jpeg: add encodingProcess and numColorComponents SOF members (backport #2874)

.github/workflows/on_PR_linux_special_builds.yml

@@ -23,6 +23,7 @@ jobs:
 
       - name: install dependencies
         run: |
+          sudo apt-get install -y libxml2-dev libxslt-dev python3-dev
           python3 -m pip install conan==1.59.0 gcovr ninja
 
       - name: Conan common config

include/exiv2/jpgimage.hpp

@@ -86,6 +86,9 @@ class EXIV2API JpegBase : public Image {
   virtual int writeHeader(BasicIo& oIo) const = 0;
   //@}
 
+  int num_color_components_{-1};      //!< image number of color components
+  std::string sof_encoding_process_;  //!< image encoding process
+
  private:
   //! @name Manipulators
   //@{
@@ -153,6 +156,18 @@ class EXIV2API JpegImage : public JpegBase {
   //! @name Accessors
   //@{
   [[nodiscard]] std::string mimeType() const override;
+  /*!
+    @brief Get the number of color components of the JPEG Image
+  */
+  [[nodiscard]] int numColorComponents() const {
+    return num_color_components_;
+  }
+  /*!
+    @brief Get the encoding process of the JPEG Image derived from the Start of Frame (SOF) markers
+  */
+  [[nodiscard]] std::string encodingProcess() const {
+    return sof_encoding_process_;
+  }
   //@}
 
  protected:

samples/CMakeLists.txt

@@ -1,5 +1,6 @@
 # CMakeLists.txt for exiv2 library sample applications
 
+<<<<<<< HEAD
 set( SAMPLES
      addmoddel.cpp
      convert-test.cpp
@@ -29,6 +30,38 @@ set( SAMPLES
      xmpprint.cpp
      xmpsample.cpp
      xmpdump.cpp
+=======
+set(SAMPLES
+    addmoddel.cpp
+    convert-test.cpp
+    easyaccess-test.cpp
+    exifcomment.cpp
+    exifdata-test.cpp
+    exifdata.cpp
+    exifprint.cpp
+    exifvalue.cpp
+    ini-test.cpp
+    iotest.cpp
+    iptceasy.cpp
+    iptcprint.cpp
+    iptctest.cpp
+    jpegparsetest.cpp
+    key-test.cpp
+    largeiptc-test.cpp
+    mmap-test.cpp
+    mrwthumb.cpp
+    prevtest.cpp
+    stringto-test.cpp
+    taglist.cpp
+    tiff-test.cpp
+    write-test.cpp
+    write2-test.cpp
+    xmpparse.cpp
+    xmpparser-test.cpp
+    xmpprint.cpp
+    xmpsample.cpp
+    xmpdump.cpp
+>>>>>>> 6b5cb9841 (Add test to validate jpeg parse)
 )
 
 ##

samples/jpegparsetest.cpp

@@ -0,0 +1,28 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+// Sample program to parse jpeg files and obtain its specific metadata
+
+#include <exiv2/exiv2.hpp>
+#include <exiv2/image_types.hpp>
+#include <iostream>
+
+int main(int argc, char* const argv[]) {
+  try {
+    if (argc != 2) {
+      std::cout << "Usage: " << argv[0] << " file\n";
+      return EXIT_FAILURE;
+    }
+
+    auto image = Exiv2::ImageFactory::open(argv[1]);
+    if (image->imageType() == Exiv2::ImageType::jpeg) {
+      auto jpegImage = dynamic_cast<Exiv2::JpegImage*>(image.get());
+      jpegImage->readMetadata();
+      std::cout << "Number of color components: " << jpegImage->numColorComponents() << "\n";
+      std::cout << "Encoding process: " << jpegImage->encodingProcess() << "\n";
+    }
+
+    return EXIT_SUCCESS;
+  } catch (Exiv2::Error& e) {
+    std::cout << "Caught Exiv2 exception '" << e << "'\n";
+    return EXIT_FAILURE;
+  }
+}

src/jpgimage.cpp

@@ -7,10 +7,12 @@
 #include "error.hpp"
 #include "futils.hpp"
 #include "helper_functions.hpp"
+#include "i18n.h"  // NLS support.
 #include "image_int.hpp"
 #include "jpgimage.hpp"
 #include "photoshop.hpp"
 #include "safe_op.hpp"
+#include "tags_int.hpp"
 #include "utils.hpp"
 
 #ifdef _WIN32
@@ -49,14 +51,40 @@ constexpr byte rst1_ = 0xd0;  //!< JPEG Restart 0 Marker (from 0xD0 to 0xD7 ther
 
