deal with stream-specific crypt filters

git-svn-id: svn+q:///qpdf/trunk@827 71b93d88-0707-0410-a8cf-f5a4172ac649

deal with stream-specific crypt filters
git-svn-id: svn+q:///qpdf/trunk@827 71b93d88-0707-0410-a8cf-f5a4172ac649
Jay Berkenbilt
1 parent 70ae58c0
Showing 9 changed files with 483 additions and 24 deletions
TODO
include/qpdf/QPDFObjectHandle.hh
libqpdf/QPDFObjectHandle.cc
libqpdf/QPDF_Stream.cc
libqpdf/QPDF_encryption.cc
qpdf/qpdf.testcov
qpdf/qtest/qpdf.test
qpdf/qtest/qpdf/decrypted-crypt-filter.pdf
qpdf/qtest/qpdf/metadata-crypt-filter.pdf
 2.1
 ===
- * Really need to handle /Crypt filter for Metadata.  Search for crypt
-   below.
-
  * Update documentation to reflect new command line flags and any
    other relevant changes.  Should read through ChangeLog and the
    manual before releasing 2.1.
@@ -83,10 +80,13 @@ General
    filters.  There is an example in the spec of using a crypt filter
    on a metadata stream.
-   When we write encrypted files, we must remember to omit any
-   encryption filter settings from original streams.
-
-   We need a way to test this.
+   For now, we notice /Crypt filters and decode parameters consistent
+   with the example in the PDF specification, and the right thing
+   happens for metadata filters that happen to be uncompressed or
+   otherwise compressed in a way we can filter.  This should handle
+   all normal cases, but it's more or less just a guess since I don't
+   have any test files that actually use stream-specific crypt filters
+   in them.
  * The second xref stream for linearized files has to be padded only
    because we need file_size as computed in pass 1 to be accurate.  If
@@ -77,14 +77,14 @@ class DLL_EXPORT QPDFObjectHandle
     bool isNumber();
     double getNumericValue();
-    // Methods for name objects
+    // Methods for name objects; see also name and array objects
     std::string getName();
     // Methods for string objects
     std::string getStringValue();
     std::string getUTF8Value();
-    // Methods for array objects
+    // Methods for array objects; see also name and array objects
     int getArrayNItems();
     QPDFObjectHandle getArrayItem(int n);
@@ -93,6 +93,9 @@ class DLL_EXPORT QPDFObjectHandle
     QPDFObjectHandle getKey(std::string const&);
     std::set<std::string> getKeys();
+    // Methods for name and array objects
+    bool isOrHasName(std::string const&);
+
     // Mutator methods.  Use with caution.
     // Recursively copy this object, making it direct.  Throws an
@@ -258,6 +258,29 @@ QPDFObjectHandle::getKeys()
     return dynamic_cast<QPDF_Dictionary*>(obj.getPointer())->getKeys();
 }
+// Array and Name accessors
+bool
+QPDFObjectHandle::isOrHasName(std::string const& value)
+{
+    if (isName() && (getName() == value))
+    {
+	return true;
+    }
+    else if (isArray())
+    {
+	int n = getArrayNItems();
+	for (int i = 0; i < n; ++i)
+	{
+	    QPDFObjectHandle item = getArrayItem(0);
+	    if (item.isName() && (item.getName() == value))
+	    {
+		return true;
+	    }
+	}
+    }
+    return false;
+}
+
 // Dictionary mutators
 void
@@ -136,6 +136,13 @@ QPDF_Stream::filterable(std::vector&lt;std::string&gt;&amp; filters,
 		    filterable = false;
 		}
 	    }
+	    else if (((key == "/Type") || (key == "/Name")) &&
+		     decode_obj.getKey("/Type").isName() &&
+		     (decode_obj.getKey("/Type").getName() ==
