Add C API for checking PDF files

Jay Berkenbilt
1 parent b72a38bf
Showing 10 changed files with 236 additions and 0 deletions
ChangeLog
include/qpdf/qpdf-c.h
libqpdf/qpdf-c.cc
qpdf/qpdf-ctest.c
qpdf/qpdf.testcov
qpdf/qtest/qpdf.test
qpdf/qtest/qpdf/c-check-clear-in.pdf
qpdf/qtest/qpdf/c-check-clear.out
qpdf/qtest/qpdf/c-check-warn-in.pdf
qpdf/qtest/qpdf/c-check-warn.out
 2018-02-17  Jay Berkenbilt  <ejb@ql.org>
  
+	* Add qpdf_check_pdf to the "C" API. This method just attempts to
+	read the entire file and produce no output, making possible to
+	assess whether the file has any errors that qpdf can detect.
+
 	* Major enhancements to handling of type errors within the qpdf
 	library. This fix is intended to eliminate those annoying cases
 	where qpdf would exit with a message like "operation for
@@ -180,6 +180,14 @@ extern &quot;C&quot; {
     QPDF_DLL
     void qpdf_set_suppress_warnings(qpdf_data qpdf, QPDF_BOOL value);
  
+    /* CHECK FUNCTIONS */
+
+    /* Attempt to read the entire PDF file to see if there are any
+     * errors qpdf can detect.
+     */
+    QPDF_DLL
+    QPDF_ERROR_CODE qpdf_check_pdf(qpdf_data qpdf);
+
     /* READ FUNCTIONS */
  
     /* READ PARAMETER FUNCTIONS -- must be called before qpdf_read */
@@ -4,6 +4,7 @@
 #include <qpdf/QPDFWriter.hh>
 #include <qpdf/QTC.hh>
 #include <qpdf/QPDFExc.hh>
+#include <qpdf/Pl_Discard.hh>
  
 #include <list>
 #include <string>
@@ -82,6 +83,15 @@ static void call_write(qpdf_data qpdf)
     qpdf->qpdf_writer->write();
 }
  
+static void call_check(qpdf_data qpdf)
+{
+    QPDFWriter w(*qpdf->qpdf);
+    Pl_Discard discard;
+    w.setOutputPipeline(&discard);
+    w.setDecodeLevel(qpdf_dl_all);
+    w.write();
+}
+
 static QPDF_ERROR_CODE trap_errors(qpdf_data qpdf, void (*fn)(qpdf_data))
 {
     QPDF_ERROR_CODE status = QPDF_SUCCESS;
@@ -236,6 +246,13 @@ char const* qpdf_get_error_message_detail(qpdf_data qpdf, qpdf_error e)
     return e->exc->getMessageDetail().c_str();
 }
  
+QPDF_ERROR_CODE qpdf_check_pdf(qpdf_data qpdf)
+{
+    QPDF_ERROR_CODE status = trap_errors(qpdf, &call_check);
+    QTC::TC("qpdf", "qpdf-c called qpdf_check_pdf");
+    return status;
+}
+
 void qpdf_set_suppress_warnings(qpdf_data qpdf, QPDF_BOOL value)
 {
     QTC::TC("qpdf", "qpdf-c called qpdf_set_suppress_warnings");
@@ -484,6 +484,18 @@ static void test22(char const* infile,
     report_errors();
 }
  
+static void test23(char const* infile,
+		   char const* password,
+		   char const* outfile,
+		   char const* outfile2)
+{
+    QPDF_ERROR_CODE status = 0;
+    qpdf_read(qpdf, infile, password);
+    status = qpdf_check_pdf(qpdf);
+    printf("status: %d\n", status);
+    report_errors();
+}
+
 int main(int argc, char* argv[])
 {
     char* p = 0;
@@ -546,6 +558,7 @@ int main(int argc, char* argv[])
 	  (n == 20) ? test20 :
 	  (n == 21) ? test21 :
 	  (n == 22) ? test22 :
+	  (n == 23) ? test23 :
 	  0);
  
     if (fn == 0)
@@ -333,3 +333,4 @@ QPDFObjectHandle dictionary ignoring removeKey 0
 QPDFObjectHandle dictionary ignoring removereplace 0
 QPDFObjectHandle numeric non-numeric 0
 QPDFObjectHandle erase array bounds 0
+qpdf-c called qpdf_check_pdf 0
@@ -2495,6 +2495,21 @@ foreach my $d (@enc_key)
  
