org.apache.jena.riot

Class RDFParserBuilder

java.lang.Object

org.apache.jena.riot.RDFParserBuilder

public class RDFParserBuilder
extends Object

An RDFParser is a process that will generate triples; RDFParserBuilder provides the means to setup the parser.

An RDFParser has a predefined source; the target for output is given when the "parse" method is called. It can be used multiple times in which case the same source is reread. The destination can vary. The application is responsible for concurrency of the destination of the parse operation. The process is

     StreamRDF destination = ...
     RDFParser parser = RDFParser.create()
          .source("filename.ttl")
          .build();
     parser.parse(destination); 
 

or using a short cut:

     RDFParser parser = RDFParser.create()
          .source("filename.ttl")
          .parse(destination); 
 

Method Summary

Modifier and Type	Method and Description
RDFParserBuilder	base(String base) Set the base URI for parsing.
RDFParser	build() Build an RDFParser.
RDFParserBuilder	canonicalLiterals(boolean flag) Deprecated. Use canonicalValues and one of langTagCanonical() and langTagLowerCase() This operation is equivalent to this.canonicalValues(flag); if ( flag ) this.langTagCanonical(); else this.langTagAsGiven(); return this;
RDFParserBuilder	canonicalValues(boolean flag) Convert the lexical form of literals to a canonical form.
RDFParserBuilder	checking(boolean flag) Set whether to perform checking, NTriples and NQuads default to no checking, other languages to checking.
RDFParserBuilder	clone() Duplicate this builder with current settings.
RDFParserBuilder	context(Context context) Set the context for the parser when built.
static RDFParserBuilder	create()
RDFParserBuilder	errorHandler(ErrorHandler handler) Set the ErrorHandler to use.
RDFParserBuilder	factory(FactoryRDF factory) Set the FactoryRDF to use.
RDFParserBuilder	forceLang(Lang lang) Force the choice RDF syntax to be lang, and ignore any indications such as file extension or HTTP Content-Type.
RDFParserBuilder	fromString(String string) Use the given string as the content to parse.
RDFParserBuilder	httpAccept(String acceptHeader) Set the HTTP "Accept" header.
RDFParserBuilder	httpClient(org.apache.http.client.HttpClient httpClient) Set the HttpClient to use.
RDFParserBuilder	httpHeader(String header, String value) Set an HTTP header.
RDFParserBuilder	labelToNode(LabelToNode labelToNode) Use the given LabelToNode, the policy for converting blank node labels in RDF syntax to Jena's Node objects (usually a blank node).
RDFParserBuilder	lang(Lang lang) Set the hint Lang.
RDFParserBuilder	langTagAsGiven() The form of the language tags as given in the data is preserved.
RDFParserBuilder	langTagCanonical() Language tags are case-normalized as defined by RFC 5646.
RDFParserBuilder	langTagLowerCase() Convert language tags to lower case.
void	parse(Dataset dataset) Parse the source, sending the results to a Dataset.
void	parse(DatasetGraph dataset) Parse the source, sending the results to a DatasetGraph.
void	parse(Graph graph) Parse the source, sending the results to a Graph.
void	parse(Model model) Parse the source, sending the results to a Model.
void	parse(StreamRDF stream) Parse the source, sending the results to a StreamRDF.
RDFParserBuilder	resolveURIs(boolean flag) Choose whether to resolve URIs. This does not affect all languages: N-Triples and N-Quads never resolve URIs. Relative URIs are bad data. Only set this to false for debugging and development purposes.
RDFParserBuilder	set(Symbol symbol, Object value) Added a setting to the context for the parser when built.
RDFParserBuilder	source(InputStream input) Set the source to InputStream.
RDFParserBuilder	source(Path path) Set the source to Path.
RDFParserBuilder	source(Reader reader) Deprecated. Use fromString(java.lang.String), or an InputStream or a StringReader.
RDFParserBuilder	source(String uriOrFile) Set the source to a URI; this includes OS file names.
RDFParserBuilder	source(StringReader reader) Set the source to StringReader.
RDFParserBuilder	streamManager(StreamManager streamManager) Set the StreamManager to use when opening a URI (including files by name, but not by Path).
RDFParserBuilder	strict(boolean strictMode) Set the parser built to "strict" mode.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

create

public static RDFParserBuilder create()

source

public RDFParserBuilder source(Path path)

Set the source to Path. This clears any other source setting.

The parser can be reused.

Parameters:

path -

Returns:

this

source

public RDFParserBuilder source(String uriOrFile)

Set the source to a URI; this includes OS file names. File URL should be of the form file:///.... This clears any other source setting.

The parser can be reused.

