You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tika.apache.org by ta...@apache.org on 2017/01/11 13:57:59 UTC
tika git commit: TIKA-2232 add unit test for OCR of jbig2 embedded in
PDF.
Repository: tika
Updated Branches:
refs/heads/master d1b1ad3d9 -> ba26f6ee0
TIKA-2232 add unit test for OCR of jbig2 embedded in PDF.
Project: http://git-wip-us.apache.org/repos/asf/tika/repo
Commit: http://git-wip-us.apache.org/repos/asf/tika/commit/ba26f6ee
Tree: http://git-wip-us.apache.org/repos/asf/tika/tree/ba26f6ee
Diff: http://git-wip-us.apache.org/repos/asf/tika/diff/ba26f6ee
Branch: refs/heads/master
Commit: ba26f6ee01574702f5eaa56bf45aaf06e043d6df
Parents: d1b1ad3
Author: tballison <ta...@mitre.org>
Authored: Wed Jan 11 08:57:51 2017 -0500
Committer: tballison <ta...@mitre.org>
Committed: Wed Jan 11 08:57:51 2017 -0500
----------------------------------------------------------------------
.../apache/tika/parser/image/ImageParserTest.java | 3 ++-
.../org/apache/tika/parser/pdf/PDFParserTest.java | 16 ++++++++++++++++
2 files changed, 18 insertions(+), 1 deletion(-)
----------------------------------------------------------------------
http://git-wip-us.apache.org/repos/asf/tika/blob/ba26f6ee/tika-parsers/src/test/java/org/apache/tika/parser/image/ImageParserTest.java
----------------------------------------------------------------------
diff --git a/tika-parsers/src/test/java/org/apache/tika/parser/image/ImageParserTest.java b/tika-parsers/src/test/java/org/apache/tika/parser/image/ImageParserTest.java
index 212b578..06625f6 100644
--- a/tika-parsers/src/test/java/org/apache/tika/parser/image/ImageParserTest.java
+++ b/tika-parsers/src/test/java/org/apache/tika/parser/image/ImageParserTest.java
@@ -20,6 +20,7 @@ import static org.junit.Assert.assertEquals;
import java.io.InputStream;
+import org.apache.tika.TikaTest;
import org.apache.tika.metadata.Metadata;
import org.apache.tika.metadata.TikaCoreProperties;
import org.apache.tika.parser.ParseContext;
@@ -27,7 +28,7 @@ import org.apache.tika.parser.Parser;
import org.junit.Test;
import org.xml.sax.helpers.DefaultHandler;
-public class ImageParserTest {
+public class ImageParserTest extends TikaTest {
private final Parser parser = new ImageParser();
http://git-wip-us.apache.org/repos/asf/tika/blob/ba26f6ee/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java
----------------------------------------------------------------------
diff --git a/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java b/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java
index 2d056c7..b017457 100644
--- a/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java
+++ b/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java
@@ -1255,6 +1255,22 @@ public class PDFParserTest extends TikaTest {
}
@Test
+ public void testJBIG2OCROnly() throws Exception {
+ if (!canRunOCR()) {
+ return;
+ }
+ PDFParserConfig config = new PDFParserConfig();
+ config.setOcrStrategy(PDFParserConfig.OCR_STRATEGY.OCR_ONLY);
+ ParseContext context = new ParseContext();
+ context.set(PDFParserConfig.class, config);
+ context.set(Parser.class, new AutoDetectParser());
+ //make sure everything works with regular xml _and_ with recursive
+ XMLResult xmlResult = getXML("testPDF_JBIG2.pdf", context);
+ assertContains("Norconex", xmlResult.xml);
+ }
+
+
+ @Test
public void testInitializationViaConfig() throws Exception {
InputStream is = getClass().getResourceAsStream("/org/apache/tika/parser/pdf/tika-config.xml");
assertNotNull(is);