svn commit: r1178389 - in /tika/trunk/tika-parsers/src/test: java/org/apache/tika/TestParsers.java java/org/apache/tika/parser/pdf/PDFParserTest.java resources/test-documents/testAnnotations.pdf resources/test-documents/testComment2.pdf

[mi...@apache.org]

Author: mikemccand
Date: Mon Oct  3 12:36:03 2011
New Revision: 1178389

URL: http://svn.apache.org/viewvc?rev=1178389&view=rev
Log:
TIKA-738: move (ignored) test case to PDFParserTest

Added:
    tika/trunk/tika-parsers/src/test/resources/test-documents/testAnnotations.pdf
      - copied unchanged from r1178250, tika/trunk/tika-parsers/src/test/resources/test-documents/testComment2.pdf
Removed:
    tika/trunk/tika-parsers/src/test/resources/test-documents/testComment2.pdf
Modified:
    tika/trunk/tika-parsers/src/test/java/org/apache/tika/TestParsers.java
    tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java

Modified: tika/trunk/tika-parsers/src/test/java/org/apache/tika/TestParsers.java
URL: http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/test/java/org/apache/tika/TestParsers.java?rev=1178389&r1=1178388&r2=1178389&view=diff
==============================================================================
--- tika/trunk/tika-parsers/src/test/java/org/apache/tika/TestParsers.java (original)
+++ tika/trunk/tika-parsers/src/test/java/org/apache/tika/TestParsers.java Mon Oct  3 12:36:03 2011
@@ -191,10 +191,6 @@ public class TestParsers extends TikaTes
         final String[] extensions = new String[] {"ppt", "pptx", "doc", "docx", "pdf", "rtf"};
         for(String extension : extensions) {
             verifyComment(extension, "testComment");
-            // TIKA-717: re-enable once we fix this:
-            //if (extension.equals("pdf")) {
-            //verifyComment(extension, "testComment2");
-            //}
         }
     }
 }

Modified: tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java
URL: http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java?rev=1178389&r1=1178388&r2=1178389&view=diff
==============================================================================
--- tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java (original)
+++ tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java Mon Oct  3 12:36:03 2011
@@ -212,4 +212,22 @@ public class PDFParserTest extends TikaT
         // chars, so we cannot test this here:
         //assertContains("\uD800\uDF32\uD800\uDF3f\uD800\uDF44\uD800\uDF39\uD800\uDF43\uD800\uDF3A", content);
     }
+
+    // TIKA-738: re-enable this
+    public void IGNOREtestAnnotations() throws Exception {
+        Parser parser = new AutoDetectParser(); // Should auto-detect!
+        ContentHandler handler = new BodyContentHandler();
+        Metadata metadata = new Metadata();
+        ParseContext context = new ParseContext();
+        InputStream stream = getResourceAsStream("/test-documents/testAnnotations.pdf");
+        try {
+            parser.parse(stream, handler, metadata, context);
+        } finally {
+            stream.close();
+        }
+        String content = handler.toString();
+        content = content.replaceAll("[\\s\u00a0]+"," ");
+        assertContains("Here is some text", content);
+        assertContains("Here is a comment", content);
+    }
 }