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 2020/02/12 17:01:15 UTC
[tika] branch master updated: TIKA-3041 -- keep inline image names
the same as they were.
This is an automated email from the ASF dual-hosted git repository.
tallison pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/tika.git
The following commit(s) were added to refs/heads/master by this push:
new 9c855c9 TIKA-3041 -- keep inline image names the same as they were.
9c855c9 is described below
commit 9c855c94f57d0a18352aec319f698b479f394c25
Author: tallison <ta...@apache.org>
AuthorDate: Wed Feb 12 12:00:55 2020 -0500
TIKA-3041 -- keep inline image names the same as they were.
---
.../java/org/apache/tika/parser/pdf/ImageGraphicsEngine.java | 2 +-
.../src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java | 9 ++++-----
2 files changed, 5 insertions(+), 6 deletions(-)
diff --git a/tika-parsers/src/main/java/org/apache/tika/parser/pdf/ImageGraphicsEngine.java b/tika-parsers/src/main/java/org/apache/tika/parser/pdf/ImageGraphicsEngine.java
index 96d8302..b660560 100644
--- a/tika-parsers/src/main/java/org/apache/tika/parser/pdf/ImageGraphicsEngine.java
+++ b/tika-parsers/src/main/java/org/apache/tika/parser/pdf/ImageGraphicsEngine.java
@@ -265,7 +265,7 @@ class ImageGraphicsEngine extends PDFGraphicsStreamEngine {
//this is the metadata for this particular image
Metadata metadata = new Metadata();
String suffix = getSuffix(image, metadata);
- String fileName = "image-" + imageNumber + "." + suffix;
+ String fileName = "image" + imageNumber + "." + suffix;
AttributesImpl attr = new AttributesImpl();
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 ce8df1b..3717a3f 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
@@ -654,7 +654,7 @@ public class PDFParserTest extends TikaTest {
assertEquals(5, metadatas.size());
assertNull(metadatas.get(0).get(TikaCoreProperties.RESOURCE_NAME_KEY));
- assertEquals("image-0.jpg", metadatas.get(1).get(TikaCoreProperties.RESOURCE_NAME_KEY));
+ assertEquals("image0.jpg", metadatas.get(1).get(TikaCoreProperties.RESOURCE_NAME_KEY));
assertEquals("Press Quality(1).joboptions", metadatas.get(3).get(TikaCoreProperties.RESOURCE_NAME_KEY));
assertEquals("Unit10.doc", metadatas.get(4).get(TikaCoreProperties.RESOURCE_NAME_KEY));
assertEquals(MediaType.image("jpeg").toString(), metadatas.get(1).get(Metadata.CONTENT_TYPE));
@@ -674,7 +674,6 @@ public class PDFParserTest extends TikaTest {
List<Metadata> metadatas = getRecursiveMetadata("testPDF_JBIG2.pdf", context);
- debug(metadatas);
assertEquals(2, metadatas.size());
assertContains("test images compressed using JBIG2", metadatas.get(0).get(AbstractRecursiveParserWrapperHandler.TIKA_CONTENT));
@@ -687,7 +686,7 @@ public class PDFParserTest extends TikaTest {
assertEquals("Invalid width.", "352", metadatas.get(1).get("width"));
assertNull(metadatas.get(0).get(TikaCoreProperties.RESOURCE_NAME_KEY));
- assertEquals("image-0.jb2",
+ assertEquals("image0.jb2",
metadatas.get(1).get(TikaCoreProperties.RESOURCE_NAME_KEY));
assertEquals(MediaType.image("x-jbig2").toString(),
metadatas.get(1).get(Metadata.CONTENT_TYPE));
@@ -931,7 +930,7 @@ public class PDFParserTest extends TikaTest {
//regular attachment
assertContains("<div source=\"attachment\" class=\"embedded\" id=\"Unit10.doc\" />", r.xml);
//inline image
- assertContains("<img src=\"embedded:image-1.tif\" alt=\"image-1.tif\" />", r.xml);
+ assertContains("<img src=\"embedded:image1.tif\" alt=\"image1.tif\" />", r.xml);
//doc embedded inside an annotation
r = getXML("testPDFFileEmbInAnnotation.pdf");
@@ -1476,7 +1475,7 @@ public class PDFParserTest extends TikaTest {
List<Metadata> metadataList = getRecursiveMetadata("testPDF_PDFBOX-52.pdf", context);
int max = 0;
- Matcher matcher = Pattern.compile("image-(\\d+)").matcher("");
+ Matcher matcher = Pattern.compile("image(\\d+)").matcher("");
for (Metadata m : metadataList) {
String n = m.get(TikaCoreProperties.RESOURCE_NAME_KEY);