FOP-2879: Add caching to avoid parsing content stream
git-svn-id: https://svn.apache.org/repos/asf/xmlgraphics/fop-pdf-images/trunk@1864947 13f79535-47bb-0310-9956-ffa450edef68
diff --git a/src/java/org/apache/fop/render/pdf/pdfbox/PDFBoxAdapter.java b/src/java/org/apache/fop/render/pdf/pdfbox/PDFBoxAdapter.java
index a49fd89..55cf5af 100644
--- a/src/java/org/apache/fop/render/pdf/pdfbox/PDFBoxAdapter.java
+++ b/src/java/org/apache/fop/render/pdf/pdfbox/PDFBoxAdapter.java
@@ -369,10 +369,12 @@
// }
}
if (newStream == null) {
- PDFWriter writer = new PDFWriter(uniqueName, currentMCID);
- newStream = writer.writeText(pdStream);
- currentMCID = writer.getCurrentMCID();
-
+ newStream = (String) clonedVersion.get(key);
+ if (newStream == null) {
+ PDFWriter writer = new PDFWriter(uniqueName, currentMCID);
+ newStream = writer.writeText(pdStream);
+ clonedVersion.put(key, newStream);
+ }
}
pdStream = new PDStream(sourceDoc, new ByteArrayInputStream(newStream.getBytes("ISO-8859-1")));
mergeXObj(sourcePageResources, fontinfo, uniqueName);
diff --git a/src/java/org/apache/fop/render/pdf/pdfbox/PDFWriter.java b/src/java/org/apache/fop/render/pdf/pdfbox/PDFWriter.java
index 34ffb72..a375ad1 100644
--- a/src/java/org/apache/fop/render/pdf/pdfbox/PDFWriter.java
+++ b/src/java/org/apache/fop/render/pdf/pdfbox/PDFWriter.java
@@ -149,8 +149,4 @@
dictArgs.add(cn.getKey());
dictArgs.add(updatedID);
}
-
- protected int getCurrentMCID() {
- return currentMCID;
- }
}
diff --git a/test/java/org/apache/fop/render/pdf/PDFBoxAdapterTestCase.java b/test/java/org/apache/fop/render/pdf/PDFBoxAdapterTestCase.java
index 09c3f95..cb78dd2 100644
--- a/test/java/org/apache/fop/render/pdf/PDFBoxAdapterTestCase.java
+++ b/test/java/org/apache/fop/render/pdf/PDFBoxAdapterTestCase.java
@@ -30,6 +30,7 @@
import java.net.URI;
import java.util.Collections;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.LinkedHashMap;
import java.util.Map;
@@ -478,55 +479,60 @@
@Test
public void testPDFCache() throws IOException {
- PDFDocument pdfdoc = new PDFDocument("");
- Map<Object, Object> pdfCache = new LinkedHashMap<Object, Object>();
- Map<Object, Object> objectCachePerFile = loadPDFWithCache(pdfdoc, pdfCache, LOOP);
+ LoadPDFWithCache loadPDFWithCache = new LoadPDFWithCache();
+ loadPDFWithCache.run(LOOP);
- Object item = pdfCache.values().iterator().next();
+ Object item = loadPDFWithCache.pdfCache.values().iterator().next();
Assert.assertEquals(item.getClass(), PDFStream.class);
- item = pdfCache.keySet().iterator().next();
+ item = loadPDFWithCache.pdfCache.keySet().iterator().next();
Assert.assertEquals(item.getClass(), Integer.class);
- Assert.assertEquals(pdfCache.size(), 12);
+ Assert.assertEquals(loadPDFWithCache.pdfCache.size(), 12);
- int pdfDictionary = 0;
- int strings = 0;
- for (Map.Entry<Object, Object> o : objectCachePerFile.entrySet()) {
- if (o.getValue().getClass().equals(PDFDictionary.class)) {
- pdfDictionary++;
- }
- if (o.getKey() instanceof String) {
