APIs and tests for retrieving raw/decoded data from image objects

Added FPDFImageObj_GetImageDataDecoded() for retrieving the uncompressed
data of an image, and FPDFImageObj_GetImageDataRaw() for retrieving the
raw data of an image.
    * Refactored out DecodeStreamMaybeCopyAndReturnLength(), which is
      used to decode both attachment data and image data.
    * Within DecodeStreamMaybeCopyAndReturnLength(), used a different
      decoder function which takes care of multiple filters if exist. As
      a result, CPDF_StreamParser::DecodeInlineStream() which was made
      static previously is now moved back into namespace.

Bug=pdfium:677

Change-Id: I22a22c99acaca98ef8c15f88911f2646a2c854d5
Reviewed-on: https://pdfium-review.googlesource.com/9811
Commit-Queue: Jane Liu <janeliulwq@google.com>
Reviewed-by: Lei Zhang <thestig@chromium.org>

1 files changed, 51 insertions, 0 deletions

diff --git a/fpdfsdk/fpdfedit_embeddertest.cpp b/fpdfsdk/fpdfedit_embeddertest.cpp
index dcaeb945d7..f1bbb87422 100644
--- a/fpdfsdk/fpdfedit_embeddertest.cpp
+++ b/fpdfsdk/fpdfedit_embeddertest.cpp
@@ -5,6 +5,7 @@
 #include <memory>
 #include <string>
 #include <utility>
+#include <vector>
 
 #include "core/fpdfapi/font/cpdf_font.h"
 #include "core/fpdfapi/page/cpdf_page.h"
@@ -979,3 +980,53 @@ TEST_F(FPDFEditEmbeddertest, ExtractImageBitmap) {
   FPDFBitmap_Destroy(bitmap);
   UnloadPage(page);
 }
+
+TEST_F(FPDFEditEmbeddertest, GetImageData) {
+  EXPECT_TRUE(OpenDocument("embedded_images.pdf"));
+  FPDF_PAGE page = LoadPage(0);
+  ASSERT_TRUE(page);
+  ASSERT_EQ(39, FPDFPage_CountObject(page));
+
+  // Retrieve an image object with flate-encoded data stream.
+  FPDF_PAGEOBJECT obj = FPDFPage_GetObject(page, 33);
+  ASSERT_EQ(FPDF_PAGEOBJ_IMAGE, FPDFPageObj_GetType(obj));
+
+  // Check that the raw image data has the correct length and hash value.
+  unsigned long len = FPDFImageObj_GetImageDataRaw(obj, nullptr, 0);
+  std::vector<char> buf(len);
+  EXPECT_EQ(4091u, FPDFImageObj_GetImageDataRaw(obj, buf.data(), len));
+  EXPECT_EQ("f73802327d2e88e890f653961bcda81a",
+            GenerateMD5Base16(reinterpret_cast<uint8_t*>(buf.data()), len));
+
+  // Check that the decoded image data has the correct length and hash value.
+  len = FPDFImageObj_GetImageDataDecoded(obj, nullptr, 0);
+  buf.clear();
+  buf.resize(len);
+  EXPECT_EQ(28776u, FPDFImageObj_GetImageDataDecoded(obj, buf.data(), len));
+  EXPECT_EQ("cb3637934bb3b95a6e4ae1ea9eb9e56e",
+            GenerateMD5Base16(reinterpret_cast<uint8_t*>(buf.data()), len));
+
+  // Retrieve an image obejct with DCTDecode-encoded data stream.
+  obj = FPDFPage_GetObject(page, 37);
+  ASSERT_EQ(FPDF_PAGEOBJ_IMAGE, FPDFPageObj_GetType(obj));
+
+  // Check that the raw image data has the correct length and hash value.
+  len = FPDFImageObj_GetImageDataRaw(obj, nullptr, 0);
+  buf.clear();
+  buf.resize(len);
+  EXPECT_EQ(4370u, FPDFImageObj_GetImageDataRaw(obj, buf.data(), len));
+  EXPECT_EQ("6aae1f3710335023a9e12191be66b64b",
+            GenerateMD5Base16(reinterpret_cast<uint8_t*>(buf.data()), len));
+
+  // Check that the decoded image data has the correct length and hash value,
+  // which should be the same as those of the raw data, since this image is
+  // encoded by a single DCTDecode filter and decoding is a noop.
+  len = FPDFImageObj_GetImageDataDecoded(obj, nullptr, 0);
+  buf.clear();
+  buf.resize(len);
+  EXPECT_EQ(4370u, FPDFImageObj_GetImageDataDecoded(obj, buf.data(), len));
+  EXPECT_EQ("6aae1f3710335023a9e12191be66b64b",
+            GenerateMD5Base16(reinterpret_cast<uint8_t*>(buf.data()), len));
+
+  UnloadPage(page);
+}