/**
 * Licensed to the Apache Software Foundation (ASF) under one
 * or more contributor license agreements. See the NOTICE file
 * distributed with this work for additional information
 * regarding copyright ownership. The ASF licenses this file
 * to you under the Apache License, Version 2.0 (the
 * "License"); you may not use this file except in compliance
 * with the License.  You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package org.apache.commons.rdf.experimental;

import java.io.IOException;
import java.io.InputStream;
import java.nio.charset.StandardCharsets;
import java.nio.file.Path;
import java.util.Optional;
import java.util.concurrent.Future;
import java.util.function.Consumer;

import org.apache.commons.rdf.api.BlankNode;
import org.apache.commons.rdf.api.Dataset;
import org.apache.commons.rdf.api.Graph;
import org.apache.commons.rdf.api.IRI;
import org.apache.commons.rdf.api.Quad;
import org.apache.commons.rdf.api.RDFSyntax;
import org.apache.commons.rdf.api.RDFTerm;
import org.apache.commons.rdf.api.RDF;
import org.apache.commons.rdf.api.Triple;

/**
 * Parse an RDF source into a target (e.g. a Graph/Dataset).
 * <h2>Experimental</h2> This interface (and its implementations) should be
 * considered <strong>at risk</strong>; they might change or be removed in the
 * next minor update of Commons RDF. It may move to the the
 * {@link org.apache.commons.rdf.api} package when it has stabilized.
 * <h2>Description</h2>
 * <p>
 * This interface follows the
 * <a href="https://en.wikipedia.org/wiki/Builder_pattern">Builder pattern</a>,
 * allowing to set parser settings like {@link #contentType(RDFSyntax)} and
 * {@link #base(IRI)}. A caller MUST call one of the <code>source</code> methods
 * (e.g. {@link #source(IRI)}, {@link #source(Path)},
 * {@link #source(InputStream)}), and MUST call one of the <code>target</code>
 * methods (e.g. {@link #target(Consumer)}, {@link #target(Dataset)},
 * {@link #target(Graph)}) before calling {@link #parse()} on the returned
 * RDFParser - however methods can be called in any order.
 * <p>
 * The call to {@link #parse()} returns a {@link Future}, allowing asynchronous
 * parse operations. Callers are recommended to check {@link Future#get()} to
 * ensure parsing completed successfully, or catch exceptions thrown during
 * parsing.
 * <p>
 * Setting a method that has already been set will override any existing value
 * in the returned builder - regardless of the parameter type (e.g.
 * {@link #source(IRI)} will override a previous {@link #source(Path)}. Settings
 * can be unset by passing <code>null</code> - note that this may require
 * casting, e.g. <code>contentType( (RDFSyntax) null )</code> to undo a previous
 * call to {@link #contentType(RDFSyntax)}.
 * <p>
 * It is undefined if a RDFParser is mutable or thread-safe, so callers should
 * always use the returned modified RDFParser from the builder methods. The
 * builder may return itself after modification, or a cloned builder with the
 * modified settings applied. Implementations are however encouraged to be
 * immutable, thread-safe and document this. As an example starting point, see
 * <code>org.apache.commons.rdf.simple.AbstractRDFParser</code>.
 * <p>
 * Example usage:
 * </p>
 *
 * <pre>
 * Graph g1 = rDFTermFactory.createGraph();
 * new ExampleRDFParserBuilder().source(Paths.get("/tmp/graph.ttl")).contentType(RDFSyntax.TURTLE).target(g1).parse()
 *         .get(30, TimeUnit.Seconds);
 * </pre>
 *
 */
public interface RDFParser {

    /**
     * The result of {@link RDFParser#parse()} indicating parsing completed.
     * <p>
     * This is a marker interface that may be subclassed to include parser
     * details, e.g. warning messages or triple counts.
     */
    interface ParseResult {
    }

