents of none XML/HTML documents.

This option can be useful for implementing crawler like applications, e.g. an URL checker. In those cases net traffic can be reduced.

withIndent yes/no : output option, indent document before output, default is no

withInputEncoding encodingName : input option

Set default document encoding (utf8, isoLatin1, usAscii, iso8859_2, ... , iso8859_16, ...). Only XML, HTML and text documents are decoded, default decoding for XML/HTML is utf8, for text iso latin1 (no decoding).

withMimeTypeFile filename : input option, set the mime type table for file: documents by given file. The format of this config file must be in the syntax of a debian linux "mime.types" config file

Specify a content handler for documents of a given mime type

withOutputEncoding encoding , output option, default is the default input encoding or utf8, if input encoding is not set

withOutputXML : output option, default writing

Default is writing XML: quote special XML chars >,<,",',& where neccessary, add XML processing instruction and encode document with respect to withOutputEncoding

Write XHTML: quote all special XML chars, use HTML entity refs or char refs for none ASCII chars

Write XML: quote only special XML chars, don't substitute chars by HTML entities, and don't generate empty elements for HTML elements, which may contain any contents, e.g. src=.../script instead of src=... /

suppreses all char and entitiy substitution

withParseByMimeType yes/no : read option, select the parser by the mime type of the document (pulled out of the HTTP header).

When the mime type is set to "text/html" the configured HTML parser is taken, when it's set to "text/xml" or "text/xhtml" the configured XML parser is taken. If the mime type is something else, no further processing is performed, the contents is given back to the application in form of a single text node. If the default document encoding is set to isoLatin1, this even enables processing of arbitray binary data.

withParseHTML yes/no: read option, use HTML parser, default is no (use XML parser)

withPreserveComment yes/no : read option, preserve comments during canonicalization, default is no

withProxy "host:port" : input option, configure a proxy for HTTP access, e.g. www-cache:3128

withRedirect yes/no : input option, automatically follow redirected URIs, default is yes

withRemoveWS yes/no : read and write option, remove all whitespace, used for document indentation, default is no

withStrictInput yes/no : input option, input of file and HTTP contents is read eagerly, default is no

withSubstDTDEntities yes/no: read option, substitute general entities defined in DTD, default is yes. switching this option and the validate option off can lead to faster parsing, because then there is no need to access the DTD

withSubstHTMLEntities yes/no: read option, substitute general entities defined in HTML DTD, default is no. switching this option on and the substDTDEntities and validate options off can lead to faster parsing because there is no need to access a DTD, but still the HTML general entities are substituted

withTace level : system option, set the trace level, (0..4)

withValidate yes/no: read option, validate document against DTD, default is yes

withWarnings yes/no : system option, issue warnings during reading, HTML parsing and processing, default is yes