From 7bbf5a2e79389026895870566196b5e13f224f72 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 14:06:49 -0500
Subject: [PATCH 01/15] add base64 encoder/decoder to bytevector

---
 taglib/toolkit/tbytevector.cpp | 92 ++++++++++++++++++++++++++++++++++
 taglib/toolkit/tbytevector.h   | 13 +++++
 tests/test_bytevector.cpp      | 40 +++++++++++++++
 3 files changed, 145 insertions(+)

diff --git a/taglib/toolkit/tbytevector.cpp b/taglib/toolkit/tbytevector.cpp
index 6262bec6..eb559390 100644
--- a/taglib/toolkit/tbytevector.cpp
+++ b/taglib/toolkit/tbytevector.cpp
@@ -950,6 +950,98 @@ ByteVector ByteVector::toHex() const
   return encoded;
 }
 
+
+
+
+
+ByteVector & ByteVector::fromBase64()
+{
+  static const unsigned char base64[256]={
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x3e,0x80,0x80,0x80,0x3f,
+    0x34,0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x80,0x80,0x80,0xff,0x80,0x80,
+    0x80,0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,
+    0x0f,0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17,0x18,0x19,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27,0x28,
+    0x29,0x2a,0x2b,0x2c,0x2d,0x2e,0x2f,0x30,0x31,0x32,0x33,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
+    0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80
+  };
+
+  detach();
+  uint len = size();
+  const unsigned char * src = (unsigned char*) data();
+  unsigned char * dst = (unsigned char*) data();
+  while(4<=len) {
+    if(base64[src[0]]==0x80) break;
+    if(base64[src[1]]==0x80) break;
+    if(base64[src[2]]==0x80) break;
+    if(base64[src[3]]==0x80) break;
+    *dst++=((base64[src[0]]<<2)&0xfc)|((base64[src[1]]>>4)&0x03);
+    if(src[2]!='=') {
+      *dst++=((base64[src[1]]&0x0f)<<4)|((base64[src[2]]>>2)&0x0f);
+      if(src[3]!='=') {
+        *dst++=((base64[src[2]]&0x03)<<6)|(base64[src[3]]&0x3f);
+      }
+      else {
+        break;
+      }
+    }
+    else {
+      break;
+    }
+    src+=4;
+    len-=4;
+    }
+  resize(dst-(unsigned char*)data());
+  return *this;
+}
+
+
+
+
+ByteVector ByteVector::toBase64() const
+{
+  static const char alphabet[]="ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
+  if (size()) {
+    uint len=size();
+    ByteVector output(4 * ((len-1)/3+1)); // note roundup
+
+    const char * src = data();
+    char * dst = output.data();
+    while(3<=len) {
+      *dst++=alphabet[(src[0]>>2)&0x3f];
+      *dst++=alphabet[((src[0]&0x03)<<4)|((src[1]>>4)&0x0f)];
+      *dst++=alphabet[((src[1]&0x0f)<<2)|((src[2]>>6)&0x03)];
+      *dst++=alphabet[src[2]&0x3f];
+      src+=3;
+      len-=3;
+    }
+    if(len) {
+      *dst++=alphabet[(src[0]>>2)&0x3f];
+      if(len>1) {
+        *dst++=alphabet[((src[0]&0x03)<<4)|((src[1]>>4)&0x0f)];
+        *dst++=alphabet[((src[1]&0x0f)<<2)];
+      }
+      else {
+        *dst++=alphabet[(src[0]&0x03)<<4];
+        *dst++='=';
+      }
+    *dst++='=';
+    }
+    return output;
+  }
+  return ByteVector();
+}
+
+
 ////////////////////////////////////////////////////////////////////////////////
 // protected members
 ////////////////////////////////////////////////////////////////////////////////
diff --git a/taglib/toolkit/tbytevector.h b/taglib/toolkit/tbytevector.h
index 793b414c..0f96c77f 100644
--- a/taglib/toolkit/tbytevector.h
+++ b/taglib/toolkit/tbytevector.h
@@ -573,6 +573,19 @@ namespace TagLib {
      */
     ByteVector toHex() const;
 
+    /*!
+     * Returns a base64 encoded copy of the byte vector
+     */
+    ByteVector toBase64() const;
+
+    /*!
+     * Decodes the base64 encoded byte vector in-memory. Returns a reference
+     * to this vector. Calls detach before decoding.
+     */
+    ByteVector & fromBase64();
+
+
+
   protected:
     /*
      * If this ByteVector is being shared via implicit sharing, do a deep copy
diff --git a/tests/test_bytevector.cpp b/tests/test_bytevector.cpp
index 1c2ca904..19f51518 100644
--- a/tests/test_bytevector.cpp
+++ b/tests/test_bytevector.cpp
@@ -43,6 +43,7 @@ class TestByteVector : public CppUnit::TestFixture
   CPPUNIT_TEST(testReplace);
   CPPUNIT_TEST(testIterator);
   CPPUNIT_TEST(testResize);
+  CPPUNIT_TEST(testBase64);
   CPPUNIT_TEST_SUITE_END();
 
 public:
@@ -379,6 +380,45 @@ public:
     CPPUNIT_ASSERT_EQUAL(-1, c.find('C'));
   }
 
