Skip to content

Commit

Permalink
bug19280 - Java tutorial update to example12()
Browse files Browse the repository at this point in the history 
The free-text indexing example in the Java
tutorial (both html and code) needed to be
updated.  I have improved it, but some of the
free-text functionality seems missing. I have
upgraded the example (and text) so that it
isn't actually *wrong* anymore, but it will
need additional work in the future. I have
left the bug issue open.

No release notes. The java examples ran. The
html file looked ok in the browser.

Change-Id: I8ab74aa763376446ace78cfd5855233dd64b625e
  • Loading branch information
@BruceDClayton
BruceDClayton committed May 12, 2010
1 parent f417d3a commit 711b6db
Show file tree
Hide file tree
Showing 2 changed files with 43 additions and 16 deletions.
32 changes: 28 additions & 4 deletions src/tutorial/TutorialExamples.java
View file
Expand Up @@ -1242,25 +1242,47 @@ public static void example12 () throws Exception {
ValueFactory f = conn.getValueFactory();
String exns = "http://example.org/people/";
conn.setNamespace("ex", exns);
conn.createFreetextIndex("fti2", new URI[]{f.createURI(exns,"fullname")});
// Create index1
conn.createFreetextIndex("index1", new URI[]{f.createURI(exns,"fullname")});
// Create parts of person resources.
URI alice = f.createURI(exns, "alice1");
URI carroll = f.createURI(exns, "carroll");
URI persontype = f.createURI(exns, "Person");
URI fullname = f.createURI(exns, "fullname");
Literal alicename = f.createLiteral("Alice B. Toklas");
Literal lewisCarroll = f.createLiteral("Lewis Carroll");
// Create parts of book resources.
URI book = f.createURI(exns, "book1");
URI booktype = f.createURI(exns, "Book");
URI booktitle = f.createURI(exns, "title");
URI booktitle = f.createURI(exns, "title");
URI author = f.createURI(exns, "author");
Literal wonderland = f.createLiteral("Alice in Wonderland");
// Add Alice B. Toklas triples
conn.clear();
conn.add(alice, RDF.TYPE, persontype);
conn.add(alice, fullname, alicename);
// Add Alice in Wonderland triples
conn.add(book, RDF.TYPE, booktype);
conn.add(book, booktitle, wonderland);
conn.add(book, author, carroll);
// Add Lewis Carroll triples
conn.add(carroll, RDF.TYPE, persontype);
conn.add(carroll, fullname, lewisCarroll);
// Check triples
RepositoryResult<Statement> statements = conn.getStatements(null, null, null, false);
try {
while (statements.hasNext()) {
println(statements.next());
}
} finally {
statements.close();
}

println("\nWhole-word match for 'Alice'.");
String queryString =
"SELECT ?s ?p ?o " +
"WHERE { ?s ?p ?o . ?s fti:match 'Alice' . }";
"WHERE { ?s ?p ?o . " +
" ?s fti:match 'Alice' . }";
TupleQuery tupleQuery = conn.prepareTupleQuery(QueryLanguage.SPARQL, queryString);
TupleQueryResult result = (TupleQueryResult)tupleQuery.evaluate();
int count = 0;
Expand All @@ -1272,7 +1294,8 @@ public static void example12 () throws Exception {
count += 1;
}
result.close();



println("\nWildcard match for 'Ali*'.");
queryString =
"SELECT ?s ?p ?o " +
Expand Down Expand Up @@ -1322,6 +1345,7 @@ public static void example12 () throws Exception {
result.close();
}