    /**
     * Specify which {@link RDF} to use for generating {@link RDFTerm}s.
     * <p>
     * This option may be used together with {@link #target(Graph)} to override
     * the implementation's default factory and graph.
     * <p>
     * <strong>Warning:</strong> Using the same {@link RDF} for multiple
     * {@link #parse()} calls may accidentally merge {@link BlankNode}s having
     * the same label, as the parser may use the
     * {@link RDF#createBlankNode(String)} method from the parsed blank node
     * labels.
     *
     * @see #target(Graph)
     * @param rdfTermFactory
     *            {@link RDF} to use for generating RDFTerms.
     * @return An {@link RDFParser} that will use the specified rdfTermFactory
     */
    RDFParser rdfTermFactory(RDF rdfTermFactory);

    /**
     * Specify the content type of the RDF syntax to parse.
     * <p>
     * This option can be used to select the RDFSyntax of the source, overriding
     * any <code>Content-Type</code> headers or equivalent.
     * <p>
     * The character set of the RDFSyntax is assumed to be
     * {@link StandardCharsets#UTF_8} unless overridden within the document
     * (e.g. {@code <?xml version="1.0" encoding="iso-8859-1"?>} in
     * {@link RDFSyntax#RDFXML}).
     * <p>
     * This method will override any contentType set with
     * {@link #contentType(String)}.
     *
     * @see #contentType(String)
     * @param rdfSyntax
     *            An {@link RDFSyntax} to parse the source according to, e.g.
     *            {@link RDFSyntax#TURTLE}.
     * @throws IllegalArgumentException
     *             If this RDFParser does not support the specified RDFSyntax.
     * @return An {@link RDFParser} that will use the specified content type.
     */
    RDFParser contentType(RDFSyntax rdfSyntax) throws IllegalArgumentException;

    /**
     * Specify the content type of the RDF syntax to parse.
     * <p>
     * This option can be used to select the RDFSyntax of the source, overriding
     * any <code>Content-Type</code> headers or equivalent.
     * <p>
     * The content type MAY include a <code>charset</code> parameter if the RDF
     * media types permit it; the default charset is
     * {@link StandardCharsets#UTF_8} unless overridden within the document.
     * <p>
     * This method will override any contentType set with
     * {@link #contentType(RDFSyntax)}.
     *
     * @see #contentType(RDFSyntax)
     * @param contentType
     *            A content-type string, e.g. <code>application/ld+json</code>
     *            or <code>text/turtle;charset="UTF-8"</code> as specified by
     *            <a href="https://tools.ietf.org/html/rfc7231#section-3.1.1.1">
     *            RFC7231</a>.
     * @return An {@link RDFParser} that will use the specified content type.
     * @throws IllegalArgumentException
     *             If the contentType has an invalid syntax, or this RDFParser
     *             does not support the specified contentType.
     */
    RDFParser contentType(String contentType) throws IllegalArgumentException;

    /**
     * Specify a {@link Graph} to add parsed triples to.
     * <p>
     * If the source supports datasets (e.g. the {@link #contentType(RDFSyntax)}
     * set has {@link RDFSyntax#supportsDataset} is true)), then only quads in
     * the <em>default graph</em> will be added to the Graph as {@link Triple}s.
     * <p>
     * It is undefined if any triples are added to the specified {@link Graph}
     * if {@link #parse()} throws any exceptions. (However implementations are
     * free to prevent this using transaction mechanisms or similar). If
     * {@link Future#get()} does not indicate an exception, the parser
     * implementation SHOULD have inserted all parsed triples to the specified
     * graph.
     * <p>
     * Calling this method will override any earlier targets set with
     * {@link #target(Graph)}, {@link #target(Consumer)} or
     * {@link #target(Dataset)}.
     * <p>
     * The default implementation of this method calls {@link #target(Consumer)}
     * with a {@link Consumer} that does {@link Graph#add(Triple)} with
     * {@link Quad#asTriple()} if the quad is in the default graph.
     *
     * @param graph
     *            The {@link Graph} to add triples to.
     * @return An {@link RDFParser} that will insert triples into the specified
     *         graph.
     */
    default RDFParser target(final Graph graph) {
        return target(q -> {
            if (!q.getGraphName().isPresent()) {
                graph.add(q.asTriple());
            }
        });
    }

