[tika] 01/02: TIKA-3477 -- don't close embedded word doc because that in turn closes the root. Unrelated issue, avoid NPE if ooxml part doesn't exist.

[ta...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

tallison pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tika.git

commit 070222a6a968aca6aff6fdf438379dfe4723e7e0
Author: tallison <ta...@apache.org>
AuthorDate: Wed Jul 14 10:19:25 2021 -0400

    TIKA-3477 -- don't close embedded word doc because that in turn closes the root.  Unrelated issue, avoid NPE if ooxml part doesn't exist.
---
 .../apache/tika/parser/microsoft/WordExtractor.java   | 19 ++++++++++---------
 .../microsoft/ooxml/AbstractOOXMLExtractor.java       |  3 +++
 2 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/WordExtractor.java b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/WordExtractor.java
index f8ff5e1..7816073 100644
--- a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/WordExtractor.java
+++ b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/WordExtractor.java
@@ -85,11 +85,8 @@ public class WordExtractor extends AbstractPOIFSExtractor {
 
     private final Deque<FormattingUtils.Tag> formattingState = new ArrayDeque<>();
 
-    private final Metadata metadata;
-
     public WordExtractor(ParseContext context, Metadata metadata) {
-        super(context);
-        this.metadata = metadata;
+        super(context, metadata);
     }
 
     private static int countParagraphs(Range... ranges) {
@@ -218,14 +215,16 @@ public class WordExtractor extends AbstractPOIFSExtractor {
             p = pictures.nextUnclaimed();
         }
 
-        // Handle any embeded office documents
+        // Handle any embedded office documents
         try {
+
             DirectoryEntry op = (DirectoryEntry) root.getEntry("ObjectPool");
             for (Entry entry : op) {
                 if (entry.getName().startsWith("_") && entry instanceof DirectoryEntry) {
                     handleEmbeddedOfficeDoc((DirectoryEntry) entry, xhtml);
                 }
             }
+
         } catch (FileNotFoundException e) {
             //swallow
         }
@@ -237,7 +236,7 @@ public class WordExtractor extends AbstractPOIFSExtractor {
             return;
         }
         for (SavedByEntry sbe : savedByTable.getEntries()) {
-            metadata.add(TikaCoreProperties.ORIGINAL_RESOURCE_NAME, sbe.getSaveLocation());
+            parentMetadata.add(TikaCoreProperties.ORIGINAL_RESOURCE_NAME, sbe.getSaveLocation());
         }
     }
 
@@ -596,9 +595,11 @@ public class WordExtractor extends AbstractPOIFSExtractor {
     protected void parseWord6(DirectoryNode root, XHTMLContentHandler xhtml)
             throws IOException, SAXException {
         Word6Extractor extractor;
-        try (HWPFOldDocument doc = new HWPFOldDocument(root)) {
-            extractor = new Word6Extractor(doc);
-        }
+        //DO NOT put this in a try/autoclose.  This will close the root
+        //which we don't want to do because there may be other
+        //documents in the root.
+        HWPFOldDocument doc = new HWPFOldDocument(root);
+        extractor = new Word6Extractor(doc);
 
         for (String p : extractor.getParagraphText()) {
             xhtml.element("p", p);
diff --git a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/ooxml/AbstractOOXMLExtractor.java b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/ooxml/AbstractOOXMLExtractor.java
index cd6c6e5..f23ae12 100644
--- a/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/ooxml/AbstractOOXMLExtractor.java
+++ b/tika-parsers/tika-parsers-standard/tika-parsers-standard-modules/tika-parser-microsoft-module/src/main/java/org/apache/tika/parser/microsoft/ooxml/AbstractOOXMLExtractor.java
@@ -167,6 +167,9 @@ public abstract class AbstractOOXMLExtractor implements OOXMLExtractor {
             for (PackageRelationship rel : opcPackage
                     .getRelationshipsByType(PackageRelationshipTypes.THUMBNAIL)) {
                 PackagePart tPart = opcPackage.getPart(rel);
+                if (tPart == null) {
+                    continue;
+                }
                 InputStream tStream = tPart.getInputStream();
                 Metadata thumbnailMetadata = new Metadata();
                 String thumbName = tPart.getPartName().getName();