/**
* Ask, Construct, and Describe queries
*/
Expand Down
27 changes: 15 additions & 12 deletions src/tutorial/java-tutorial-40.html
View file
Expand Up @@ -1042,28 +1042,31 @@ <h2 id="Namespaces">Namespaces (example11()) &nbsp;&nbsp;&nbsp;<a class="returnl
<h2 id="Free Text Search">Free Text Search (example12()) &nbsp;&nbsp;&nbsp;<a class="returnlink" href="#Contents">Return to Top</a></h2>
<p>
It is common for users to build RDF applications that combine
some form of "keyword search" with their queries. For example, a user
might want to retrieve all triples for which the string "Alice" appears
as a word within the third (object) argument to the triple. AllegroGraph
provides a capability for including free text matching within a SPARQL
query. It requires, however, that you register the predicates that will participate in text searches so they can be indexed. </p>
some form of "keyword search" with their queries. For example, a user might want to retrieve all triples for which the string "Alice" appears as a word within the third (object) argument to the triple. AllegroGraph provides a capability for including free text matching within a SPARQL query. It requires, however, that you create and configure indexes appropriate to the searches you want to pursue. </p>
<p> The example example12() begins by borrowing the connection object from example1(). Then it creates a namespace string and registers the namespace with the connection object, as in the previous example. </p>
<pre class="input"> public static void example12 () throws Exception { <br> AGRepositoryConnection conn = example1(false);<br> ValueFactory f = conn.getValueFactory();<br> String exns = &quot;http://example.org/people/&quot;;<br> conn.setNamespace(&quot;ex&quot;, exns);</pre>
<p>We have to register the predicates that will participate in text indexing. In the example12() example below, we have
called the connection method <strong>registerFreeTextPredicate()</strong> to register the predicate "http://example.org/people/fullname" for text indexing. Generating the predicate's URI is a separate step. </p>
<pre class="input"> conn.registerFreetextPredicate(f.createURI(exns,&quot;fullname&quot;));</pre>
<p>The next step is to create two new resources, &quot;Alice1&quot; named &quot;Alice B. Toklas,&quot; and &quot;book1&quot; with the title &quot;Alice in Wonderland.&quot; Notice that we did not register the book title predicate for text indexing. </p>
<pre class="input"> URI alice = f.createURI(exns, &quot;alice1&quot;);<br> URI persontype = f.createURI(exns, &quot;Person&quot;);<br> URI fullname = f.createURI(exns, &quot;fullname&quot;); <br> Literal alicename = f.createLiteral(&quot;Alice B. Toklas&quot;);<br> URI book = f.createURI(exns, &quot;book1&quot;);<br> URI booktype = f.createURI(exns, &quot;Book&quot;);<br> URI booktitle = f.createURI(exns, &quot;title&quot;); <br> Literal wonderland = f.createLiteral(&quot;Alice in Wonderland&quot;);</pre>
<p>We have to create an index. AllegroGraph lets you create any number of text indexes, each for a specific purpose. In this case we are indexing the literal values we find in the &quot;fullname&quot; predicate, which we will use in resources that describe people. The createFreeTextIndex() method has many configurable parameters. Their default settings are appropriate to this situation. All we have to provide is a name for the index and the URI of the predicate (or predicates) that contain the text to be indexed. </p>
<pre class="input"> conn.createFreetextIndex(&quot;index1&quot;, new URI[]{f.createURI(exns,&quot;fullname&quot;)});</pre>
<p>The next step is to create three new resources. One is a person, &quot;Alice,&quot; whose full name is &quot;Alice B. Toklas.&quot; There will also be a book, &quot;Alice in Wonderland,&quot; which is linked to an author resource, &quot;Lewis Carroll.&quot; The first step is to create some URIs and literal values from which to assemble the triples: </p>