+  void testBase64()
+  {
+    ByteVector sempty;
+    ByteVector t0("a"); // test 1 byte
+    ByteVector t1("any carnal pleasure.");
+    ByteVector t2("any carnal pleasure");
+    ByteVector t3("any carnal pleasur");
+    ByteVector s0("a"); // test 1 byte
+    ByteVector s1("any carnal pleasure.");
+    ByteVector s2("any carnal pleasure");
+    ByteVector s3("any carnal pleasur");
+    ByteVector eempty;
+    ByteVector e0("YQ==");
+    ByteVector e1("YW55IGNhcm5hbCBwbGVhc3VyZS4=");
+    ByteVector e2("YW55IGNhcm5hbCBwbGVhc3VyZQ==");
+    ByteVector e3("YW55IGNhcm5hbCBwbGVhc3Vy");
+
+    // Encode
+    CPPUNIT_ASSERT_EQUAL(eempty, sempty.toBase64());
+    CPPUNIT_ASSERT_EQUAL(e0, s0.toBase64());
+    CPPUNIT_ASSERT_EQUAL(e1, s1.toBase64());
+    CPPUNIT_ASSERT_EQUAL(e2, s2.toBase64());
+    CPPUNIT_ASSERT_EQUAL(e3, s3.toBase64());
+
+    // Decode
+    CPPUNIT_ASSERT_EQUAL(sempty, eempty.toBase64());
+    CPPUNIT_ASSERT_EQUAL(s0, e0.fromBase64());
+    CPPUNIT_ASSERT_EQUAL(s1, e1.fromBase64());
+    CPPUNIT_ASSERT_EQUAL(s2, e2.fromBase64());
+    CPPUNIT_ASSERT_EQUAL(s3, e3.fromBase64());
+
+    CPPUNIT_ASSERT_EQUAL(t0, s0.toBase64().fromBase64());
+    CPPUNIT_ASSERT_EQUAL(t1, s1.toBase64().fromBase64());
+    CPPUNIT_ASSERT_EQUAL(t2, s2.toBase64().fromBase64());
+    CPPUNIT_ASSERT_EQUAL(t3, s3.toBase64().fromBase64());
+
+  }
+
+
 };
 
 CPPUNIT_TEST_SUITE_REGISTRATION(TestByteVector);

From 0c14bd0ed075d140ef55fc62efc81b8c3ad841c8 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 14:10:24 -0500
Subject: [PATCH 02/15] Add FLAC::Picture support to Xiph Comment

---
 taglib/ogg/xiphcomment.cpp | 87 +++++++++++++++++++++++++++++++++-----
 taglib/ogg/xiphcomment.h   | 26 ++++++++++++
 tests/test_ogg.cpp         | 28 ++++++++++++
 3 files changed, 130 insertions(+), 11 deletions(-)

diff --git a/taglib/ogg/xiphcomment.cpp b/taglib/ogg/xiphcomment.cpp
index 9462607f..c307e618 100644
--- a/taglib/ogg/xiphcomment.cpp
+++ b/taglib/ogg/xiphcomment.cpp
@@ -26,17 +26,22 @@
 #include <tbytevector.h>
 #include <tdebug.h>
 
+#include <flacpicture.h>
 #include <xiphcomment.h>
 #include <tpropertymap.h>
 
 using namespace TagLib;
 
+
+typedef List<FLAC::Picture*> PictureList;
+
 class Ogg::XiphComment::XiphCommentPrivate
 {
 public:
   FieldListMap fieldListMap;
   String vendorID;
   String commentField;
+  PictureList pictureList;
 };
 
 ////////////////////////////////////////////////////////////////////////////////
@@ -181,6 +186,8 @@ TagLib::uint Ogg::XiphComment::fieldCount() const
   for(; it != d->fieldListMap.end(); ++it)
     count += (*it).second.size();
 
+  count += d->pictureList.size();
+
   return count;
 }
 
@@ -275,6 +282,36 @@ bool Ogg::XiphComment::contains(const String &key) const
   return d->fieldListMap.contains(key) && !d->fieldListMap[key].isEmpty();
 }
 
+void Ogg::XiphComment::removePicture(FLAC::Picture *picture, bool del)
+{
+  List<FLAC::Picture *>::Iterator it = d->pictureList.find(picture);
+  if(it != d->pictureList.end())
+    d->pictureList.erase(it);
+
+  if(del)
+    delete picture;
+}
+
+void Ogg::XiphComment::removePictures()
+{
+  PictureList newList;
+  for(uint i = 0; i < d->pictureList.size(); i++) {
+    delete d->pictureList[i];
+  }
+  d->pictureList = newList;
+}
+
+void Ogg::XiphComment::addPicture(FLAC::Picture * picture)
+{
+  d->pictureList.append(picture);
+}
+
+
+List<FLAC::Picture *> Ogg::XiphComment::pictureList()
+{
+  return d->pictureList;
+}
+
 ByteVector Ogg::XiphComment::render() const
 {
   return render(true);
@@ -321,6 +358,13 @@ ByteVector Ogg::XiphComment::render(bool addFramingBit) const
     }
   }
 
+  for(PictureList::ConstIterator it = d->pictureList.begin(); it != d->pictureList.end(); ++it) {
+    ByteVector picture = (*it)->render().toBase64();
+    data.append(ByteVector::fromUInt(picture.size()+23,false));
+    data.append("METADATA_BLOCK_PICTURE=");
+    data.append(picture);
+  }
+
   // Append the "framing bit".
 
   if(addFramingBit)
@@ -363,20 +407,41 @@ void Ogg::XiphComment::parse(const ByteVector &data)
     const uint commentLength = data.toUInt(pos, false);
     pos += 4;
 
-    String comment = String(data.mid(pos, commentLength), String::UTF8);
-    pos += commentLength;
-    if(pos > data.size()) {
+    ByteVector entry = data.mid(pos, commentLength);
+
+    // Don't go past data end
+    pos+=commentLength;
+    if (pos>data.size())
       break;
+
+    // Handle Pictures separately
+    if(entry.startsWith("METADATA_BLOCK_PICTURE=")) {
+
+      // Decode base64 picture data
+      ByteVector picturedata = entry.mid(23, entry.size()-23).fromBase64();
+
+      if(picturedata.size()==0) {
+        debug("Empty picture data. Discarding content");
+        continue;
+        }
+
+      FLAC::Picture * picture = new FLAC::Picture();
+      if(picture->parse(picturedata))
+        d->pictureList.append(picture);
+      else
+        debug("Unable to parse METADATA_BLOCK_PICTURE. Discarding content.");
     }
+    else {
 