- strings++;
- }
- }
- Assert.assertEquals(pdfDictionary, 26);
- Assert.assertEquals(strings, 33);
- Assert.assertEquals(objectCachePerFile.size(), 45);
+ Iterator<Object> iterator = loadPDFWithCache.objectCachePerFile.values().iterator();
+ iterator.next();
+ item = iterator.next();
+ Assert.assertEquals(item.getClass(), PDFDictionary.class);
+ item = loadPDFWithCache.objectCachePerFile.keySet().iterator().next();
+ Assert.assertEquals(item.getClass(), String.class);
+ Assert.assertEquals(loadPDFWithCache.objectCachePerFile.size(), 46);
}
- private Map<Object, Object> loadPDFWithCache(PDFDocument pdfdoc, Map<Object, Object> pdfCache, String pdf)
- throws IOException {
- Map<Object, Object> objectCachePerFile = new LinkedHashMap<Object, Object>();
- PDFPage pdfpage = getPDFPage(pdfdoc);
- pdfdoc.assignObjectNumber(pdfpage);
- pdfpage.setDocument(pdfdoc);
- PDFBoxAdapter adapter = new PDFBoxAdapter(
- pdfpage, objectCachePerFile, new HashMap<Integer, PDFArray>(), pdfCache);
- PDDocument doc = PDDocument.load(new File(pdf));
- PDPage page = doc.getPage(0);
- adapter.createStreamFromPDFBoxPage(doc, page, "key", new AffineTransform(), null, new Rectangle());
- doc.close();
- return objectCachePerFile;
+ @Test
+ public void testPDFCache2() throws IOException {
+ LoadPDFWithCache loadPDFWithCache = new LoadPDFWithCache();
+ String stream = loadPDFWithCache.run(LOOP);
+ String cachedStream = (String) loadPDFWithCache.objectCachePerFile.get(LOOP);
+ Assert.assertTrue(cachedStream.contains("EMC"));
+ Assert.assertTrue(stream.endsWith(cachedStream));
+ }
+
+ private static class LoadPDFWithCache {
+ private PDFDocument pdfdoc = new PDFDocument("");
+ private Map<Object, Object> pdfCache = new LinkedHashMap<Object, Object>();
+ private Map<Object, Object> objectCachePerFile = new LinkedHashMap<Object, Object>();
+ private String run(String pdf) throws IOException {
+ PDFPage pdfpage = getPDFPage(pdfdoc);
+ pdfdoc.assignObjectNumber(pdfpage);
+ pdfpage.setDocument(pdfdoc);
+ PDFBoxAdapter adapter = new PDFBoxAdapter(
+ pdfpage, objectCachePerFile, new HashMap<Integer, PDFArray>(), pdfCache);
+ PDDocument doc = PDDocument.load(new File(pdf));
+ PDPage page = doc.getPage(0);
+ String stream = adapter.createStreamFromPDFBoxPage(
+ doc, page, pdf, new AffineTransform(), null, new Rectangle());
+ doc.close();
+ return stream;
+ }
}
@Test
public void testPDFSize() throws IOException {
- PDFDocument pdfdoc = new PDFDocument("");
- Map<Object, Object> pdfCache = new HashMap<Object, Object>();
- loadPDFWithCache(pdfdoc, pdfCache, ANNOT);
- loadPDFWithCache(pdfdoc, pdfCache, ANNOT);
+ LoadPDFWithCache loadPDFWithCache = new LoadPDFWithCache();
+ loadPDFWithCache.run(ANNOT);
+ loadPDFWithCache.run(ANNOT);
ByteArrayOutputStream bos = new ByteArrayOutputStream();
- pdfdoc.output(bos);
- Assert.assertEquals(pdfCache.size(), 2);
+ loadPDFWithCache.pdfdoc.output(bos);
+ Assert.assertEquals(loadPDFWithCache.pdfCache.size(), 2);
Assert.assertTrue(bos.size() <= 6418);
}