<pre class="input"> // Create parts of person resources. <br> URI alice = f.createURI(exns, &quot;alice1&quot;);<br> URI carroll = f.createURI(exns, &quot;carroll&quot;);<br> URI persontype = f.createURI(exns, &quot;Person&quot;);<br> URI fullname = f.createURI(exns, &quot;fullname&quot;); <br> Literal alicename = f.createLiteral(&quot;Alice B. Toklas&quot;);<br> Literal lewisCarroll = f.createLiteral(&quot;Lewis Carroll&quot;);<br> // Create parts of book resources.<br> URI book = f.createURI(exns, &quot;book1&quot;);<br> URI booktype = f.createURI(exns, &quot;Book&quot;);<br> URI booktitle = f.createURI(exns, &quot;title&quot;);<br> URI author = f.createURI(exns, &quot;author&quot;);<br> Literal wonderland = f.createLiteral(&quot;Alice in Wonderland&quot;);</pre>
<p>Clear the repository, so our new triples are the only ones available. </p>
<pre class="input">
conn.clear() </pre>
<p>Add the resource for the new person, Alice B. Toklas: </p>
<pre class="input"> conn.add(alice, RDF.TYPE, persontype);<br> conn.add(alice, fullname, alicename);</pre>
<p>Add the new book, <em>Alice in Wonderland</em>. </p>
<pre class="input"> conn.add(book, RDF.TYPE, booktype); <br> conn.add(book, booktitle, wonderland); </pre>
<pre class="input"> conn.add(book, RDF.TYPE, booktype); <br> conn.add(book, booktitle, wonderland);
conn.add(book, author, carroll);</pre>
<p>Note that the book's author predicate links to the resource URI of the Lewis Carroll author resource. The author is a Person with a fullname:</p>
<pre class="input"> conn.add(carroll, RDF.TYPE, persontype);<br> conn.add(carroll, fullname, lewisCarroll);</pre>
<p>Let's use getStatements() to retrieve the new triples and inspect them:</p>
<pre class="output">(http://example.org/people/alice1, http://www.w3.org/1999/02/22-rdf-syntax-ns#type, http://example.org/people/Person) [null]<br>(http://example.org/people/alice1, http://example.org/people/fullname, &quot;Alice B. Toklas&quot;) [null]
<br>(http://example.org/people/book1, http://www.w3.org/1999/02/22-rdf-syntax-ns#type, http://example.org/people/Book) [null]<br>(http://example.org/people/book1, http://example.org/people/title, &quot;Alice in Wonderland&quot;) [null]<br>(http://example.org/people/book1, http://example.org/people/author, http://example.org/people/carroll) [null]
<br>(http://example.org/people/carroll, http://www.w3.org/1999/02/22-rdf-syntax-ns#type, http://example.org/people/Person) [null]<br>(http://example.org/people/carroll, http://example.org/people/fullname, &quot;Lewis Carroll&quot;) [null]</pre>
<p>Now we set up the SPARQL query that looks for triples containing &quot;Alice&quot; in the object position. </p>
<p>The text match occurs through a &quot;magic&quot; predicate called <strong>fti:match</strong>. This is not an RDF &quot;predicate&quot; but a LISP &quot;predicate,&quot; meaning that it behaves as a true/false test. This predicate has two arguments. One is the subject URI of the resources to search. The other is the string pattern to search for, such as &quot;Alice&quot;. Only registered text predicates will be searched. Only full-word matches will be found. </p>
<pre class="input"> String queryString = <br> &quot;SELECT ?s ?p ?o &quot; +<br> &quot;WHERE { ?s ?p ?o . ?s fti:match 'Alice' . }&quot;;</pre>
<pre class="input"> String queryString = <br> &quot;SELECT ?s ?p ?o &quot; +<br> &quot;WHERE { ?s ?p ?o .
?s fti:match 'Alice' . }&quot;;</pre>
<p>There is no need to include a prefix declaration for the 'fti' nickname. That is because 'fti' is included among the built-in namespace/nickname mappings in AllegroGraph.</p>
<p>When we execute our SPARQL query, it matches the "Alice" within the literal "Alice B. Toklas" because that literal occurs in a triple having the registered <strong>fullname</strong> predicate, but it does not match the "Alice" in the literal "Alice in Wonderland" because the <strong>booktitle</strong> predicate was not registered for text indexing. This query returns <em>all triples</em> of a resource that had a successful match in at least one object value. </p>
<pre class="input"> TupleQuery tupleQuery = conn.prepareTupleQuery(QueryLanguage.SPARQL, queryString);<br> TupleQueryResult result = (TupleQueryResult)tupleQuery.evaluate();<br> int count = 0;<br> while (result.hasNext()) {<br> BindingSet bindingSet = result.next();<br> if (count &lt; 5) {<br> println(bindingSet);<br> }<br> count += 1;<br> }</pre>
Expand Down

0 comments on commit 711b6db

Please sign in to comment.