-    int commentSeparatorPosition = comment.find("=");
-    if(commentSeparatorPosition == -1) {
-      break;
+      // Check for field separator
+      int sep = entry.find('=');
+      if (sep == -1)
+        break;
+
+      // Parse key and value
+      String key = String(entry.mid(0,sep), String::UTF8);
+      String value = String(entry.mid(sep+1, commentLength-sep), String::UTF8);
+      addField(key, value, false);
     }
-
-    String key = comment.substr(0, commentSeparatorPosition);
-    String value = comment.substr(commentSeparatorPosition + 1);
-
-    addField(key, value, false);
   }
 }
diff --git a/taglib/ogg/xiphcomment.h b/taglib/ogg/xiphcomment.h
index 54f3070b..8fcc6873 100644
--- a/taglib/ogg/xiphcomment.h
+++ b/taglib/ogg/xiphcomment.h
@@ -32,6 +32,7 @@
 #include "tstring.h"
 #include "tstringlist.h"
 #include "tbytevector.h"
+#include "flacpicture.h"
 #include "taglib_export.h"
 
 namespace TagLib {
@@ -205,6 +206,31 @@ namespace TagLib {
        */
       ByteVector render(bool addFramingBit) const;
 
+
+      /*!
+       * Returns a list of pictures attached to the xiph comment.
+       */
+      List<FLAC::Picture *> pictureList();
+
+      /*!
+       * Removes an picture. If \a del is true the picture's memory
+       * will be freed; if it is false, it must be deleted by the user.
+       */
+      void removePicture(FLAC::Picture *picture, bool del = true);
+
+      /*!
+       * Remove all pictures.
+       */
+      void removePictures();
+
+      /*!
+       * Add a new picture to the comment block. The comment block takes ownership of the
+       * picture and will handle freeing its memory.
+       *
+       * \note The file will be saved only after calling save().
+       */
+      void addPicture(FLAC::Picture *picture);
+
     protected:
       /*!
        * Reads the tag from the file specified in the constructor and fills the
diff --git a/tests/test_ogg.cpp b/tests/test_ogg.cpp
index 210fca2f..05b23f64 100644
--- a/tests/test_ogg.cpp
+++ b/tests/test_ogg.cpp
@@ -21,6 +21,7 @@ class TestOGG : public CppUnit::TestFixture
   CPPUNIT_TEST(testDictInterface1);
   CPPUNIT_TEST(testDictInterface2);
   CPPUNIT_TEST(testAudioProperties);
+  CPPUNIT_TEST(testPicture);
   CPPUNIT_TEST_SUITE_END();
 
 public:
@@ -134,6 +135,33 @@ public:
     CPPUNIT_ASSERT_EQUAL(112000, f.audioProperties()->bitrateNominal());
     CPPUNIT_ASSERT_EQUAL(0, f.audioProperties()->bitrateMinimum());
   }
+
+  void testPicture()
+  {
+    ScopedFileCopy copy("empty", ".ogg");
+    string newname = copy.fileName();
+
+    Vorbis::File *f = new Vorbis::File(newname.c_str());
+    FLAC::Picture *newpic = new FLAC::Picture();
+    newpic->setType(FLAC::Picture::BackCover);
+    newpic->setWidth(5);
+    newpic->setHeight(6);
+    newpic->setColorDepth(16);
+    newpic->setNumColors(7);
+    newpic->setMimeType("image/jpeg");
+    newpic->setDescription("new image");
+    newpic->setData("JPEG data");
+    f->tag()->addPicture(newpic);
+    f->save();
+    delete f;
+
+    f = new Vorbis::File(newname.c_str());
+    List<FLAC::Picture *> lst = f->tag()->pictureList();
+    CPPUNIT_ASSERT_EQUAL(TagLib::uint(1), lst.size());
+    delete f;
+  }
+
+
 };
 
 CPPUNIT_TEST_SUITE_REGISTRATION(TestOGG);

From d3351a0012d89bd78e032859597e4f4ef40efdd5 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 14:14:06 -0500
Subject: [PATCH 03/15] Delete pictures in XiphComment destructor

---
 taglib/ogg/xiphcomment.cpp | 1 +
 1 file changed, 1 insertion(+)

diff --git a/taglib/ogg/xiphcomment.cpp b/taglib/ogg/xiphcomment.cpp
index c307e618..2783d922 100644
--- a/taglib/ogg/xiphcomment.cpp
+++ b/taglib/ogg/xiphcomment.cpp
@@ -61,6 +61,7 @@ Ogg::XiphComment::XiphComment(const ByteVector &data) : TagLib::Tag()
 
 Ogg::XiphComment::~XiphComment()
 {
+  removePictures();
   delete d;
 }
 

From 62bb3c95c80e6c1646a54cb3b96bc01bdebe4ff2 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 14:51:10 -0500
Subject: [PATCH 04/15] Make ByteVector::fromBase64 as static member function

---
 taglib/ogg/xiphcomment.cpp     |  2 +-
 taglib/toolkit/tbytevector.cpp | 16 +++++++++-------
 taglib/toolkit/tbytevector.h   |  7 ++-----
 tests/test_bytevector.cpp      | 16 ++++++++--------
 4 files changed, 20 insertions(+), 21 deletions(-)

diff --git a/taglib/ogg/xiphcomment.cpp b/taglib/ogg/xiphcomment.cpp
index 2783d922..15f8cbaf 100644
--- a/taglib/ogg/xiphcomment.cpp
+++ b/taglib/ogg/xiphcomment.cpp
@@ -419,7 +419,7 @@ void Ogg::XiphComment::parse(const ByteVector &data)
     if(entry.startsWith("METADATA_BLOCK_PICTURE=")) {
 
       // Decode base64 picture data
-      ByteVector picturedata = entry.mid(23, entry.size()-23).fromBase64();
+      ByteVector picturedata = ByteVector::fromBase64(entry.mid(23));
 
       if(picturedata.size()==0) {
         debug("Empty picture data. Discarding content");
diff --git a/taglib/toolkit/tbytevector.cpp b/taglib/toolkit/tbytevector.cpp
index eb559390..f18d2329 100644
--- a/taglib/toolkit/tbytevector.cpp
+++ b/taglib/toolkit/tbytevector.cpp
@@ -954,7 +954,7 @@ ByteVector ByteVector::toHex() const
 
 
 
-ByteVector & ByteVector::fromBase64()
+ByteVector ByteVector::fromBase64(const ByteVector & input)
 {
   static const unsigned char base64[256]={
     0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
@@ -975,10 +975,12 @@ ByteVector & ByteVector::fromBase64()
     0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80
   };
 
