Expose marked content IDs for images and alt text.

 - Adds two new public bits of API:
    - StructElement_GetMCID(STRUCTELEMENT) to return the marked content
      ID of the struct element.
    - IMAGEOBJ_METADATA.mcid to contain the marked content ID of the
      image.
 - Restores ContentMark::GetMCID, which was removed with other dead
   code in https://pdfium-review.googlesource.com/c/pdfium/+/19170.
 - Adds a couple calls to tests, including a simple new PDF to test
   the struct tree marked content ID.

Bug:pdfium:957
Change-Id: I92856e43d741df989e53a575a08258da19a39f22
Reviewed-on: https://pdfium-review.googlesource.com/20632
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>

1 files changed, 14 insertions, 0 deletions

diff --git a/core/fpdfapi/page/cpdf_contentmark.cpp b/core/fpdfapi/page/cpdf_contentmark.cpp
index 1d70edeb51..c54ef55b1f 100644
--- a/core/fpdfapi/page/cpdf_contentmark.cpp
+++ b/core/fpdfapi/page/cpdf_contentmark.cpp
@@ -25,6 +25,11 @@ const CPDF_ContentMarkItem& CPDF_ContentMark::GetItem(size_t i) const {
   return m_Ref.GetObject()->GetItem(i);
 }
 
+int CPDF_ContentMark::GetMarkedContentID() const {
+  const MarkData* pData = m_Ref.GetObject();
+  return pData ? pData->GetMarkedContentID() : -1;
+}
+
 void CPDF_ContentMark::AddMark(const ByteString& name,
                                CPDF_Dictionary* pDict,
                                bool bDirect) {
@@ -53,6 +58,15 @@ const CPDF_ContentMarkItem& CPDF_ContentMark::MarkData::GetItem(
   return m_Marks[index];
 }
 
+int CPDF_ContentMark::MarkData::GetMarkedContentID() const {
+  for (const auto& mark : m_Marks) {
+    const CPDF_Dictionary* pDict = mark.GetParam();
+    if (pDict && pDict->KeyExist("MCID"))
+      return pDict->GetIntegerFor("MCID");
+  }
+  return -1;
+}
+
 void CPDF_ContentMark::MarkData::AddMark(const ByteString& name,
                                          CPDF_Dictionary* pDict,
                                          bool bDirect) {