Parameters:

uriOrFile -

Returns:

this

fromString

public RDFParserBuilder fromString(String string)

Use the given string as the content to parse. This clears any other source setting.

The syntax must be set with .lang(...).

The parser can be reused.

Parameters:

string - The characters to be parsed.

Returns:

this

source

public RDFParserBuilder source(InputStream input)

Set the source to InputStream. This clears any other source setting.

The syntax must be set with .lang(...).

The InputStream will be closed when the parser is called and the parser can not be reused.

Parameters:

input -

Returns:

this

source

public RDFParserBuilder source(StringReader reader)

Set the source to StringReader. This clears any other source setting. The StringReader will be closed when the parser is called and the parser can not be reused.

The syntax must be set with .lang(...).

Consider using fromString(java.lang.String) instead.

Parameters:

reader -

Returns:

this

source

@Deprecated
public RDFParserBuilder source(Reader reader)

Deprecated. Use fromString(java.lang.String), or an InputStream or a StringReader.

Set the source to Reader. This clears any other source setting. The Reader will be closed when the parser is called and the parser can not be reused.

The syntax must be set with .lang(...).

Parameters:

reader -

Returns:

this

streamManager

public RDFParserBuilder streamManager(StreamManager streamManager)

Set the StreamManager to use when opening a URI (including files by name, but not by Path).

Parameters:

streamManager -

Returns:

this

lang

public RDFParserBuilder lang(Lang lang)

Set the hint Lang. This is the RDF syntax used when there is no way to deduce the syntax (e.g. read from a InputStream, no recognized file extension, no recognized HTTP Content-Type provided).

Parameters:

lang -

Returns:

this

strict

public RDFParserBuilder strict(boolean strictMode)

Set the parser built to "strict" mode. The default is system wide setting of SysRIOT.isStrictMode().

Parameters:

strictMode -

Returns:

this

forceLang

public RDFParserBuilder forceLang(Lang lang)

Force the choice RDF syntax to be lang, and ignore any indications such as file extension or HTTP Content-Type.

Parameters:

lang -

Returns:

this

See Also:

Lang

httpAccept

public RDFParserBuilder httpAccept(String acceptHeader)

Set the HTTP "Accept" header. The default if not set is WebContent.defaultRDFAcceptHeader.

Parameters:

acceptHeader -

Returns:

this

httpHeader

public RDFParserBuilder httpHeader(String header,
                                   String value)

Set an HTTP header. Any previous setting is lost.

Consider setting up an HttpClient if more complicated setting to an HTTP request is required.

httpClient

public RDFParserBuilder httpClient(org.apache.http.client.HttpClient httpClient)

Set the HttpClient to use. This will override any HTTP header settings set for this builder.

base

public RDFParserBuilder base(String base)

Set the base URI for parsing. The default is to have no base URI.

resolveURIs

public RDFParserBuilder resolveURIs(boolean flag)

Choose whether to resolve URIs.
This does not affect all languages: N-Triples and N-Quads never resolve URIs.
Relative URIs are bad data.
Only set this to false for debugging and development purposes.

canonicalLiterals

@Deprecated
public RDFParserBuilder canonicalLiterals(boolean flag)

Deprecated. Use canonicalValues and one of langTagCanonical() and langTagLowerCase()

This operation is equivalent to

   this.canonicalValues(flag);
    if ( flag )
        this.langTagCanonical();
    else
        this.langTagAsGiven();
    return this;
 

Convert the lexical form of literals to a canonical form.

canonicalValues

public RDFParserBuilder canonicalValues(boolean flag)

Convert the lexical form of literals to a canonical form.

Two literals can be different RDF terms for the same value.

Examples include (first shown of the pair is the canonical form):

     "1"^^xsd:integer and  "+01"^^xsd:integer 
     "1.0E0"^^xsd:double and  "1"^^xsd:double
 

The canonical forms follow XSD 1.1 <href="https://www.w3.org/TR/xmlschema11-2/#canonical-lexical-representation">2.3.1 Canonical Mapping</a> except in the case of xsd:decimal where it follows the older XSD 1.0 which makes it legal for Turtle's short form ("1.0"^^xsd:Decimal rather than "1"^^xsd:decimal). See XSD 1.0 3.2.3.2 Canonical representation

The effect on literals where the lexical form does not represent a valid value (for example, "3000"^^xsd:byte) is undefined.

This option is off by default.

This option can slow parsing down.

For consistent loading of data, it is recommended that data is cleaned and canonicalized before loading so the conversion is done once.

See Also:

langTagLowerCase(), langTagCanonical()

langTagLowerCase

public RDFParserBuilder langTagLowerCase()

Convert language tags to lower case.

This is the suggested form in RDF 1.1 for comparsions. However, this is not the recommended canonical form in RFC 5646.