 // Start of Frame markers, nondifferential Huffman-coding frames
 constexpr byte sof0_ = 0xc0;  //!< JPEG Start-Of-Frame marker
+constexpr byte sof1_ = 0xc1;  //!< JPEG Start-Of-Frame marker
+constexpr byte sof2_ = 0xc2;  //!< JPEG Start-Of-Frame marker
 constexpr byte sof3_ = 0xc3;  //!< JPEG Start-Of-Frame marker
 
 // Start of Frame markers, differential Huffman-coding frames
 constexpr byte sof5_ = 0xc5;  //!< JPEG Start-Of-Frame marker
+constexpr byte sof6_ = 0xc6;  //!< JPEG Start-Of-Frame marker
+constexpr byte sof7_ = 0xc6;  //!< JPEG Start-Of-Frame marker
 
 // Start of Frame markers, differential arithmetic-coding frames
+constexpr byte sof9_ = 0xc9;   //!< JPEG Start-Of-Frame marker
+constexpr byte sof10_ = 0xca;  //!< JPEG Start-Of-Frame marker
+constexpr byte sof11_ = 0xcb;  //!< JPEG Start-Of-Frame marker
+constexpr byte sof13_ = 0xcd;  //!< JPEG Start-Of-Frame marker
+constexpr byte sof14_ = 0xce;  //!< JPEG Start-Of-Frame marker
 constexpr byte sof15_ = 0xcf;  //!< JPEG Start-Of-Frame marker
 
+// JPEG process SOF markers
+constexpr Internal::TagDetails jpegProcessMarkerTags[] = {
+    {sof0_, N_("Baseline DCT, Huffman coding")},
+    {sof1_, N_("Extended sequential DCT, Huffman coding")},
+    {sof2_, N_("Progressive DCT, Huffman coding")},
+    {sof3_, N_("Lossless, Huffman coding")},
+    {sof5_, N_("Sequential DCT, differential Huffman coding")},
+    {sof6_, N_("Progressive DCT, differential Huffman coding")},
+    {sof7_, N_("Lossless, Differential Huffman coding")},
+    {sof9_, N_("Extended sequential DCT, arithmetic coding")},
+    {sof10_, N_("Progressive DCT, arithmetic coding")},
+    {sof11_, N_("Lossless, arithmetic coding")},
+    {sof13_, N_("Sequential DCT, differential arithmetic coding")},
+    {sof14_, N_("Progressive DCT, differential arithmetic coding")},
+    {sof15_, N_("Lossless, differential arithmetic coding")},
+};
+
 constexpr auto exifId_ = "Exif\0\0";  //!< Exif identifier
 // constexpr auto jfifId_ = "JFIF\0";                         //!< JFIF identifier
 constexpr auto xmpId_ = "http://ns.adobe.com/xap/1.0/\0";  //!< XMP packet identifier
@@ -159,6 +187,13 @@ void JpegBase::readMetadata() {
       std::copy(sizebuf.begin(), sizebuf.end(), buf.begin());
     }
 
+    if (auto itSofMarker = Exiv2::find(jpegProcessMarkerTags, marker)) {
+      sof_encoding_process_ = itSofMarker->label_;
+      if (size >= 7 && buf.c_data(7)) {
+        num_color_components_ = *buf.c_data(7);
+      }
+    }
+
     if (!foundExifData && marker == app1_ && size >= 8  // prevent out-of-bounds read in memcmp on next line
         && buf.cmpBytes(2, exifId_, 6) == 0) {
       ByteOrder bo = ExifParser::decode(exifData_, buf.c_data(8), size - 8);

tests/bash_tests/test_jpegparse.py

@@ -0,0 +1,13 @@
+# -*- coding: utf-8 -*-
+
+import system_tests
+
+class TestJpegParse(metaclass=system_tests.CaseMeta):
+    url = "https://github.com/Exiv2/exiv2/pull/2874"
+
+    # Test extraction of jpeg specific metadata
+    filename = "$data_path/exiv2-fujifilm-finepix-s2pro.jpg"
+    commands = ["$jpegparsetest $filename"]
+    stderr = [""] * len(commands)
+    stdout = ["""Number of color components: 3\nEncoding process: Baseline DCT, Huffman coding\n"""]
+    retval = [0] * len(commands)

tests/suite.conf

@@ -35,6 +35,7 @@ tiff_test: ${ENV:exiv2_path}/tiff-test
 largeiptc_test: ${ENV:exiv2_path}/largeiptc-test
 easyaccess_test: ${ENV:exiv2_path}/easyaccess-test
 taglist: ${ENV:exiv2_path}/taglist
+jpegparsetest: ${ENV:exiv2_path}/jpegparsetest
 
 [variables]
 kerOffsetOutOfRange: Offset out of range