Add option to save pass 1 of lineariziation

This is useful only for debugging the linearization code.

ChangeLog

@@ -1,3 +1,13 @@
+2018-02-04  Jay Berkenbilt  <ejb@ql.org>
+
+	* Add QPDFWriter::setLinearizationPass1Filename method and
+	--linearize-pass1 command line option to allow specification of a
+	file into which QPDFWriter will write its intermediate
+	linearization pass 1 file. This is useful only for debugging qpdf.
+	qpdf creates linearized files by computing the output in two
+	passes. Ordinarily the first pass is discarded and not written
+	anywhere. This option allows it to be inspected.
+
 2018-02-04  Jay Berkenbilt  <ejb@ql.org>
 
 	* 7.1.1: release

TODO

@@ -177,44 +177,6 @@ I find it useful to make reference to them in this list
  * See if we can avoid preserving unreferenced objects in object
    streams even when preserving the object streams.
 
- * For debugging linearization bugs, consider adding an option to save
-   pass 1 of linearization.  This code is sufficient.  Change the
-   interface to allow specification of a pass1 file, which would
-   change the behavior as in this patch.
-
-------------------------------
-Index: QPDFWriter.cc
-===================================================================
---- QPDFWriter.cc	(revision 932)
-+++ QPDFWriter.cc	(working copy)
-@@ -1965,11 +1965,15 @@
- 
-     // Write file in two passes.  Part numbers refer to PDF spec 1.4.
- 
-+    FILE* XXX = 0;
-     for (int pass = 1; pass <= 2; ++pass)
-     {
- 	if (pass == 1)
- 	{
--	    pushDiscardFilter();
-+//	    pushDiscardFilter();
-+	    XXX = QUtil::safe_fopen("/tmp/pass1.pdf", "w");
-+	    pushPipeline(new Pl_StdioFile("pass1", XXX));
-+	    activatePipelineStack();
- 	}
- 
- 	// Part 1: header
-@@ -2204,6 +2208,8 @@
- 
- 	    // Restore hint offset
- 	    this->xref[hint_id] = QPDFXRefEntry(1, hint_offset, 0);
-+	    fclose(XXX);
-+	    XXX = 0;
- 	}
-     }
- }
-------------------------------
-
  * Provide APIs for embedded files.  See *attachments*.pdf in test
    suite.  The private method findAttachmentStreams finds at least
    cases for modern versions of Adobe Reader (>= 1.7, maybe earlier).

include/qpdf/QPDFWriter.hh

@@ -367,6 +367,15 @@ class QPDFWriter
     QPDF_DLL
     void setLinearization(bool);
 
+    // For debugging QPDF: provide the name of a file to write pass1
+    // of linearization to. The only reason to use this is to debug
+    // QPDF. To linearize, QPDF writes out the file in two passes.
+    // Usually the first pass is discarded, but lots of computations
+    // are made in pass 1. If a linearized file comes out wrong, it
+    // can be helpful to look at the first pass.
+    QPDF_DLL
+    void setLinearizationPass1Filename(std::string const&);
+
     // Create PCLm output. This is only useful for clients that know
     // how to create PCLm files. If a file is structured exactly as
     // PCLm requires, this call will tell QPDFWriter to write the PCLm
@@ -571,6 +580,7 @@ class QPDFWriter
         std::string deterministic_id_data;
 
         // For linearization only
+        std::string lin_pass1_filename;
         std::map<int, int> obj_renumber_no_gen;
         std::map<int, int> object_to_object_stream_no_gen;
     };

libqpdf/QPDFWriter.cc

@@ -356,6 +356,12 @@ QPDFWriter::setLinearization(bool val)
     }
 }
 
+void
+QPDFWriter::setLinearizationPass1Filename(std::string const& filename)
+{
+    this->m->lin_pass1_filename = filename;
+}
+
 void
 QPDFWriter::setPCLm(bool val)
 {
@@ -2957,11 +2963,24 @@ QPDFWriter::writeLinearized()
 
     // Write file in two passes.  Part numbers refer to PDF spec 1.4.
 