Providing all data is converted consistently, language tag equality is maintained for either lower case or RFC canonicalization styles.

This option can slow parsing down.

See Also:

langTagCanonical()

langTagCanonical

public RDFParserBuilder langTagCanonical()

Language tags are case-normalized as defined by RFC 5646. Example: en-GB, not en-gb.

This does not affect the RDF 1.1 requirement that the value-space of language tags is lower-case.

Providing all data is converted consistently, lang tag equality is maintained for either lower case or RFC canonicalization.

This option can slow parsing down.

See Also:

langTagLowerCase()

langTagAsGiven

public RDFParserBuilder langTagAsGiven()

The form of the language tags as given in the data is preserved. This is the default behaviour of parsing.

See Also:

langTagLowerCase(), langTagCanonical()

checking

public RDFParserBuilder checking(boolean flag)

Set whether to perform checking, NTriples and NQuads default to no checking, other languages to checking.

Checking adds warnings over and above basic syntax errors.

• URIs - whether IRs confirm to all the rules of the URI scheme
• Literals: whether the lexical form conforms to the rules for the datatype.
• Triples and quads: check slots have a valid kind of RDF term (parsers usually make this a syntax error anyway).

See also errorHandler(ErrorHandler) to control the output. The default is to log. This can also be used to turn warnings into exceptions.

errorHandler

public RDFParserBuilder errorHandler(ErrorHandler handler)

Set the ErrorHandler to use. This replaces any previous setting. The default is use slf4j logger "RIOT".

Parameters:

handler -

Returns:

this

factory

public RDFParserBuilder factory(FactoryRDF factory)

Set the FactoryRDF to use. FactoryRDF control how parser output is turned into Node and how Triples and Quads are built. This replaces any previous setting.
The default is use RiotLib.factoryRDF() which is provides Node reuse.
The FactoryRDF also determines how blank node labels in RDF syntax are mapped to BlankNodeId. Use

    new Factory(myLabelToNode) 
 

to create an FactoryRDF and set the LabelToNode step.

Parameters:

factory -

Returns:

this

See Also:

labelToNode

labelToNode

public RDFParserBuilder labelToNode(LabelToNode labelToNode)

Use the given LabelToNode, the policy for converting blank node labels in RDF syntax to Jena's Node objects (usually a blank node).
Only applies when the FactoryRDF is not set in the RDFParserBuilder, otherwise the FactoryRDF controls the label-to-node process.
SyntaxLabels.createLabelToNode() is the default policy.
LabelToNode.createUseLabelAsGiven() uses the label in the RDF syntax directly. This does not produce safe RDF and should only be used for development and debugging.

Parameters:

labelToNode -

Returns:

this

See Also:

factory

context

public RDFParserBuilder context(Context context)

Set the context for the parser when built. If a context is already partly set for this builder, merge the new settings into the outstanding context. If the context argument is null, do nothing.

Parameters:

context -

Returns:

this

See Also:

Context

set

public RDFParserBuilder set(Symbol symbol,
                            Object value)

Added a setting to the context for the parser when built. A value of "null" removes a previous setting.

Parameters:

symbol -

value -

Returns:

this

See Also:

Context

parse

public void parse(StreamRDF stream)

Parse the source, sending the results to a StreamRDF. Short form for build().parse(stream).

Parameters:

stream -

parse

public void parse(Graph graph)

Parse the source, sending the results to a Graph. The source must be for triples; any quads are discarded. Short form for build().parse(graph) which sends triples and prefixes to the Graph.

Parameters:

graph -

parse

public void parse(Model model)

Parse the source, sending the results to a Model. The source must be for triples; any quads are discarded. Short form for build().parse(model) which sends triples and prefixes to the Model.

Parameters:

model -

parse

public void parse(DatasetGraph dataset)

Parse the source, sending the results to a DatasetGraph. Short form for build().parse(dataset) which sends triples and prefixes to the DatasetGraph.

Parameters:

dataset -

parse

public void parse(Dataset dataset)

Parse the source, sending the results to a Dataset. Short form for build().parse(dataset) which sends triples and prefixes to the Dataset.

Parameters:

dataset -

build

public RDFParser build()

Build an RDFParser. The parser takes it's configuration from this builder and can not then be changed. The source must be set. When a parser is used, it is takes the source and sends output to an StreamRDF.

Shortcuts:

• parse(DatasetGraph) - parse the source and output to a DatasetGraph
• parse(Graph) - parse the source and output to a Graph
• parse(StreamRDF) - parse the source and output to a StreamRDF

Returns:

RDFParser

clone

public RDFParserBuilder clone()

Duplicate this builder with current settings. Changes to setting to this builder do not affect the clone.

Overrides:

clone in class Object