-  detach();
-  uint len = size();
-  const unsigned char * src = (unsigned char*) data();
-  unsigned char * dst = (unsigned char*) data();
+  uint len = input.size();
+
+  ByteVector output(len);
+
+  const unsigned char * src = (unsigned char*) input.data();
+  unsigned char * dst = (unsigned char*)output.data();
   while(4<=len) {
     if(base64[src[0]]==0x80) break;
     if(base64[src[1]]==0x80) break;
@@ -1000,8 +1002,8 @@ ByteVector & ByteVector::fromBase64()
     src+=4;
     len-=4;
     }
-  resize(dst-(unsigned char*)data());
-  return *this;
+  output.resize(dst-(unsigned char*)output.data());
+  return output;
 }
 
 
diff --git a/taglib/toolkit/tbytevector.h b/taglib/toolkit/tbytevector.h
index 0f96c77f..4b669b9e 100644
--- a/taglib/toolkit/tbytevector.h
+++ b/taglib/toolkit/tbytevector.h
@@ -579,12 +579,9 @@ namespace TagLib {
     ByteVector toBase64() const;
 
     /*!
-     * Decodes the base64 encoded byte vector in-memory. Returns a reference
-     * to this vector. Calls detach before decoding.
+     * Decodes the base64 encoded byte vector.      
      */
-    ByteVector & fromBase64();
-
-
+    static ByteVector fromBase64(const ByteVector &);
 
   protected:
     /*
diff --git a/tests/test_bytevector.cpp b/tests/test_bytevector.cpp
index 19f51518..18097adf 100644
--- a/tests/test_bytevector.cpp
+++ b/tests/test_bytevector.cpp
@@ -406,15 +406,15 @@ public:
 
     // Decode
     CPPUNIT_ASSERT_EQUAL(sempty, eempty.toBase64());
-    CPPUNIT_ASSERT_EQUAL(s0, e0.fromBase64());
-    CPPUNIT_ASSERT_EQUAL(s1, e1.fromBase64());
-    CPPUNIT_ASSERT_EQUAL(s2, e2.fromBase64());
-    CPPUNIT_ASSERT_EQUAL(s3, e3.fromBase64());
+    CPPUNIT_ASSERT_EQUAL(s0, ByteVector::fromBase64(e0));
+    CPPUNIT_ASSERT_EQUAL(s1, ByteVector::fromBase64(e1));
+    CPPUNIT_ASSERT_EQUAL(s2, ByteVector::fromBase64(e2));
+    CPPUNIT_ASSERT_EQUAL(s3, ByteVector::fromBase64(e3));
 
-    CPPUNIT_ASSERT_EQUAL(t0, s0.toBase64().fromBase64());
-    CPPUNIT_ASSERT_EQUAL(t1, s1.toBase64().fromBase64());
-    CPPUNIT_ASSERT_EQUAL(t2, s2.toBase64().fromBase64());
-    CPPUNIT_ASSERT_EQUAL(t3, s3.toBase64().fromBase64());
+    CPPUNIT_ASSERT_EQUAL(t0, ByteVector::fromBase64(s0.toBase64()));
+    CPPUNIT_ASSERT_EQUAL(t1, ByteVector::fromBase64(s1.toBase64()));
+    CPPUNIT_ASSERT_EQUAL(t2, ByteVector::fromBase64(s2.toBase64()));
+    CPPUNIT_ASSERT_EQUAL(t3, ByteVector::fromBase64(s3.toBase64()));
 
   }
 

From 59a1b7a4916e3cf47945ba965961249db21fb9db Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 14:53:17 -0500
Subject: [PATCH 05/15] add forgotten const cast

---
 taglib/toolkit/tbytevector.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/taglib/toolkit/tbytevector.cpp b/taglib/toolkit/tbytevector.cpp
index f18d2329..a87922d5 100644
--- a/taglib/toolkit/tbytevector.cpp
+++ b/taglib/toolkit/tbytevector.cpp
@@ -979,7 +979,7 @@ ByteVector ByteVector::fromBase64(const ByteVector & input)
 
   ByteVector output(len);
 
-  const unsigned char * src = (unsigned char*) input.data();
+  const unsigned char * src = (const unsigned char*) input.data();
   unsigned char * dst = (unsigned char*)output.data();
   while(4<=len) {
     if(base64[src[0]]==0x80) break;

From c4a02a17991a857ad01d9ed61aad7da167faec4e Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 14:57:19 -0500
Subject: [PATCH 06/15] remove redundant size specificier in mid usage

---
 taglib/ogg/xiphcomment.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/taglib/ogg/xiphcomment.cpp b/taglib/ogg/xiphcomment.cpp
index 15f8cbaf..431c7980 100644
--- a/taglib/ogg/xiphcomment.cpp
+++ b/taglib/ogg/xiphcomment.cpp
@@ -441,7 +441,7 @@ void Ogg::XiphComment::parse(const ByteVector &data)
 
       // Parse key and value
       String key = String(entry.mid(0,sep), String::UTF8);
-      String value = String(entry.mid(sep+1, commentLength-sep), String::UTF8);
+      String value = String(entry.mid(sep+1), String::UTF8);
       addField(key, value, false);
     }
   }

From c963d3ba04d0cd4ad227c2703d978c023d659833 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 15:05:05 -0500
Subject: [PATCH 07/15] correctly filter out invalid base64 characters