    /**
     * Specify a {@link Dataset} to add parsed quads to.
     * <p>
     * It is undefined if any quads are added to the specified {@link Dataset}
     * if {@link #parse()} throws any exceptions. (However implementations are
     * free to prevent this using transaction mechanisms or similar). On the
     * other hand, if {@link #parse()} does not indicate an exception, the
     * implementation SHOULD have inserted all parsed quads to the specified
     * dataset.
     * <p>
     * Calling this method will override any earlier targets set with
     * {@link #target(Graph)}, {@link #target(Consumer)} or
     * {@link #target(Dataset)}.
     * <p>
     * The default implementation of this method calls {@link #target(Consumer)}
     * with a {@link Consumer} that does {@link Dataset#add(Quad)}.
     *
     * @param dataset
     *            The {@link Dataset} to add quads to.
     * @return An {@link RDFParser} that will insert triples into the specified
     *         dataset.
     */
    default RDFParser target(final Dataset dataset) {
        return target(dataset::add);
    }

    /**
     * Specify a consumer for parsed quads.
     * <p>
     * The quads will include triples in all named graphs of the parsed source,
     * including any triples in the default graph. When parsing a source format
     * which do not support datasets, all quads delivered to the consumer will
     * be in the default graph (e.g. their {@link Quad#getGraphName()} will be
     * as {@link Optional#empty()}), while for a source
     * <p>
     * It is undefined if any quads are consumed if {@link #parse()} throws any
     * exceptions. On the other hand, if {@link #parse()} does not indicate an
     * exception, the implementation SHOULD have produced all parsed quads to
     * the specified consumer.
     * <p>
     * Calling this method will override any earlier targets set with
     * {@link #target(Graph)}, {@link #target(Consumer)} or
     * {@link #target(Dataset)}.
     * <p>
     * The consumer is not assumed to be thread safe - only one
     * {@link Consumer#accept(Object)} is delivered at a time for a given
     * {@link RDFParser#parse()} call.
     * <p>
     * This method is typically called with a functional consumer, for example:
     *
     * <pre>
     * {@code
     * List<Quad> quads = new ArrayList<Quad>;
     * parserBuilder.target(quads::add).parse();
     * }
     * </pre>
     *
     * @param consumer
     *            A {@link Consumer} of {@link Quad}s
     * @return An {@link RDFParser} that will call the consumer for into the
     *         specified dataset.
     */
    RDFParser target(Consumer<Quad> consumer);

    /**
     * Specify a base IRI to use for parsing any relative IRI references.
     * <p>
     * Setting this option will override any protocol-specific base IRI (e.g.
     * <code>Content-Location</code> header) or the {@link #source(IRI)} IRI,
     * but does not override any base IRIs set within the source document (e.g.
     * <code>@base</code> in Turtle documents).
     * <p>
     * If the source is in a syntax that does not support relative IRI
     * references (e.g. {@link RDFSyntax#NTRIPLES}), setting the
     * <code>base</code> has no effect.
     * <p>
     * This method will override any base IRI set with {@link #base(String)}.
     *
     * @see #base(String)
     * @param base
     *            An absolute IRI to use as a base.
     * @return An {@link RDFParser} that will use the specified base IRI.
     */
    RDFParser base(IRI base);

    /**
     * Specify a base IRI to use for parsing any relative IRI references.
     * <p>
     * Setting this option will override any protocol-specific base IRI (e.g.
     * <code>Content-Location</code> header) or the {@link #source(IRI)} IRI,
     * but does not override any base IRIs set within the source document (e.g.
     * <code>@base</code> in Turtle documents).
     * <p>
     * If the source is in a syntax that does not support relative IRI
     * references (e.g. {@link RDFSyntax#NTRIPLES}), setting the
     * <code>base</code> has no effect.
     * <p>
     * This method will override any base IRI set with {@link #base(IRI)}.
     *
     * @see #base(IRI)
     * @param base
     *            An absolute IRI to use as a base.
     * @return An {@link RDFParser} that will use the specified base IRI.
     * @throws IllegalArgumentException
     *             If the base is not a valid absolute IRI string
     */
    RDFParser base(String base) throws IllegalArgumentException;

