Merge pull request #66 from veewee/reader-improvements

Add reader MatchingNode results and a signal to stop reading

Author: veewee

docs/reader.md

@@ -7,21 +7,31 @@ As a result, the reader provides a generator of XML strings that match your matc
 ## Example
 
 ```php
-use VeeWee\Xml\Dom\Document;
+use VeeWee\Xml\Dom\Configurator;
 use VeeWee\Xml\Reader\Reader;
+use VeeWee\Xml\Reader\Signal;
 use VeeWee\Xml\Reader\Matcher;
 
 $reader   = Reader::fromXmlFile('large-data.xml');
 $provider = $reader->provide(
-    Matcher\all(
+    $matcher = Matcher\all(
         Matcher\node_name('item'),
         Matcher\node_attribute('locale', 'nl-BE')
-    )
+    ),
+    // Optionally, you can provide a signal to stop reading at a given point:
+    $signal = new Signal()
 );
 
 foreach ($provider as $nlItem) {
-    $dom = Document::fromXmlString($nlItem);
     // Do something with it
+    $xml = $nlItem->xml();
+    $dom = $nlItem->intoDocument(Configurator\canonicalize());
+    $decoded = $nlItem->decode(Configurator\canonicalize());
+    $matched = $nlItem->matches($matcher);
+    $sequence = $nlItem->nodeSequence();
+    
+    // If you have loaded sufficient items, you can stop reading the XML file:
+    $signal->stop();
 }
 ```
 
@@ -54,7 +64,8 @@ The reader will keep only small parts of the XML in memory by reading the XML st
 When the reader detects the first `breakfast_menu` element, it will ask the provided matchers if you are interested in this tag.
 A matcher is a function that returns `true` when interested or `false` when it is not interested in this element.
 When the matcher returns `true`, the reader will read the complete outer XML of current tag and `yield` this matching XML to your logic.
-This means that the memory-safety of YOUR reader is based on the part inside the XML you are interested in:
+This XML is wrapped in a `MatchingNode` which also contains the `NodeSequence` and some handy shortcut functions to e.g. convert the XML into a DOM Document.
+Do note that, the memory-safety of YOUR reader is based on the part inside the XML you are interested in:
 If you only match on the root node, it will yield the complete XML and therefore won't be memory-safe.
 
 After deciding if you are interested in the previous tag, it jumps over to the next tag: `breakfast_menu > food[position() = 1 AND @soldOUt=false AND @bestSeller = true]` and asks the matcher if you are interested in this.

src/Xml/Reader/MatchingNode.php

@@ -0,0 +1,65 @@
+<?php
+declare(strict_types=1);
+
+namespace VeeWee\Xml\Reader;
+
+use DOMDocument;
+use VeeWee\Xml\Dom\Document;
+use VeeWee\Xml\Encoding\Exception\EncodingException;
+use VeeWee\Xml\Exception\RuntimeException;
+use VeeWee\Xml\Reader\Node\NodeSequence;
+use function VeeWee\Xml\Encoding\xml_decode;
+
+final class MatchingNode
+{
+    /**
+     * @param non-empty-string $xml
+     */
+    public function __construct(
+        private readonly string $xml,
+        private readonly NodeSequence $nodeSequence
+    ) {
+    }
+
+    /**
+     * @return non-empty-string
+     */
+    public function xml(): string
+    {
+        return $this->xml;
+    }
+
+    public function nodeSequence(): NodeSequence
+    {
+        return $this->nodeSequence;
+    }
+
+    /**
+     * @param list<callable(DOMDocument): DOMDocument> $configurators
+     *
+     * @throws RuntimeException
+     */
+    public function intoDocument(callable ... $configurators): Document
+    {
+        return Document::fromXmlString($this->xml, ...$configurators);
+    }
+
+    /**
+     * @param list<callable(DOMDocument): DOMDocument> $configurators
+     *
+     * @throws RuntimeException
+     * @throws EncodingException
+     */
+    public function decode(callable ... $configurators): array
+    {
+        return xml_decode($this->xml, ...$configurators);
+    }
+
+    /**
+     * @param callable(NodeSequence): bool $matcher
+     */
+    public function matches(callable $matcher): bool
+    {
+        return $matcher($this->nodeSequence);
+    }
+}