---
 taglib/toolkit/tbytevector.cpp | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/taglib/toolkit/tbytevector.cpp b/taglib/toolkit/tbytevector.cpp
index a87922d5..377dc034 100644
--- a/taglib/toolkit/tbytevector.cpp
+++ b/taglib/toolkit/tbytevector.cpp
@@ -960,7 +960,7 @@ ByteVector ByteVector::fromBase64(const ByteVector & input)
     0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
     0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
     0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x3e,0x80,0x80,0x80,0x3f,
-    0x34,0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x80,0x80,0x80,0xff,0x80,0x80,
+    0x34,0x35,0x36,0x37,0x38,0x39,0x3a,0x3b,0x3c,0x3d,0x80,0x80,0x80,0x80,0x80,0x80,
     0x80,0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0x0a,0x0b,0x0c,0x0d,0x0e,
     0x0f,0x10,0x11,0x12,0x13,0x14,0x15,0x16,0x17,0x18,0x19,0x80,0x80,0x80,0x80,0x80,
     0x80,0x1a,0x1b,0x1c,0x1d,0x1e,0x1f,0x20,0x21,0x22,0x23,0x24,0x25,0x26,0x27,0x28,
@@ -984,12 +984,12 @@ ByteVector ByteVector::fromBase64(const ByteVector & input)
   while(4<=len) {
     if(base64[src[0]]==0x80) break;
     if(base64[src[1]]==0x80) break;
-    if(base64[src[2]]==0x80) break;
-    if(base64[src[3]]==0x80) break;
     *dst++=((base64[src[0]]<<2)&0xfc)|((base64[src[1]]>>4)&0x03);
     if(src[2]!='=') {
+      if(base64[src[2]]==0x80) break;
       *dst++=((base64[src[1]]&0x0f)<<4)|((base64[src[2]]>>2)&0x0f);
       if(src[3]!='=') {
+        if(base64[src[3]]==0x80) break;
         *dst++=((base64[src[2]]&0x03)<<6)|(base64[src[3]]&0x3f);
       }
       else {

From f4c1beb16147dee202824ff693d55da69eb110be Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 21:25:06 -0500
Subject: [PATCH 08/15] Make fromBase64 more strict and update code to follow
 taglib coding style

---
 taglib/ogg/xiphcomment.cpp     | 22 ++++----
 taglib/toolkit/tbytevector.cpp | 99 +++++++++++++++++++++-------------
 2 files changed, 74 insertions(+), 47 deletions(-)

diff --git a/taglib/ogg/xiphcomment.cpp b/taglib/ogg/xiphcomment.cpp
index 431c7980..6b87e3ea 100644
--- a/taglib/ogg/xiphcomment.cpp
+++ b/taglib/ogg/xiphcomment.cpp
@@ -33,7 +33,7 @@
 using namespace TagLib;
 
 
-typedef List<FLAC::Picture*> PictureList;
+typedef List<FLAC::Picture *> PictureList;
 
 class Ogg::XiphComment::XiphCommentPrivate
 {
@@ -285,7 +285,7 @@ bool Ogg::XiphComment::contains(const String &key) const
 
 void Ogg::XiphComment::removePicture(FLAC::Picture *picture, bool del)
 {
-  List<FLAC::Picture *>::Iterator it = d->pictureList.find(picture);
+  PictureList::Iterator it = d->pictureList.find(picture);
   if(it != d->pictureList.end())
     d->pictureList.erase(it);
 
@@ -410,9 +410,10 @@ void Ogg::XiphComment::parse(const ByteVector &data)
 
     ByteVector entry = data.mid(pos, commentLength);
 
+    pos += commentLength;
+
     // Don't go past data end
-    pos+=commentLength;
-    if (pos>data.size())
+    if (pos > data.size())
       break;
 
     // Handle Pictures separately
@@ -421,12 +422,13 @@ void Ogg::XiphComment::parse(const ByteVector &data)
       // Decode base64 picture data
       ByteVector picturedata = ByteVector::fromBase64(entry.mid(23));
 
-      if(picturedata.size()==0) {
+      if(picturedata.size() == 0) {
         debug("Empty picture data. Discarding content");
         continue;
-        }
+      }
 
       FLAC::Picture * picture = new FLAC::Picture();
+
       if(picture->parse(picturedata))
         d->pictureList.append(picture);
       else
@@ -436,11 +438,13 @@ void Ogg::XiphComment::parse(const ByteVector &data)
 
       // Check for field separator
       int sep = entry.find('=');
-      if (sep == -1)
-        break;
+      if (sep < 1) {
+        debug("Discarding invalid comment field.");
+        continue;
+      }
 
       // Parse key and value
-      String key = String(entry.mid(0,sep), String::UTF8);
+      String key = String(entry.mid(0, sep), String::UTF8);
       String value = String(entry.mid(sep+1), String::UTF8);
       addField(key, value, false);
     }
diff --git a/taglib/toolkit/tbytevector.cpp b/taglib/toolkit/tbytevector.cpp
index 377dc034..14550ad1 100644
--- a/taglib/toolkit/tbytevector.cpp
+++ b/taglib/toolkit/tbytevector.cpp
@@ -950,13 +950,9 @@ ByteVector ByteVector::toHex() const
   return encoded;
 }
 
-
-
-
-
 ByteVector ByteVector::fromBase64(const ByteVector & input)
 {
-  static const unsigned char base64[256]={
+  static const unsigned char base64[256] = {
     0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
     0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,
     0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x80,0x3e,0x80,0x80,0x80,0x3f,
@@ -980,61 +976,88 @@ ByteVector ByteVector::fromBase64(const ByteVector & input)
   ByteVector output(len);
 
   const unsigned char * src = (const unsigned char*) input.data();
-  unsigned char * dst = (unsigned char*)output.data();
-  while(4<=len) {
-    if(base64[src[0]]==0x80) break;
-    if(base64[src[1]]==0x80) break;
-    *dst++=((base64[src[0]]<<2)&0xfc)|((base64[src[1]]>>4)&0x03);
-    if(src[2]!='=') {
-      if(base64[src[2]]==0x80) break;
-      *dst++=((base64[src[1]]&0x0f)<<4)|((base64[src[2]]>>2)&0x0f);
-      if(src[3]!='=') {
-        if(base64[src[3]]==0x80) break;
-        *dst++=((base64[src[2]]&0x03)<<6)|(base64[src[3]]&0x3f);
+  unsigned char *       dst = (unsigned char*) output.data();
+
+  while(4 <= len) {
+
+    // Check invalid character
+    if(base64[src[0]] == 0x80) 
+      break;
+
+    // Check invalid character
+    if(base64[src[1]] == 0x80) 
+      break;
+
+    // Decode first byte
+    *dst++ = ((base64[src[0]] << 2) & 0xfc) | ((base64[src[1]] >> 4) & 0x03);
+
+    if(src[2] != '=') {
+
+      // Check invalid character
+      if(base64[src[2]] == 0x80) 
+        break;
+
+      // Decode second byte
+      *dst++ = ((base64[src[1]] & 0x0f) << 4) | ((base64[src[2]] >> 2) & 0x0f);
+
+      if(src[3] != '=') {
+
+        // Check invalid character
+        if(base64[src[3]] == 0x80) 
+          break;
+
+        // Decode third byte
+        *dst++ = ((base64[src[2]] & 0x03) << 6) | (base64[src[3]] & 0x3f);
       }
       else {
+        // assume end of data
+        len -= 4;
         break;
       }
     }
     else {
+      // assume end of data
+      len -= 4;
       break;
     }
-    src+=4;
-    len-=4;
+    src += 4;
+    len -= 4;
     }
-  output.resize(dst-(unsigned char*)output.data());
-  return output;
+
+  // Only return output if we processed all bytes
+  if(len == 0) {
+    output.resize(dst - (unsigned char*) output.data());
+    return output;
+  }
+  return ByteVector();
 }
 