    /**
     * Specify a source {@link InputStream} to parse.
     * <p>
     * The source set will not be read before the call to {@link #parse()}.
     * <p>
     * The InputStream will not be closed after parsing. The InputStream does
     * not need to support {@link InputStream#markSupported()}.
     * <p>
     * The parser might not consume the complete stream (e.g. an RDF/XML parser
     * may not read beyond the closing tag of
     * <code>&lt;/rdf:Description&gt;</code>).
     * <p>
     * The {@link #contentType(RDFSyntax)} or {@link #contentType(String)}
     * SHOULD be set before calling {@link #parse()}.
     * <p>
     * The character set is assumed to be {@link StandardCharsets#UTF_8} unless
     * the {@link #contentType(String)} specifies otherwise or the document
     * declares its own charset (e.g. RDF/XML with a
     * <code>&lt;?xml encoding="iso-8859-1"&gt;</code> header).
     * <p>
     * The {@link #base(IRI)} or {@link #base(String)} MUST be set before
     * calling {@link #parse()}, unless the RDF syntax does not permit relative
     * IRIs (e.g. {@link RDFSyntax#NTRIPLES}).
     * <p>
     * This method will override any source set with {@link #source(IRI)},
     * {@link #source(Path)} or {@link #source(String)}.
     *
     * @param inputStream
     *            An InputStream to consume
     * @return An {@link RDFParser} that will use the specified source.
     */
    RDFParser source(InputStream inputStream);

    /**
     * Specify a source file {@link Path} to parse.
     * <p>
     * The source set will not be read before the call to {@link #parse()}.
     * <p>
     * The {@link #contentType(RDFSyntax)} or {@link #contentType(String)}
     * SHOULD be set before calling {@link #parse()}.
     * <p>
     * The character set is assumed to be {@link StandardCharsets#UTF_8} unless
     * the {@link #contentType(String)} specifies otherwise or the document
     * declares its own charset (e.g. RDF/XML with a
     * <code>&lt;?xml encoding="iso-8859-1"&gt;</code> header).
     * <p>
     * The {@link #base(IRI)} or {@link #base(String)} MAY be set before calling
     * {@link #parse()}, otherwise {@link Path#toUri()} will be used as the base
     * IRI.
     * <p>
     * This method will override any source set with {@link #source(IRI)},
     * {@link #source(InputStream)} or {@link #source(String)}.
     *
     * @param file
     *            A Path for a file to parse
     * @return An {@link RDFParser} that will use the specified source.
     */
    RDFParser source(Path file);

    /**
     * Specify an absolute source {@link IRI} to retrieve and parse.
     * <p>
     * The source set will not be read before the call to {@link #parse()}.
     * <p>
     * If this builder does not support the given IRI protocol (e.g.
     * <code>urn:uuid:ce667463-c5ab-4c23-9b64-701d055c4890</code>), this method
     * should succeed, while the {@link #parse()} should throw an
     * {@link IOException}.
     * <p>
     * The {@link #contentType(RDFSyntax)} or {@link #contentType(String)} MAY
     * be set before calling {@link #parse()}, in which case that type MAY be
     * used for content negotiation (e.g. <code>Accept</code> header in HTTP),
     * and SHOULD be used for selecting the RDFSyntax.
     * <p>
     * The character set is assumed to be {@link StandardCharsets#UTF_8} unless
     * the protocol's equivalent of <code>Content-Type</code> specifies
     * otherwise or the document declares its own charset (e.g. RDF/XML with a
     * <code>&lt;?xml encoding="iso-8859-1"&gt;</code> header).
     * <p>
     * The {@link #base(IRI)} or {@link #base(String)} MAY be set before calling
     * {@link #parse()}, otherwise the source IRI will be used as the base IRI.
     * <p>
     * This method will override any source set with {@link #source(Path)},
     * {@link #source(InputStream)} or {@link #source(String)}.
     *
     * @param iri
     *            An IRI to retrieve and parse
     * @return An {@link RDFParser} that will use the specified source.
     */
    RDFParser source(IRI iri);

