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 2026/09/03 00:43:09 UTC
(tika) 01/01: add a placeholder marker
This is an automated email from the ASF dual-hosted git repository.
tballison pushed a commit to branch placeholder-stream
in repository https://gitbox.apache.org/repos/asf/tika.git
commit 92bd7cecf4eac8e22f800182fb565ccfdd8d1011
Author: tallison <ta...@apache.org>
AuthorDate: Wed Sep 2 20:42:55 2026 -0400
add a placeholder marker
---
CHANGES.txt | 5 +++++
.../src/main/java/org/apache/tika/io/TikaInputSource.java | 8 ++++++++
.../src/main/java/org/apache/tika/io/TikaInputStream.java | 14 +++++++++++++-
.../main/java/org/apache/tika/renderer/RenderResult.java | 3 ++-
.../apache/tika/parser/microsoft/MetafileRendering.java | 2 +-
.../org/apache/tika/parser/microsoft/OfficeParser.java | 11 ++++++++---
.../java/org/apache/tika/parser/pdf/AbstractPDF2XHTML.java | 2 +-
.../apache/tika/parser/pdf/image/ImageGraphicsEngine.java | 2 +-
8 files changed, 39 insertions(+), 8 deletions(-)
diff --git a/CHANGES.txt b/CHANGES.txt
index 80e898aa72..54605e07c1 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -1,5 +1,10 @@
Release 4.1.0 - unreleased
+ * Placeholder streams -- the empty stand-ins parsers hand parseEmbedded
+ for content that is never extracted -- report an unknown length rather
+ than their own zero, and the macro-failure entry is registered without
+ parsing its sentinel (TIKA-4874).
+
* Raster previews for the vector thumbnails of Office documents: the new
poi-metafile-renderer draws EMF and WMF images through POI (a PNG of
a configurable width; Word's bitmap-in-WMF thumbnails from the bitmap
diff --git a/tika-core/src/main/java/org/apache/tika/io/TikaInputSource.java b/tika-core/src/main/java/org/apache/tika/io/TikaInputSource.java
index 9e7b4a8d30..78500097c8 100644
--- a/tika-core/src/main/java/org/apache/tika/io/TikaInputSource.java
+++ b/tika-core/src/main/java/org/apache/tika/io/TikaInputSource.java
@@ -63,6 +63,14 @@ interface TikaInputSource extends Closeable {
*/
long getLength();
+ /**
+ * Whether this source stands in for content that is never extracted. Spooling one
+ * measures nothing, so its unknown length must not cost a temp file to confirm.
+ */
+ default boolean isPlaceholder() {
+ return false;
+ }
+
/**
* Enables full rewind capability.
* <p>
diff --git a/tika-core/src/main/java/org/apache/tika/io/TikaInputStream.java b/tika-core/src/main/java/org/apache/tika/io/TikaInputStream.java
index b58cd548db..670e2581dd 100644
--- a/tika-core/src/main/java/org/apache/tika/io/TikaInputStream.java
+++ b/tika-core/src/main/java/org/apache/tika/io/TikaInputStream.java
@@ -158,6 +158,18 @@ public class TikaInputStream extends TaggedInputStream {
return new TikaInputStream(inputSource, tmp, ext);
}
+ /**
+ * An empty stream standing in for content that is never extracted -- a metadata-only
+ * entry, a rendering carried as an open container. It reports an <em>unknown</em>
+ * length, so nothing mistakes the placeholder's size for the document's. Pair it with
+ * {@link org.apache.tika.parser.MetadataOnlyParse} to register an entry without
+ * parsing it, unless an open container supplies the content.
