Fix repaired xref stream offsets and support indirect Contents arrays for pages.

CHANGES.md

@@ -9,6 +9,8 @@ v1.6.2 - YYYY-MM-DD
 - Added missing range checks to `pdfioArrayCopy` and `pdfioDictCopy`.
 - Refactored PDF encryption code to fix unlocking with certain files.
 - Improved xref table loop detection (Issue #148)
+- Fixed xref reconstruction for objects lacking a `Type` value.
+- Fixed `pdfioPageOpenStream` for indirect `Contents` arrays.
 - Fixed an error propagation bug when reading too-long values (Issue #146)
 - Fixed a Clang warning.
 

pdfio-file.c

@@ -2621,19 +2621,19 @@ repair_xref(
 
 	    _pdfioTokenFlush(&tb);
 
-            if (type && !strcmp(line, "stream"))
+            if (!strcmp(line, "stream"))
             {
               // Possible object or XRef stream...
 	      obj->stream_offset = _pdfioFileTell(pdf);
 
-	      if (!strcmp(type, "ObjStm") && num_sobjs < (sizeof(sobjs) / sizeof(sobjs[0])))
+	      if (type && !strcmp(type, "ObjStm") && num_sobjs < (sizeof(sobjs) / sizeof(sobjs[0])))
 	      {
 	        PDFIO_DEBUG("repair_xref: Object stream...\n");
 		sobjs[num_sobjs] = obj;
 		num_sobjs ++;
 	      }
 
-	      if (!strcmp(type, "XRef") && !pdf->trailer_dict)
+	      if (type && !strcmp(type, "XRef") && !pdf->trailer_dict)
 	      {
 		// Save the trailer dictionary...
 		pdfio_obj_t *encrypt_obj;

pdfio-object.c

@@ -547,6 +547,8 @@ pdfioObjOpenStream(pdfio_obj_t *obj,	// I - Object
   pdfio_stream_t	*st;		// Stream
 
 
+  PDFIO_DEBUG("pdfioObjOpenStream(obj=%p(%lu), decode=%s)\n", (void *)obj, obj ? (unsigned long)obj->number : 0, decode ? "true" : "false");
+
   // Range check input...
   if (!obj)
     return (NULL);
@@ -566,7 +568,10 @@ pdfioObjOpenStream(pdfio_obj_t *obj,	// I - Object
 
   // No stream if there is no dict or offset to a stream...
   if (obj->value.type != PDFIO_VALTYPE_DICT || !obj->stream_offset)
+  {
+    PDFIO_DEBUG("pdfioObjOpenStream: value.type=%d, stream_offset=%ld\n", obj->value.type, (long)obj->stream_offset);
     return (NULL);
+  }
 
   // Open the stream...
   if ((st = _pdfioStreamOpen(obj, decode)) != NULL)

pdfio-page.c

@@ -105,6 +105,10 @@ pdfioPageOpenStream(
 static _pdfio_value_t *			// O - Value or NULL on error
 get_contents(pdfio_obj_t *page)		// I - Page object
 {
+  _pdfio_value_t *contents;		// Contents value
+  pdfio_obj_t	*obj;			// Contents object
+
+
   // Range check input...
   if (!page)
     return (NULL);
@@ -119,5 +123,24 @@ get_contents(pdfio_obj_t *page)		// I - Page object
   if (page->value.type != PDFIO_VALTYPE_DICT)
     return (NULL);
 
-  return (_pdfioDictGetValue(page->value.value.dict, "Contents"));
+  contents = _pdfioDictGetValue(page->value.value.dict, "Contents");
+
+  if (contents->type == PDFIO_VALTYPE_INDIRECT)
+  {
+    // See if the indirect object is a stream or an array of indirect object
+    // references...
+    if ((obj = pdfioFileFindObj(page->pdf, contents->value.indirect.number)) != NULL)
+    {
+      if (obj->value.type == PDFIO_VALTYPE_NONE)
+      {
+	if (!_pdfioObjLoad(obj))
+	  return (NULL);
+      }
+
+      if (obj->value.type == PDFIO_VALTYPE_ARRAY)
+        contents = &(obj->value);
+    }
+  }
+
+  return (contents);
 }

testpdfio.c

@@ -552,10 +552,13 @@ do_test_file(const char *filename,	// I - PDF filename
 
 	      if (!pdfioDictGetRect(dict, "MediaBox", &media_box))
 	      {
-		if ((obj = pdfioDictGetObj(dict, "Parent")) != NULL)
+	        pdfio_obj_t *parent;	// Parent object
+
+		while ((parent = pdfioDictGetObj(dict, "Parent")) != NULL)
 		{
-		  dict = pdfioObjGetDict(obj);
+		  dict = pdfioObjGetDict(parent);
-		  pdfioDictGetRect(dict, "MediaBox", &media_box);
+		  if (pdfioDictGetRect(dict, "MediaBox", &media_box))
+		    break;
 		}
 	      }