src/Xml/Reader/Reader.php

@@ -60,18 +60,25 @@ public static function fromXmlString(string $xml, callable ... $configurators):
     /**
      * @param callable(NodeSequence): bool $matcher
      *
-     * @return Generator<string>
+     * @return Generator<MatchingNode>
      *
      * @throws RuntimeException
      */
-    public function provide(callable $matcher): Generator
+    public function provide(callable $matcher, ?Signal $signal = null): Generator
     {
+        $signal ??= new Signal();
         $reader = ($this->factory)();
         $pointer = Pointer::create();
 
         yield from stop_on_first_issue(
-            static fn (): bool => $reader->read(),
-            static function () use ($reader, $pointer, $matcher) : ?string {
+            static function () use ($reader, $signal): bool {
+                if($signal->stopRequested()) {
+                    return !$reader->close();
+                }
+
+                return $reader->read();
+            },
+            static function () use ($reader, $pointer, $matcher) : ?MatchingNode {
                 if ($reader->nodeType === XMLReader::END_ELEMENT) {
                     $pointer->leaveElement();
 
@@ -93,13 +100,14 @@ static function () use ($reader): array {
                     );
 
                     $pointer->enterElement($element);
-                    $result = $matcher($pointer->getNodeSequence()) ? $reader->readOuterXml() : null;
+                    $outerXml = $matcher($pointer->getNodeSequence()) ? $reader->readOuterXml() : null;
+                    $match = $outerXml ? new MatchingNode($outerXml, $pointer->getNodeSequence()) : null;
 
                     if ($isEmptyElement) {
                         $pointer->leaveElement();
                     }
 
-                    return $result;
+                    return $match;
                 }
 
                 return null;

src/Xml/Reader/Signal.php

@@ -0,0 +1,19 @@
+<?php
+declare(strict_types=1);
+
+namespace VeeWee\Xml\Reader;
+
+final class Signal
+{
+    private bool $stopRequested = false;
+
+    public function stop(): void
+    {
+        $this->stopRequested = true;
+    }
+
+    public function stopRequested(): bool
+    {
+        return $this->stopRequested;
+    }
+}

tests/Xml/Reader/Configurator/SubstituteEntitiesTest.php

@@ -5,7 +5,9 @@
 namespace VeeWee\Tests\Xml\Reader\Configurator;
 
 use PHPUnit\Framework\TestCase;
+use VeeWee\Xml\Reader\MatchingNode;
 use VeeWee\Xml\Reader\Reader;
+use function Psl\Vec\map;
 use function VeeWee\Xml\Reader\Configurator\substitute_entities;
 use function VeeWee\Xml\Reader\Matcher\node_name;
 
@@ -21,11 +23,11 @@ public function test_it_can_substitute_entities(): void
             [
                 '<user>my entity value</user>',
             ],
-            [...$iterator]
+            map($iterator, static fn (MatchingNode $match): string => $match->xml())
         );
     }
 
-    
+
     public function test_it_can_skip_substituting_entities(): void
     {
         $xml = $this->buildXml();
@@ -36,7 +38,7 @@ public function test_it_can_skip_substituting_entities(): void
             [
                 '<user>&entity;</user>',
             ],
-            [...$iterator]
+            map($iterator, static fn (MatchingNode $match): string => $match->xml())
         );
     }
 

tests/Xml/Reader/Configurator/XsdSchemaTest.php

@@ -7,16 +7,18 @@
 use PHPUnit\Framework\TestCase;
 use VeeWee\Tests\Xml\Helper\FillFileTrait;
 use VeeWee\Xml\Exception\RuntimeException;
+use VeeWee\Xml\Reader\MatchingNode;
 use VeeWee\Xml\Reader\Reader;
 use XMLReader;
+use function Psl\Vec\map;
 use function VeeWee\Xml\Reader\Configurator\xsd_schema;
 use function VeeWee\Xml\Reader\Matcher\node_name;
 
 final class XsdSchemaTest extends TestCase
 {
     use FillFileTrait;
 
-    
+
     public function test_it_can_iterate_if_the_schema_matches(): void
     {
         [$xsdFile, $xsdHandle] = $this->createXsdFile();
@@ -37,13 +39,13 @@ public function test_it_can_iterate_if_the_schema_matches(): void
                 '<user>Bos</user>',
                 '<user>Mos</user>'
             ],
-            [...$iterator]
+            map($iterator, static fn (MatchingNode $match): string => $match->xml())
         );
 
         fclose($xsdHandle);
     }
 
-    
+
     public function test_it_triggers_an_error_on_invalid_schema(): void
     {
         [$xsdFile, $xsdHandle] = $this->createXsdFile();
@@ -65,7 +67,7 @@ public function test_it_triggers_an_error_on_invalid_schema(): void
         fclose($xsdHandle);
     }
 