-
-
-
 ByteVector ByteVector::toBase64() const
 {
-  static const char alphabet[]="ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
+  static const char alphabet[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
   if (size()) {
-    uint len=size();
-    ByteVector output(4 * ((len-1)/3+1)); // note roundup
+    uint len = size();
+    ByteVector output(4 * ((len - 1) / 3 + 1)); // note roundup
 
     const char * src = data();
     char * dst = output.data();
-    while(3<=len) {
-      *dst++=alphabet[(src[0]>>2)&0x3f];
-      *dst++=alphabet[((src[0]&0x03)<<4)|((src[1]>>4)&0x0f)];
-      *dst++=alphabet[((src[1]&0x0f)<<2)|((src[2]>>6)&0x03)];
-      *dst++=alphabet[src[2]&0x3f];
-      src+=3;
-      len-=3;
+    while(3 <= len) {
+      *dst++ = alphabet[(src[0] >> 2) & 0x3f];
+      *dst++ = alphabet[((src[0] & 0x03) << 4) | ((src[1] >> 4) & 0x0f)];
+      *dst++ = alphabet[((src[1] & 0x0f) << 2) | ((src[2] >> 6) & 0x03)];
+      *dst++ = alphabet[src[2] & 0x3f];
+      src += 3;
+      len -= 3;
     }
     if(len) {
-      *dst++=alphabet[(src[0]>>2)&0x3f];
+      *dst++ = alphabet[(src[0] >> 2) & 0x3f];
       if(len>1) {
-        *dst++=alphabet[((src[0]&0x03)<<4)|((src[1]>>4)&0x0f)];
-        *dst++=alphabet[((src[1]&0x0f)<<2)];
+        *dst++ = alphabet[((src[0] & 0x03) << 4)|((src[1] >> 4) & 0x0f)];
+        *dst++ = alphabet[((src[1] & 0x0f) << 2)];
       }
       else {
-        *dst++=alphabet[(src[0]&0x03)<<4];
-        *dst++='=';
+        *dst++ = alphabet[(src[0] & 0x03) << 4];
+        *dst++ = '=';
       }
     *dst++='=';
     }

From 90ad17b4c5da178d91c9a20e8c3724ea9e57de53 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 21:26:30 -0500
Subject: [PATCH 09/15] Additional unit tests for base64 encoder/decoder

---
 tests/test_bytevector.cpp | 68 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 68 insertions(+)

diff --git a/tests/test_bytevector.cpp b/tests/test_bytevector.cpp
index 18097adf..450402e3 100644
--- a/tests/test_bytevector.cpp
+++ b/tests/test_bytevector.cpp
@@ -25,6 +25,7 @@
 #include <cstring>
 #include <tbytevector.h>
 #include <tbytevectorlist.h>
+#include <tstring.h>
 #include <cppunit/extensions/HelperMacros.h>
 
 using namespace std;
@@ -416,6 +417,73 @@ public:
     CPPUNIT_ASSERT_EQUAL(t2, ByteVector::fromBase64(s2.toBase64()));
     CPPUNIT_ASSERT_EQUAL(t3, ByteVector::fromBase64(s3.toBase64()));
 
+    ByteVector all((uint)256);
+
+    // in order
+    {
+      for(int i = 0; i < 256; i++){
+        all[i]=(unsigned char)i;
+        }
+      ByteVector b64 = all.toBase64();
+      ByteVector original = ByteVector::fromBase64(b64);
+      CPPUNIT_ASSERT_EQUAL(all,original);
+    }
+
+    // reverse
+    {
+      for(int i = 0; i < 256; i++){
+        all[i]=(unsigned char)255-i;
+        }
+      ByteVector b64 = all.toBase64();
+      ByteVector original = ByteVector::fromBase64(b64);
+      CPPUNIT_ASSERT_EQUAL(all,original);
+    }
+
+    // all zeroes
+    {
+      for(int i = 0; i < 256; i++){
+        all[i]=0;
+        }
+      ByteVector b64 = all.toBase64();
+      ByteVector original = ByteVector::fromBase64(b64);
+      CPPUNIT_ASSERT_EQUAL(all,original);
+    }
+
+    // all ones
+    {
+      for(int i = 0; i < 256; i++){
+        all[i]=0xff;
+        }
+      ByteVector b64 = all.toBase64();
+      ByteVector original = ByteVector::fromBase64(b64);
+      CPPUNIT_ASSERT_EQUAL(all,original);
+    }
+
+    // Missing end bytes
+    {
+      // No missing bytes
+      ByteVector m0("YW55IGNhcm5hbCBwbGVhc3VyZQ==");
+      CPPUNIT_ASSERT_EQUAL(s2,ByteVector::fromBase64(m0));
+
+      // 1 missing byte
+      ByteVector m1("YW55IGNhcm5hbCBwbGVhc3VyZQ=");
+      CPPUNIT_ASSERT_EQUAL(sempty,ByteVector::fromBase64(m1));
+
+      // 2 missing bytes
+      ByteVector m2("YW55IGNhcm5hbCBwbGVhc3VyZQ");
+      CPPUNIT_ASSERT_EQUAL(sempty,ByteVector::fromBase64(m2));
+
+      // 3 missing bytes
+      ByteVector m3("YW55IGNhcm5hbCBwbGVhc3VyZ");
+      CPPUNIT_ASSERT_EQUAL(sempty,ByteVector::fromBase64(m3));
+    }
+
+    // Grok invalid characters
+    {
+      ByteVector invalid("abd\x00\x01\x02\x03\x04");
+      CPPUNIT_ASSERT_EQUAL(sempty,ByteVector::fromBase64(invalid));
+    }
+
   }
 
