API Using paths instead of File objects in extractors

Makes coupling to File objects optional, by choosing
to use the FileTextExtractable extension.

code/extensions/FileTextExtractable.php

@@ -27,10 +27,10 @@ class FileTextExtractable extends DataExtension {
 		if (!$forceParse && $this->owner->FileContentCache) return $this->owner->FileContentCache;
 
 		// Determine which extractor can process this file.
-		$extractor = FileTextExtractor::for_file($this->owner);
+		$extractor = FileTextExtractor::for_file($this->owner->FullPath);
 		if (!$extractor) return null;
 
-		$text = $extractor->getContent($this->owner);
+		$text = $extractor->getContent($this->owner->FullPath);
 		if (!$text) return null;
 
 		$this->owner->FileContentCache = $text;

code/extractors/FileTextExtractor.php

@@ -17,11 +17,11 @@ abstract class FileTextExtractor extends Object {
 	protected static $sorted_extractor_classes = null;
 
 	/**
-	 * @param  DataObject $file
+	 * @param  String $path
 	 * @return FileTextExtractor
 	 */
-	static function for_file($file) {
-		$extension = strtolower($file->getExtension());
+	static function for_file($path) {
+		$extension = pathinfo($path, PATHINFO_EXTENSION);
 
 		if (!self::$sorted_extractor_classes) {
 			// Generate the sorted list of extractors on demand.
@@ -41,6 +41,14 @@ abstract class FileTextExtractor extends Object {
 		}
 	}
 
+	/**
+	 * Checks if the extractor is supported on the current environment,
+	 * for example if the correct binaries or libraries are available.
+	 * 
+	 * @return boolean
+	 */
+	abstract function isAvailable();
+
 	/**
 	 * Return an array of content types that the extractor can handle.
 	 * @return unknown_type
@@ -48,11 +56,12 @@ abstract class FileTextExtractor extends Object {
 	abstract function supportedExtensions();
 
 	/**
-	 * Given a file object, extract the contents as text
-	 * @param $file
+	 * Given a file path, extract the contents as text.
+	 * 
+	 * @param $path
 	 * @return unknown_type
 	 */
-	abstract function getContent($file);
+	abstract function getContent($path);
 }
 
 ?>

code/extractors/HTMLTextExtractor.php

@@ -16,9 +16,8 @@ class HTMLTextExtractor extends FileTextExtractor {
 	 */
 	public static $priority = 10;
 
-	function getContent($file) {
-		$filename = Director::baseFolder() . "/" . $file->Filename;
-		$content = file_get_contents($filename);
+	function getContent($path) {
+		$content = file_get_contents($path);
 		return strip_tags($content);
 	}
 }

code/extractors/PDFTextExtractor.php

@@ -24,10 +24,9 @@ class PDFTextExtractor extends FileTextExtractor {
 		return ( $path ? $path . '/' : '' ) . $prog;
 	}
 	
-	function getContent($file) {
-		$filename = Director::baseFolder() . "/" . $file->Filename;
-		if (!$filename) return ""; // no file
-		$content = `{$this->bin('pdftotext')} "$filename" -`;
+	function getContent($path) {
+		if (!$path) return ""; // no file
+		$content = `{$this->bin('pdftotext')} "$path" -`;
 		return $content;
 	}
 }