+ */
+ public static TikaInputStream getPlaceholder() {
+ TemporaryResources tmp = new TemporaryResources();
+ return new TikaInputStream(new PlaceholderSource(tmp), tmp, "");
+ }
+
public static TikaInputStream get(Path path) throws IOException {
return get(path, new Metadata());
}
@@ -453,7 +465,7 @@ public class TikaInputStream extends TaggedInputStream {
return -1;
}
long len = source.getLength();
- if (len == -1) {
+ if (len == -1 && !source.isPlaceholder()) {
// Force spill to get length
getPath();
len = source.getLength();
diff --git a/tika-core/src/main/java/org/apache/tika/renderer/RenderResult.java b/tika-core/src/main/java/org/apache/tika/renderer/RenderResult.java
index 25588c45bb..db6ab42bdf 100644
--- a/tika-core/src/main/java/org/apache/tika/renderer/RenderResult.java
+++ b/tika-core/src/main/java/org/apache/tika/renderer/RenderResult.java
@@ -65,7 +65,8 @@ public class RenderResult implements Closeable {
if (result instanceof Path) {
return TikaInputStream.get((Path)result, metadata);
} else {
- TikaInputStream tis = TikaInputStream.get(new byte[0]);
+ // the rendering rides in the open container, not the stream
+ TikaInputStream tis = TikaInputStream.getPlaceholder();
tis.setOpenContainer(result);
return tis;
}
diff --git a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/MetafileRendering.java b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/MetafileRendering.java
index d4efd0a12c..0032c97a64 100644
--- a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/MetafileRendering.java
+++ b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/MetafileRendering.java
@@ -119,7 +119,7 @@ final class MetafileRendering {
if (source != null && source.hasFile()) {
return TikaInputStream.get(source.getPath());
}
- return TikaInputStream.get(new byte[0]);
+ return TikaInputStream.getPlaceholder();
}
/**
diff --git a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/OfficeParser.java b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/OfficeParser.java
index 423a51f86b..925e688f58 100644
--- a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/OfficeParser.java
+++ b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/OfficeParser.java
@@ -55,6 +55,7 @@ import org.apache.tika.metadata.HttpHeaders;
import org.apache.tika.metadata.Metadata;
import org.apache.tika.metadata.TikaCoreProperties;
import org.apache.tika.mime.MediaType;
+import org.apache.tika.parser.MetadataOnlyParse;
import org.apache.tika.parser.ParseContext;
import org.apache.tika.parser.PasswordProvider;
import org.apache.tika.parser.microsoft.ooxml.OOXMLParser;
@@ -130,9 +131,13 @@ public class OfficeParser extends AbstractOfficeParser {
m.set(HttpHeaders.CONTENT_TYPE, "text/x-vbasic");
EmbeddedDocumentUtil.recordException(e, m, context);
if (embeddedDocumentExtractor.shouldParseEmbedded(m, context)) {
- embeddedDocumentExtractor.parseEmbedded(
- //pass in space character so that we don't trigger a zero-byte exception
- TikaInputStream.get(new byte[]{'\u0020'}), xhtml, m, context, true);
+ // the entry carries the exception, not content: register it without a parse
+ try (TikaInputStream tis = TikaInputStream.getPlaceholder()) {
+ context.set(MetadataOnlyParse.class, MetadataOnlyParse.INSTANCE);
+ embeddedDocumentExtractor.parseEmbedded(tis, xhtml, m, context, true);
+ } finally {
+ context.set(MetadataOnlyParse.class, null);
+ }
}
return;
}
diff --git a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-pdf-module/src/main/java/org/apache/tika/parser/pdf/AbstractPDF2XHTML.java b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-pdf-module/src/main/java/org/apache/tika/parser/pdf/AbstractPDF2XHTML.java
index aa57face74..da4a51ed43 100644
--- a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-pdf-module/src/main/java/org/apache/tika/parser/pdf/AbstractPDF2XHTML.java
+++ b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-pdf-module/src/main/java/org/apache/tika/parser/pdf/AbstractPDF2XHTML.java
@@ -640,7 +640,7 @@ class AbstractPDF2XHTML extends PDFTextStripper {
new PageRangeRequest(getCurrentPageNo(), getCurrentPageNo());
if (thisRenderer instanceof PDDocumentRenderer) {
//do not do autocloseable. We need to leave the pdDocument open!
- TikaInputStream tis = TikaInputStream.get(new byte[0]);
+ TikaInputStream tis = TikaInputStream.getPlaceholder();
tis.setOpenContainer(pdDocument);
return thisRenderer.render(tis, pageMetadata, context, pageRangeRequest)
.getResults().get(0);
diff --git a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-pdf-module/src/main/java/org/apache/tika/parser/pdf/image/ImageGraphicsEngine.java b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-pdf-module/src/main/java/org/apache/tika/parser/pdf/image/ImageGraphicsEngine.java
index 1c04c23a44..e4387802f9 100644
--- a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-pdf-module/src/main/java/org/apache/tika/parser/pdf/image/ImageGraphicsEngine.java
+++ b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-pdf-module/src/main/java/org/apache/tika/parser/pdf/image/ImageGraphicsEngine.java
@@ -451,7 +451,7 @@ public class ImageGraphicsEngine extends PDFGraphicsStreamEngine {
metadata.set(TIFF.IMAGE_LENGTH, pdImage.getHeight());
//TODO: what else can we extract from the PDImage without rendering?
//Register the image's metadata entry without decoding it (marker skips the parse).
- try (TikaInputStream tis = TikaInputStream.get(new byte[0])) {
+ try (TikaInputStream tis = TikaInputStream.getPlaceholder()) {
parseContext.set(MetadataOnlyParse.class, MetadataOnlyParse.INSTANCE);
embeddedDocumentExtractor.parseEmbedded(tis,
new EmbeddedContentHandler(xhtml), metadata, parseContext, false);