-    
+
     public function test_it_triggers_an_error_if_schema_file_does_not_exist(): void
     {
         $xml = '<root />';
@@ -80,7 +82,7 @@ public function test_it_triggers_an_error_if_schema_file_does_not_exist(): void
         fclose($xsdHandle);
     }
 
-    
+
     public function test_it_can_not_set_a_schema_if_the_reader_started_reading(): void
     {
         [$xsdFile, $xsdHandle] = $this->createXsdFile();
@@ -93,7 +95,7 @@ public function test_it_can_not_set_a_schema_if_the_reader_started_reading(): vo
         fclose($xsdHandle);
     }
 
-    
+
     public function test_it_can_not_set_a_schema_if_the_schema_is_invalid(): void
     {
         [$xsdFile, $xsdHandle] = $this->fillFile('invalid schema');

tests/Xml/Reader/Matcher/AbstractMatcherTest.php

@@ -6,8 +6,10 @@
 use Closure;
 use Generator;
 use PHPUnit\Framework\TestCase;
+use VeeWee\Xml\Reader\MatchingNode;
 use VeeWee\Xml\Reader\Node\NodeSequence;
 use VeeWee\Xml\Reader\Reader;
+use function Psl\Vec\map;
 
 abstract class AbstractMatcherTest extends TestCase
 {
@@ -23,7 +25,7 @@ abstract public static function provideMatcherCases(): Generator;
     public function test_real_xml_cases(Closure $matcher, string $xml, array $expected)
     {
         $reader = Reader::fromXmlString($xml);
-        $actual = [...$reader->provide($matcher)];
+        $actual = map($reader->provide($matcher), static fn (MatchingNode $match): string => $match->xml());
 
         static::assertSame($actual, $expected);
     }

tests/Xml/Reader/MatchingNodeTest.php

@@ -0,0 +1,73 @@
+<?php
+declare(strict_types=1);
+
+namespace VeeWee\Tests\Xml\Reader;
+
+use PHPUnit\Framework\TestCase;
+use VeeWee\Xml\Reader\MatchingNode;
+use VeeWee\Xml\Reader\Node\ElementNode;
+use VeeWee\Xml\Reader\Node\NodeSequence;
+use function Psl\Fun\identity;
+use function VeeWee\Xml\Dom\Locator\document_element;
+use function VeeWee\Xml\Dom\Mapper\xml_string;
+use function VeeWee\Xml\Reader\Matcher\element_name;
+
+final class MatchingNodeTest extends TestCase
+{
+
+    public function test_it_is_a_matching_node(): void
+    {
+        $match = new MatchingNode(
+            $xml = '<hello/>',
+            $sequence = new NodeSequence(
+                new ElementNode(1, 'hello', 'hello', '', '', [])
+            )
+        );
+
+        static::assertSame($xml, $match->xml());
+        static::assertSame($sequence, $match->nodeSequence());
+    }
+
+
+    public function test_it_can_match(): void
+    {
+        $match = new MatchingNode(
+            '<hello/>',
+            new NodeSequence(
+                new ElementNode(1, 'hello', 'hello', '', '', [])
+            )
+        );
+
+        static::assertTrue($match->matches(element_name('hello')));
+        static::assertFalse($match->matches(element_name('world')));
+    }
+
+
+    public function test_it_can_transform_into_a_dom_document(): void
+    {
+        $match = new MatchingNode(
+            $xml = '<hello/>',
+            new NodeSequence(
+                new ElementNode(1, 'hello', 'hello', '', '', [])
+            )
+        );
+
+        $document = $match->intoDocument(identity());
+
+        static::assertSame($xml, xml_string()($document->map(document_element())));
+    }
+
+    public function test_it_can_decode_the_xml(): void
+    {
+        $match = new MatchingNode(
+            $xml = '<hello/>',
+            new NodeSequence(
+                new ElementNode(1, 'hello', 'hello', '', '', [])
+            )
+        );
+
+        $decoded = $match->decode(identity());
+
+        static::assertSame(['hello' => ''], $decoded);
+    }
+}