    /**
     * Specify an absolute source IRI to retrieve and parse.
     * <p>
     * The source set will not be read before the call to {@link #parse()}.
     * <p>
     * If this builder does not support the given IRI (e.g.
     * <code>urn:uuid:ce667463-c5ab-4c23-9b64-701d055c4890</code>), this method
     * should succeed, while the {@link #parse()} should throw an
     * {@link IOException}.
     * <p>
     * The {@link #contentType(RDFSyntax)} or {@link #contentType(String)} MAY
     * be set before calling {@link #parse()}, in which case that type MAY be
     * used for content negotiation (e.g. <code>Accept</code> header in HTTP),
     * and SHOULD be used for selecting the RDFSyntax.
     * <p>
     * The character set is assumed to be {@link StandardCharsets#UTF_8} unless
     * the protocol's equivalent of <code>Content-Type</code> specifies
     * otherwise or the document declares its own charset (e.g. RDF/XML with a
     * <code>&lt;?xml encoding="iso-8859-1"&gt;</code> header).
     * <p>
     * The {@link #base(IRI)} or {@link #base(String)} MAY be set before calling
     * {@link #parse()}, otherwise the source IRI will be used as the base IRI.
     * <p>
     * This method will override any source set with {@link #source(Path)},
     * {@link #source(InputStream)} or {@link #source(IRI)}.
     *
     * @param iri
     *            An IRI to retrieve and parse
     * @return An {@link RDFParser} that will use the specified source.
     * @throws IllegalArgumentException
     *             If the base is not a valid absolute IRI string
     *
     */
    RDFParser source(String iri) throws IllegalArgumentException;

    /**
     * Parse the specified source.
     * <p>
     * A source method (e.g. {@link #source(InputStream)}, {@link #source(IRI)},
     * {@link #source(Path)}, {@link #source(String)} or an equivalent subclass
     * method) MUST have been called before calling this method, otherwise an
     * {@link IllegalStateException} will be thrown.
     * <p>
     * A target method (e.g. {@link #target(Consumer)},
     * {@link #target(Dataset)}, {@link #target(Graph)} or an equivalent
     * subclass method) MUST have been called before calling parse(), otherwise
     * an {@link IllegalStateException} will be thrown.
     * <p>
     * It is undefined if this method is thread-safe, however the
     * {@link RDFParser} may be reused (e.g. setting a different source) as soon
     * as the {@link Future} has been returned from this method.
     * <p>
     * The RDFParser SHOULD perform the parsing as an asynchronous operation,
     * and return the {@link Future} as soon as preliminary checks (such as
     * validity of the {@link #source(IRI)} and {@link #contentType(RDFSyntax)}
     * settings) have finished. The future SHOULD not mark
     * {@link Future#isDone()} before parsing is complete. A synchronous
     * implementation MAY be blocking on the <code>parse()</code> call and
     * return a Future that is already {@link Future#isDone()}.
     * <p>
     * The returned {@link Future} contains a {@link ParseResult}.
     * Implementations may subclass this interface to provide any parser
     * details, e.g. list of warnings. <code>null</code> is a possible return
     * value if no details are available, but parsing succeeded.
     * <p>
     * If an exception occurs during parsing, (e.g. {@link IOException} or
     * <code>org.apache.commons.rdf.simple.experimental.RDFParseException</code>),
     * it should be indicated as the
     * {@link java.util.concurrent.ExecutionException#getCause()} in the
     * {@link java.util.concurrent.ExecutionException} thrown on
     * {@link Future#get()}.
     *
     * @return A Future that will return the populated {@link Graph} when the
     *         parsing has finished.
     * @throws IOException
     *             If an error occurred while starting to read the source (e.g.
     *             file not found, unsupported IRI protocol). Note that IO
     *             errors during parsing would instead be the
     *             {@link java.util.concurrent.ExecutionException#getCause()} of
     *             the {@link java.util.concurrent.ExecutionException} thrown on
     *             {@link Future#get()}.
     * @throws IllegalStateException
     *             If the builder is in an invalid state, e.g. a
     *             <code>source</code> has not been set.
     */
    Future<? extends ParseResult> parse() throws IOException, IllegalStateException;
}