FIX: Document info date parsing

- fixes for date handling in the document info dictionary: - factored out getting the String literal from `addDateProperty` to `getDateLiteral` method, to simplify the methods; - no longer throwing a `PdfInvalidException` from `addDateProperty` so that bad dates don't halt docinfo processing, i.e. `ModDate` is still validated if `CreationDate` is invalid; - fixes to `jhove-bbt/scripts/create-1.33-target.sh' to adjust output affected by the above; and - bumped PDF module -> 1.12.7 and date.
openpreserve · Jan 7, 2025 · a0d2e0d · a0d2e0d
1 parent 8ee2c10
commit a0d2e0d
Show file tree

Hide file tree

Showing 5 changed files with 47 additions and 20 deletions.
diff --git a/jhove-bbt/scripts/create-1.33-target.sh b/jhove-bbt/scripts/create-1.33-target.sh
@@ -56,3 +56,15 @@ echo "TEST BASELINE: Creating baseline"
 echo " - copying ${baselineRoot} baseline to ${targetRoot}"
 cp -R "${baselineRoot}" "${targetRoot}"
 
+
+# Update release details for PDF module
+find "${targetRoot}" -type f -name "*.pdf.jhove.xml" -exec sed -i 's/<reportingModule release="1.12.7" date="2024-08-22">PDF-hul<\/reportingModule>/<reportingModule release="1.12.8" date="2025-01-24">PDF-hul<\/reportingModule>/' {} \;
+find "${targetRoot}" -type f -name "audit.jhove.xml" -exec sed -i 's/<module release="1.12.7">PDF-hul<\/module>/<module release="1.12.8">PDF-hul<\/module>/' {} \;
+find "${targetRoot}" -type f -name "audit-PDF-hul.jhove.xml" -exec sed -i 's/<release>1.12.7<\/release>/<release>1.12.8<\/release>/' {} \;
+find "${targetRoot}" -type f -name "audit-PDF-hul.jhove.xml" -exec sed -i 's/2024-08-22/2025-01-24/' {} \;
+
+# Fix the results affected by the improvements to date handling in the PDF module
+sed -i 's/<message offset/<message subMessage="For date property CreationDate" offset/' "${targetRoot}/examples/modules/PDF-hul/AA_Banner.pdf.jhove.xml"
+if [[ -f "${candidateRoot}/errors/modules/PDF-hul/pdf-hul-9-govdocs-065694.pdf.jhove.xml" ]]; then
+	cp "${candidateRoot}/errors/modules/PDF-hul/pdf-hul-9-govdocs-065694.pdf.jhove.xml" "${targetRoot}/errors/modules/PDF-hul/pdf-hul-9-govdocs-065694.pdf.jhove.xml"
+fi
diff --git a/jhove-installer/pom.xml b/jhove-installer/pom.xml
@@ -25,7 +25,7 @@
     <html.hul.version>1.4.4</html.hul.version>
     <jpeg2000.hul.version>1.4.4</jpeg2000.hul.version>
     <jpeg.hul.version>1.5.4</jpeg.hul.version>
-    <pdf.hul.version>1.12.7</pdf.hul.version>
+    <pdf.hul.version>1.12.8</pdf.hul.version>
     <tiff.hul.version>1.9.5</tiff.hul.version>
     <utf8.hul.version>1.7.4</utf8.hul.version>
     <wave.hul.version>1.8.3</wave.hul.version>

diff --git a/jhove-modules/pdf-hul/pom.xml b/jhove-modules/pdf-hul/pom.xml
@@ -6,7 +6,7 @@
     <version>1.33.0-SNAPSHOT</version>
   </parent>
   <artifactId>pdf-hul</artifactId>
-  <version>1.12.7</version>
+  <version>1.12.8</version>
   <name>JHOVE PDF Module HUL</name>
   <description>PDF module developed by Harvard University Library</description>
 </project>
diff --git a/jhove-modules/pdf-hul/src/main/java/edu/harvard/hul/ois/jhove/module/PdfModule.java b/jhove-modules/pdf-hul/src/main/java/edu/harvard/hul/ois/jhove/module/PdfModule.java
@@ -380,8 +380,8 @@ public class PdfModule extends ModuleBase {
      ******************************************************************/
 