 show_ntests();
 # ----------
+$td->notify("--- Check from C API ---");
+my @c_check_types = qw(warn clear);
+$n_tests += scalar(@c_check_types);
+
+foreach my $i (@c_check_types)
+{
+    $td->runtest("C check $i",
+                 {$td->COMMAND => "qpdf-ctest 23 c-check-$i-in.pdf '' -"},
+                 {$td->FILE => "c-check-$i.out",
+                      $td->EXIT_STATUS => 0},
+                 $td->NORMALIZE_NEWLINES);
+}
+
+show_ntests();
+# ----------
 $td->notify("--- Content Preservation Tests ---");
 # $n_tests incremented below
  
+%PDF-1.3
+1 0 obj
+<<
+  /Type /Catalog
+  /Pages 2 0 R
+>>
+endobj
+
+2 0 obj
+<<
+  /Type /Pages
+  /Kids [
+    3 0 R
+  ]
+  /Count 1
+>>
+endobj
+
+3 0 obj
+<<
+  /Type /Page
+  /Parent 2 0 R
+  /MediaBox [0 0 612 792]
+  /Contents 4 0 R
+  /Resources <<
+    /ProcSet 5 0 R
+    /Font <<
+      /F1 6 0 R
+    >>
+  >>
+>>
+endobj
+
+4 0 obj
+<<
+  /Length 44
+>>
+stream
+BT
+  /F1 24 Tf
+  72 720 Td
+  (Potato) Tj
+ET
+endstream
+endobj
+
+5 0 obj
+[
+  /PDF
+  /Text
+]
+endobj
+
+6 0 obj
+<<
+  /Type /Font
+  /Subtype /Type1
+  /Name /F1
+  /BaseFont /Helvetica
+  /Encoding /WinAnsiEncoding
+>>
+endobj
+
+xref
+0 7
+0000000000 65535 f 
+0000000009 00000 n 
+0000000063 00000 n 
+0000000135 00000 n 
+0000000307 00000 n 
+0000000403 00000 n 
+0000000438 00000 n 
+trailer <<
+  /Size 7
+  /Root 1 0 R
+>>
+startxref
+556
+%%EOF
+status: 0
+%PDF-1.3
+1 0 obj
+<<
+  /Type /Catalog
+  /Pages 2 0 R
+>>
+endobj
+
+2 0 obj
+<<
+  /Type /Pages
+  /Kids [
+    3 0 R
+  ]
+  /Count 1
+>>
+endobj
+
+3 0 obj
+<<
+  /Type /Page
+  /Parent 2 0 R
+  /MediaBox [0 0 612 792]
+  /Contents 4 0 R
+  /Resources <<
+    /ProcSet 5 0 R
+    /Font <<
+      /F1 6 0 R
+    >>
+  >>
+>>
+endobj
+
+4 0 obj
+<<
+  /Length 44
+>>
+stream
+BT
+  /F1 24 Tf
+  72 720 Td
+  (Potato) Tj
+ET
+endstream
+endobj
+
+5 0 obj
+[
+  /PDF
+  /Text
+]
+endobj
+
+6 0 obj
+<<
+  /Type /Font
+  /Subtype /Type1
+  /Name /F1
+  /BaseFont /Helvetica
+  /Encoding /WinAnsiEncoding
+>>
+endobj
+
+xref
+0 7
+0000000000 65535 f 
+0000000009 00000 n 
+0000000063 00000 n 
+0000000135 00000 n 
+0000000307 00000 n 
+0000000403 00000 n 
+0000000438 00000 n 
+trailer <<
+  /Size 7
+  /Root 1 0 R
+>>
+startxref
+1556
+%%EOF
+WARNING: c-check-warn-in.pdf: file is damaged
+WARNING: c-check-warn-in.pdf (offset 1556): xref not found
+WARNING: c-check-warn-in.pdf: Attempting to reconstruct cross-reference table
+status: 1
+warning: c-check-warn-in.pdf: file is damaged
+  code: 5
+  file: c-check-warn-in.pdf
+  pos : 0
+  text: file is damaged
+warning: c-check-warn-in.pdf (offset 1556): xref not found
+  code: 5
+  file: c-check-warn-in.pdf
+  pos : 1556
+  text: xref not found
+warning: c-check-warn-in.pdf: Attempting to reconstruct cross-reference table
+  code: 5
+  file: c-check-warn-in.pdf
+  pos : 0
+  text: Attempting to reconstruct cross-reference table