 

From 47bd01ecda5852070de3821cd954658e8560d389 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 21:32:50 -0500
Subject: [PATCH 10/15] Minor style fixes

---
 taglib/ogg/xiphcomment.cpp | 4 ++--
 tests/test_bytevector.cpp  | 1 -
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/taglib/ogg/xiphcomment.cpp b/taglib/ogg/xiphcomment.cpp
index 6b87e3ea..cb607ce4 100644
--- a/taglib/ogg/xiphcomment.cpp
+++ b/taglib/ogg/xiphcomment.cpp
@@ -361,7 +361,7 @@ ByteVector Ogg::XiphComment::render(bool addFramingBit) const
 
   for(PictureList::ConstIterator it = d->pictureList.begin(); it != d->pictureList.end(); ++it) {
     ByteVector picture = (*it)->render().toBase64();
-    data.append(ByteVector::fromUInt(picture.size()+23,false));
+    data.append(ByteVector::fromUInt(picture.size() + 23, false));
     data.append("METADATA_BLOCK_PICTURE=");
     data.append(picture);
   }
@@ -445,7 +445,7 @@ void Ogg::XiphComment::parse(const ByteVector &data)
 
       // Parse key and value
       String key = String(entry.mid(0, sep), String::UTF8);
-      String value = String(entry.mid(sep+1), String::UTF8);
+      String value = String(entry.mid(sep + 1), String::UTF8);
       addField(key, value, false);
     }
   }
diff --git a/tests/test_bytevector.cpp b/tests/test_bytevector.cpp
index 450402e3..8d23b82b 100644
--- a/tests/test_bytevector.cpp
+++ b/tests/test_bytevector.cpp
@@ -25,7 +25,6 @@
 #include <cstring>
 #include <tbytevector.h>
 #include <tbytevectorlist.h>
-#include <tstring.h>
 #include <cppunit/extensions/HelperMacros.h>
 
 using namespace std;

From f836cb4dea012826c00c371783db2b810438e750 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 21:38:01 -0500
Subject: [PATCH 11/15] Two more spaces

---
 taglib/toolkit/tbytevector.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/taglib/toolkit/tbytevector.cpp b/taglib/toolkit/tbytevector.cpp
index 14550ad1..531503f7 100644
--- a/taglib/toolkit/tbytevector.cpp
+++ b/taglib/toolkit/tbytevector.cpp
@@ -1052,7 +1052,7 @@ ByteVector ByteVector::toBase64() const
     if(len) {
       *dst++ = alphabet[(src[0] >> 2) & 0x3f];
       if(len>1) {
-        *dst++ = alphabet[((src[0] & 0x03) << 4)|((src[1] >> 4) & 0x0f)];
+        *dst++ = alphabet[((src[0] & 0x03) << 4) | ((src[1] >> 4) & 0x0f)];
         *dst++ = alphabet[((src[1] & 0x0f) << 2)];
       }
       else {

From 70e471f1d19a3be0caaf6000590eb48c71b439f5 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 22:09:27 -0500
Subject: [PATCH 12/15] Check if saved picture can be read back correctly

---
 tests/test_ogg.cpp | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/tests/test_ogg.cpp b/tests/test_ogg.cpp
index 05b23f64..3afeb196 100644
--- a/tests/test_ogg.cpp
+++ b/tests/test_ogg.cpp
@@ -158,6 +158,14 @@ public:
     f = new Vorbis::File(newname.c_str());
     List<FLAC::Picture *> lst = f->tag()->pictureList();
     CPPUNIT_ASSERT_EQUAL(TagLib::uint(1), lst.size());
+    CPPUNIT_ASSERT_EQUAL(int(5), lst[0]->width());
+    CPPUNIT_ASSERT_EQUAL(int(6), lst[0]->height());
+    CPPUNIT_ASSERT_EQUAL(int(16), lst[0]->colorDepth());
+    CPPUNIT_ASSERT_EQUAL(int(7), lst[0]->numColors());
+    CPPUNIT_ASSERT_EQUAL(String("image/jpeg"), lst[0]->mimeType());
+    CPPUNIT_ASSERT_EQUAL(String("new image"), lst[0]->description());
+    CPPUNIT_ASSERT_EQUAL(ByteVector("JPEG data"), lst[0]->data());
+  
     delete f;
   }
 

From 41a44f4ab2b1dda149e54c9b7431dd8624d22c51 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 22:34:08 -0500
Subject: [PATCH 13/15] Support reading deprecated COVERART xiphcomment

---
 taglib/ogg/xiphcomment.cpp | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/taglib/ogg/xiphcomment.cpp b/taglib/ogg/xiphcomment.cpp
index cb607ce4..0424ed55 100644
--- a/taglib/ogg/xiphcomment.cpp
+++ b/taglib/ogg/xiphcomment.cpp
@@ -434,6 +434,23 @@ void Ogg::XiphComment::parse(const ByteVector &data)
       else
         debug("Unable to parse METADATA_BLOCK_PICTURE. Discarding content.");
     }
+    else if (entry.startsWith("COVERART=")) {
+
+      // Decode base64 picture data
+      ByteVector picturedata = ByteVector::fromBase64(entry.mid(9));
+
+      if (picturedata.size() == 0) {
+        debug("Empty coverart data. Discaring content");
+        continue;
+      }
+
+      // Assume it's some type of image file
+      FLAC::Picture * picture = new FLAC::Picture();
+      picture->setData(picturedata);
+      picture->setMimeType("image/");
+      picture->setType(FLAC::Picture::Other);
+      d->pictureList.append(picture);
+    }
     else {
 