     private static final String NAME = "PDF-hul";
-    private static final String RELEASE = "1.12.7";
-    private static final int[] DATE = { 2024, 8, 22 };
+    private static final String RELEASE = "1.12.8";
+    private static final int[] DATE = { 2025, 1, 24 };
     private static final String[] FORMAT = { "PDF",
             "Portable Document Format" };
     private static final String COVERAGE = "PDF 1.0-1.6; "
@@ -2031,9 +2031,9 @@ protected boolean readDocInfoDict(RepInfo info) {
             // CreationDate requires string-to-date conversion
             // ModDate does too
             addDateProperty(_docInfoDict, _docInfoList, DICT_KEY_CREATION_DATE,
-                    PROP_NAME_CREATION_DATE);
+                    PROP_NAME_CREATION_DATE, info);
             addDateProperty(_docInfoDict, _docInfoList, DICT_KEY_MODIFIED_DATE,
-                    PROP_NAME_MODIFIED_DATE);
+                    PROP_NAME_MODIFIED_DATE, info);
             addStringProperty(_docInfoDict, _docInfoList, DICT_KEY_TRAPPED,
                     PROP_NAME_TRAPPED);
         } catch (PdfException e) {
@@ -4333,27 +4333,42 @@ protected void addStringProperty(PdfDictionary dict,
      * with a string value, to a specified List.
      */
     protected void addDateProperty(PdfDictionary dict, List<Property> propList,
-            String key, String propName) throws PdfInvalidException {
+                 String key, String propName, final RepInfo info) {
         if (_encrypted) {
             String propText = ENCRYPTED;
             propList.add(new Property(propName, PropertyType.STRING, propText));
-        } else {
-            PdfObject propObject = dict.get(key);
-            if (propObject instanceof PdfSimpleObject) {
-                Token tok = ((PdfSimpleObject) propObject).getToken();
-                if (tok instanceof Literal) {
-                    Literal lit = (Literal) tok;
-                    if (!lit.getValue().isEmpty()) {
-                        Date propDate = lit.parseDate();
-                        if (propDate != null) {
-                            propList.add(new Property(propName, PropertyType.DATE, propDate));
-                        }
-                    }
+            return;
+        }
+        PdfObject propObject = dict.get(key);
+        if (propObject == null) {
+            return;
+        }
+        Literal lit = getDateLiteral(propObject);
+        if (lit != null && !lit.getValue().isEmpty()) {
+            try {
+                Date propDate = lit.parseDate();
+                if (propDate != null) {
+                    propList.add(new Property(propName, PropertyType.DATE, propDate));
+                    return;
                 }
+            } catch (PdfInvalidException e) {
+                info.setValid(false);
+                info.setMessage(new ErrorMessage(JhoveMessages.getMessageInstance(
+                    MessageConstants.PDF_HUL_133.getId(), MessageConstants.PDF_HUL_133.getMessage(), "For date property " + propName), _parser.getOffset()));
             }
         }
     }
 
+    private Literal getDateLiteral(final PdfObject obj) {
+        if (obj instanceof PdfSimpleObject) {
+            Token tok = ((PdfSimpleObject) obj).getToken();
+            if (tok instanceof Literal) {
+                return (Literal) tok;
+            }
+        }
+        return null;
+    }
+
     /*
      * General function for adding a property with a 32-bit
      * value, with an array of Strings to interpret

diff --git a/jhove-modules/utf8-hul/pom.xml b/jhove-modules/utf8-hul/pom.xml
@@ -19,7 +19,7 @@
       <dependency>
         <groupId>org.openpreservation.jhove.modules</groupId>
         <artifactId>pdf-hul</artifactId>
-        <version>1.12.7</version>
+        <version>1.12.8</version>
         <scope>test</scope>
       </dependency>
   </dependencies>