+		      "/CryptFilterDecodeParms"))
+	    {
+		// we handle this in decryptStream
+	    }
 	    else
 	    {
 		filterable = false;
@@ -212,7 +219,8 @@ QPDF_Stream::filterable(std::vector&lt;std::string&gt;&amp; filters,
 	 iter != filters.end(); ++iter)
     {
 	std::string const& filter = *iter;
-	if (! ((filter == "/FlateDecode") ||
+	if (! ((filter == "/Crypt") ||
+	       (filter == "/FlateDecode") ||
 	       (filter == "/LZWDecode") ||
 	       (filter == "/ASCII85Decode") ||
 	       (filter == "/ASCIIHexDecode")))
@@ -266,7 +274,11 @@ QPDF_Stream::pipeStreamData(Pipeline* pipeline, bool filter,
 	     iter != filters.rend(); ++iter)
 	{
 	    std::string const& filter = *iter;
-	    if (filter == "/FlateDecode")
+	    if (filter == "/Crypt")
+	    {
+		// Ignore -- handled by pipeStreamData
+	    }
+	    else if (filter == "/FlateDecode")
 	    {
 		if (predictor == 12)
 		{
@@ -600,18 +600,19 @@ QPDF::decryptStream(Pipeline*&amp; pipeline, int objid, int generation,
 	encryption_method_e method = e_unknown;
 	std::string method_source = "/StmF from /Encrypt dictionary";
-	// NOTE: the section in the PDF specification on crypt filters
-	// seems to suggest that there might be a /Crypt key in
-	// /DecodeParms whose value is a crypt filter (.e.g., << /Name
-	// /StdCF >>), but implementation notes suggest this can only
-	// happen for metadata streams, and emperical observation
-	// suggests that they are otherwise ignored.  Not having been
-	// able to find a sample file that uses crypt filters in any
-	// way other than /StrF and /StmF, I'm not really sure what to
-	// do about this.  If we were to override the encryption on a
-	// per-stream basis using crypt filters, set method_source to
-	// something useful in the error message for unknown
-	// encryption methods (search for method_source).
+	if (stream_dict.getKey("/Filter").isOrHasName("/Crypt") &&
+	    stream_dict.getKey("/DecodeParms").isDictionary())
+	{
+	    QPDFObjectHandle decode_parms = stream_dict.getKey("/DecodeParms");
+	    if (decode_parms.getKey("/Type").isName() &&
+		(decode_parms.getKey("/Type").getName() ==
+		 "/CryptFilterDecodeParms"))
+	    {
+		QTC::TC("qpdf", "QPDF_encryption stream crypt filter");
+		method = interpretCF(decode_parms.getKey("/Name"));
+		method_source = "stream's Crypt decode parameters";
+	    }
+	}
 	if (method == e_unknown)
 	{
@@ -170,3 +170,4 @@ QPDFWriter forcing object stream disable 0
 QPDFWriter forced version disabled encryption 0
 qpdf-c called qpdf_set_r4_encryption_parameters 0
 qpdf-c called qpdf_set_static_aes_IV 0
+QPDF_encryption stream crypt filter 0
@@ -1079,7 +1079,7 @@ $td-&gt;runtest(&quot;make sure there is no xref stream&quot;,
 	     $td->NORMALIZE_NEWLINES);
 # Look at some actual V4 files
-$n_tests += 8;
+$n_tests += 10;
 foreach my $d (['--force-V4', 'V4'],
 	       ['--cleartext-metadata', 'V4-clearmeta'],
 	       ['--use-aes=y', 'V4-aes'],
@@ -1094,6 +1094,14 @@ foreach my $d ([&#39;--force-V4&#39;, &#39;V4&#39;],
 		 {$td->FILE => "a.pdf"},
 		 {$td->FILE => "$out.pdf"});
 }
+# Crypt Filter
+$td->runtest("decrypt with crypt filter",
+	     {$td->COMMAND => "qpdf --decrypt --static-id" .
+		  " metadata-crypt-filter.pdf a.pdf"},
+	     {$td->STRING => "", $td->EXIT_STATUS => 0});
+$td->runtest("check output",
+	     {$td->FILE => 'a.pdf'},
+	     {$td->FILE => 'decrypted-crypt-filter.pdf'});
 show_ntests();
 # ----------