+    FILE* lin_pass1_file = 0;
     for (int pass = 1; pass <= 2; ++pass)
     {
 	if (pass == 1)
+	{
+            if (! this->m->lin_pass1_filename.empty())
+            {
+                lin_pass1_file =
+                    QUtil::safe_fopen(
+                        this->m->lin_pass1_filename.c_str(), "wb");
+                pushPipeline(
+                    new Pl_StdioFile("linearization pass1", lin_pass1_file));
+                activatePipelineStack();
+            }
+            else
             {
                 pushDiscardFilter();
+            }
             if (this->m->deterministic_id)
             {
                 pushMD5Pipeline();
@@ -3201,6 +3220,20 @@ QPDFWriter::writeLinearized()
 
 	    // Restore hint offset
 	    this->m->xref[hint_id] = QPDFXRefEntry(1, hint_offset, 0);
+            if (lin_pass1_file)
+            {
+                // Write some debugging information
+                fprintf(lin_pass1_file, "%% hint_offset=%s\n",
+                        QUtil::int_to_string(hint_offset).c_str());
+                fprintf(lin_pass1_file, "%% hint_length=%s\n",
+                        QUtil::int_to_string(hint_length).c_str());
+                fprintf(lin_pass1_file, "%% second_xref_offset=%s\n",
+                        QUtil::int_to_string(second_xref_offset).c_str());
+                fprintf(lin_pass1_file, "%% second_xref_end=%s\n",
+                        QUtil::int_to_string(second_xref_end).c_str());
+                fclose(lin_pass1_file);
+                lin_pass1_file = 0;
+            }
 	}
     }
 }

qpdf/qpdf.cc

@@ -153,6 +153,7 @@ struct Options
     bool qdf_mode;
     bool preserve_unreferenced_objects;
     bool newline_before_endstream;
+    std::string linearize_pass1;
     std::string min_version;
     std::string force_version;
     bool show_npages;
@@ -391,6 +392,8 @@ familiar with the PDF file format or who are PDF developers.\n\
 --preserve-unreferenced   preserve unreferenced objects\n\
 --newline-before-endstream  always put a newline before endstream\n\
 --qdf                     turns on \"QDF mode\" (below)\n\
+--linearize-pass1=file    write intermediate pass of linearized file\n\
+                          for debugging\n\
 --min-version=version     sets the minimum PDF version of the output file\n\
 --force-version=version   forces this to be the PDF version of the output file\n\
 \n\
@@ -1531,6 +1534,15 @@ static void parse_options(int argc, char* argv[], Options& o)
             {
                 o.newline_before_endstream = true;
             }
+            else if (strcmp(arg, "linearize-pass1") == 0)
+            {
+                if (parameter == 0)
+                {
+                    usage("--linearize-pass1 be given as"
+                          "--linearize-pass1=filename");
+                }
+                o.linearize_pass1 = parameter;
+            }
             else if (strcmp(arg, "min-version") == 0)
             {
                 if (parameter == 0)
@@ -2214,6 +2226,10 @@ static void set_writer_options(QPDF& pdf, Options& o, QPDFWriter& w)
     {
         w.setLinearization(true);
     }
+    if (! o.linearize_pass1.empty())
+    {
+        w.setLinearizationPass1Filename(o.linearize_pass1);
+    }
     if (o.object_stream_set)
     {
         w.setObjectStreamMode(o.object_stream_mode);

qpdf/qtest/qpdf.test

@@ -240,7 +240,7 @@ foreach my $d (@bug_tests)
 show_ntests();
 # ----------
 $td->notify("--- Miscellaneous Tests ---");
-$n_tests += 93;
+$n_tests += 96;
 
 $td->runtest("qpdf version",
 	     {$td->COMMAND => "qpdf --version"},
@@ -252,6 +252,17 @@ $td->runtest("C API: qpdf version",
 	      $td->EXIT_STATUS => 0},
 	     $td->NORMALIZE_NEWLINES);
 
+$td->runtest("linearize pass 1 file",
+             {$td->COMMAND => "qpdf --linearize --static-id" .
+                  " --linearize-pass1=b.pdf minimal.pdf a.pdf"},
+             {$td->STRING => "", $td->EXIT_STATUS => 0});
+$td->runtest("check output",
+	     {$td->FILE => "a.pdf"},
+	     {$td->FILE => "minimal-linearized.pdf"});
+$td->runtest("check pass1 file",
+	     {$td->FILE => "b.pdf"},
+	     {$td->FILE => "minimal-linearize-pass1.pdf"});
+
 foreach (my $i = 1; $i <= 3; ++$i)
 {
     $td->runtest("misc tests",