       // Check for field separator

From 3e8eff16b8fc8de6a38fd6c159cb21990d1fc051 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Sat, 16 May 2015 22:37:54 -0500
Subject: [PATCH 14/15] another style fix

---
 taglib/toolkit/tbytevector.cpp | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/taglib/toolkit/tbytevector.cpp b/taglib/toolkit/tbytevector.cpp
index 531503f7..757e41c3 100644
--- a/taglib/toolkit/tbytevector.cpp
+++ b/taglib/toolkit/tbytevector.cpp
@@ -981,11 +981,11 @@ ByteVector ByteVector::fromBase64(const ByteVector & input)
   while(4 <= len) {
 
     // Check invalid character
-    if(base64[src[0]] == 0x80) 
+    if(base64[src[0]] == 0x80)
       break;
 
     // Check invalid character
-    if(base64[src[1]] == 0x80) 
+    if(base64[src[1]] == 0x80)
       break;
 
     // Decode first byte
@@ -994,7 +994,7 @@ ByteVector ByteVector::fromBase64(const ByteVector & input)
     if(src[2] != '=') {
 
       // Check invalid character
-      if(base64[src[2]] == 0x80) 
+      if(base64[src[2]] == 0x80)
         break;
 
       // Decode second byte
@@ -1003,7 +1003,7 @@ ByteVector ByteVector::fromBase64(const ByteVector & input)
       if(src[3] != '=') {
 
         // Check invalid character
-        if(base64[src[3]] == 0x80) 
+        if(base64[src[3]] == 0x80)
           break;
 
         // Decode third byte
@@ -1059,7 +1059,7 @@ ByteVector ByteVector::toBase64() const
         *dst++ = alphabet[(src[0] & 0x03) << 4];
         *dst++ = '=';
       }
-    *dst++='=';
+    *dst++ = '=';
     }
     return output;
   }

From 2184aa476f19533d2829a40c7a7d22a76c8f8545 Mon Sep 17 00:00:00 2001
From: Sander Jansen <s.jansen@gmail.com>
Date: Mon, 18 May 2015 08:24:02 -0500
Subject: [PATCH 15/15] xiph: preserve any picture data we can't decode in the
 text comments

---
 taglib/ogg/xiphcomment.cpp | 101 +++++++++++++++++++++++--------------
 1 file changed, 62 insertions(+), 39 deletions(-)

diff --git a/taglib/ogg/xiphcomment.cpp b/taglib/ogg/xiphcomment.cpp
index 0424ed55..8ec108dd 100644
--- a/taglib/ogg/xiphcomment.cpp
+++ b/taglib/ogg/xiphcomment.cpp
@@ -419,51 +419,74 @@ void Ogg::XiphComment::parse(const ByteVector &data)
     // Handle Pictures separately
     if(entry.startsWith("METADATA_BLOCK_PICTURE=")) {
 
-      // Decode base64 picture data
-      ByteVector picturedata = ByteVector::fromBase64(entry.mid(23));
+      // We need base64 encoded data including padding
+      if((entry.size() - 23) > 3 && ((entry.size() - 23) % 4) == 0) {
 
-      if(picturedata.size() == 0) {
-        debug("Empty picture data. Discarding content");
-        continue;
+        // Decode base64 picture data
+        ByteVector picturedata = ByteVector::fromBase64(entry.mid(23));
+        if (picturedata.size()) {
+
+          // Decode Flac Picture
+          FLAC::Picture * picture = new FLAC::Picture();
+          if(picture->parse(picturedata)) {
+
+            d->pictureList.append(picture);
+
+            // continue to next field
+            continue;
+          }
+          else {
+            delete picture;
+            debug("Failed to decode FlacPicture block");
+          }
+        }
+        else {
+          debug("Failed to decode base64 encoded data");
+        }
       }
-
-      FLAC::Picture * picture = new FLAC::Picture();
-
-      if(picture->parse(picturedata))
-        d->pictureList.append(picture);
-      else
-        debug("Unable to parse METADATA_BLOCK_PICTURE. Discarding content.");
-    }
-    else if (entry.startsWith("COVERART=")) {
-
-      // Decode base64 picture data
-      ByteVector picturedata = ByteVector::fromBase64(entry.mid(9));
-
-      if (picturedata.size() == 0) {
-        debug("Empty coverart data. Discaring content");
-        continue;
+      else {
+        debug("Invalid base64 encoded data");
       }
-
-      // Assume it's some type of image file
-      FLAC::Picture * picture = new FLAC::Picture();
-      picture->setData(picturedata);
-      picture->setMimeType("image/");
-      picture->setType(FLAC::Picture::Other);
-      d->pictureList.append(picture);
     }
-    else {
 
-      // Check for field separator
-      int sep = entry.find('=');
-      if (sep < 1) {
-        debug("Discarding invalid comment field.");
-        continue;
+    // Handle old picture standard
+    if (entry.startsWith("COVERART=")) {
+
+      if((entry.size() - 9) > 3 && ((entry.size() - 9) % 4) == 0) {
+
+        // Decode base64 picture data
+        ByteVector picturedata = ByteVector::fromBase64(entry.mid(9));
+        if (picturedata.size()) {
+
+          // Assume it's some type of image file
+          FLAC::Picture * picture = new FLAC::Picture();
+          picture->setData(picturedata);
+          picture->setMimeType("image/");
+          picture->setType(FLAC::Picture::Other);
+          d->pictureList.append(picture);
+
+          // continue to next field
+          continue;
+        }
+        else {
+          debug("Failed to decode base64 encoded data");
+        }
+      }
+      else {
+        debug("Invalid base64 encoded data");
       }
-
-      // Parse key and value
-      String key = String(entry.mid(0, sep), String::UTF8);
-      String value = String(entry.mid(sep + 1), String::UTF8);
-      addField(key, value, false);
     }
+
+    // Check for field separator
+    int sep = entry.find('=');
+    if (sep < 1) {
+      debug("Discarding invalid comment field.");
+      continue;
+    }
+
+    // Parse key and value
+    String key = String(entry.mid(0, sep), String::UTF8);
+    String value = String(entry.mid(sep + 1), String::UTF8);
+    addField(key, value, false);
   }
 }