[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr...
|
From: |
Ben Levitt |
|
Subject: |
[Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr... |
|
Date: |
Sun, 08 Jul 2007 23:16:57 +0000 |
CVSROOT: /sources/traverso
Module name: traverso
Changes by: Ben Levitt <benjie> 07/07/08 23:16:57
Modified files:
src/core : AbstractAudioReader.cpp AbstractAudioReader.h
Project.cpp ResampleAudioReader.cpp
ResampleAudioReader.h SFAudioReader.cpp
SFAudioReader.h VorbisAudioReader.cpp
VorbisAudioReader.h core.pro gdither.cpp
Added files:
src/core : MadAudioReader.cpp MadAudioReader.h
Log message:
Add MP3 decoding using libmad
CVSWeb URLs:
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/AbstractAudioReader.cpp?cvsroot=traverso&r1=1.1&r2=1.2
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/AbstractAudioReader.h?cvsroot=traverso&r1=1.2&r2=1.3
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/Project.cpp?cvsroot=traverso&r1=1.51&r2=1.52
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/ResampleAudioReader.cpp?cvsroot=traverso&r1=1.2&r2=1.3
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/ResampleAudioReader.h?cvsroot=traverso&r1=1.2&r2=1.3
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/SFAudioReader.cpp?cvsroot=traverso&r1=1.2&r2=1.3
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/SFAudioReader.h?cvsroot=traverso&r1=1.2&r2=1.3
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/VorbisAudioReader.cpp?cvsroot=traverso&r1=1.2&r2=1.3
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/VorbisAudioReader.h?cvsroot=traverso&r1=1.2&r2=1.3
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/core.pro?cvsroot=traverso&r1=1.30&r2=1.31
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/gdither.cpp?cvsroot=traverso&r1=1.2&r2=1.3
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/MadAudioReader.cpp?cvsroot=traverso&rev=1.1
http://cvs.savannah.gnu.org/viewcvs/traverso/src/core/MadAudioReader.h?cvsroot=traverso&rev=1.1
Patches:
Index: AbstractAudioReader.cpp
===================================================================
RCS file: /sources/traverso/traverso/src/core/AbstractAudioReader.cpp,v
retrieving revision 1.1
retrieving revision 1.2
diff -u -b -r1.1 -r1.2
--- AbstractAudioReader.cpp 6 Jul 2007 22:13:32 -0000 1.1
+++ AbstractAudioReader.cpp 8 Jul 2007 23:16:56 -0000 1.2
@@ -21,6 +21,7 @@
#include "AbstractAudioReader.h"
#include "SFAudioReader.h"
+#include "MadAudioReader.h"
#include "VorbisAudioReader.h"
#include "ResampleAudioReader.h"
@@ -68,6 +69,9 @@
if (VorbisAudioReader::can_decode(filename)) {
newReader = new VorbisAudioReader(filename);
}
+ else if (MadAudioReader::can_decode(filename)) {
+ newReader = new MadAudioReader(filename);
+ }
else if (SFAudioReader::can_decode(filename)) {
newReader = new SFAudioReader(filename);
}
Index: AbstractAudioReader.h
===================================================================
RCS file: /sources/traverso/traverso/src/core/AbstractAudioReader.h,v
retrieving revision 1.2
retrieving revision 1.3
diff -u -b -r1.2 -r1.3
--- AbstractAudioReader.h 7 Jul 2007 22:00:12 -0000 1.2
+++ AbstractAudioReader.h 8 Jul 2007 23:16:56 -0000 1.3
@@ -36,7 +36,7 @@
~AbstractAudioReader();
virtual int get_num_channels() = 0;
- virtual int get_length() = 0;
+ virtual nframes_t get_length() = 0;
virtual int get_rate() = 0;
virtual bool is_compressed() = 0;
int read_from(audio_sample_t* dst, nframes_t start, nframes_t cnt);
Index: Project.cpp
===================================================================
RCS file: /sources/traverso/traverso/src/core/Project.cpp,v
retrieving revision 1.51
retrieving revision 1.52
diff -u -b -r1.51 -r1.52
--- Project.cpp 26 Jun 2007 18:07:56 -0000 1.51
+++ Project.cpp 8 Jul 2007 23:16:56 -0000 1.52
@@ -703,6 +703,7 @@
int Project::get_rate( ) const
{
return m_rate;
+ //return audiodevice().get_sample_rate();
}
int Project::get_bitdepth( ) const
Index: ResampleAudioReader.cpp
===================================================================
RCS file: /sources/traverso/traverso/src/core/ResampleAudioReader.cpp,v
retrieving revision 1.2
retrieving revision 1.3
diff -u -b -r1.2 -r1.3
--- ResampleAudioReader.cpp 7 Jul 2007 22:00:13 -0000 1.2
+++ ResampleAudioReader.cpp 8 Jul 2007 23:16:56 -0000 1.3
@@ -50,8 +50,7 @@
m_realReader = 0;
}
- m_srcData.end_of_input = 0;
-
+ reset();
seek(0);
}
@@ -69,6 +68,14 @@
}
+// Clear the samplerateconverter to a clean state (used on seek)
+void ResampleAudioReader::reset()
+{
+ src_reset(m_srcState);
+ m_srcData.end_of_input = 0;
+}
+
+
// Get from child AudioReader
int ResampleAudioReader::get_num_channels()
{
@@ -80,7 +87,7 @@
// Get from child AudioReader, convert from file's frames to song's frames
-int ResampleAudioReader::get_length()
+nframes_t ResampleAudioReader::get_length()
{
if (m_realReader) {
if (audiodevice().get_sample_rate() ==
m_realReader->get_rate()) {
@@ -119,13 +126,16 @@
return m_realReader->seek(start);
}
+ reset();
+
m_nextFrame = start;
return m_realReader->seek(song_to_file_frame(start));
}
-//
+// If no conversion is necessary, pass the read straight to the child
AudioReader,
+// otherwise get data from childreader and use libsamplerate to convert
int ResampleAudioReader::read(audio_sample_t* dst, int sampleCount)
{
Q_ASSERT(m_realReader);
Index: ResampleAudioReader.h
===================================================================
RCS file: /sources/traverso/traverso/src/core/ResampleAudioReader.h,v
retrieving revision 1.2
retrieving revision 1.3
diff -u -b -r1.2 -r1.3
--- ResampleAudioReader.h 7 Jul 2007 22:00:13 -0000 1.2
+++ ResampleAudioReader.h 8 Jul 2007 23:16:56 -0000 1.3
@@ -33,7 +33,7 @@
~ResampleAudioReader();
int get_num_channels();
- int get_length();
+ nframes_t get_length();
int get_rate();
bool is_compressed();
bool seek(nframes_t start);
@@ -43,6 +43,8 @@
static bool can_decode(QString filename) {return false;}
protected:
+ void reset();
+
nframes_t song_to_file_frame(nframes_t frame);
nframes_t file_to_song_frame(nframes_t frame);
Index: SFAudioReader.cpp
===================================================================
RCS file: /sources/traverso/traverso/src/core/SFAudioReader.cpp,v
retrieving revision 1.2
retrieving revision 1.3
diff -u -b -r1.2 -r1.3
--- SFAudioReader.cpp 7 Jul 2007 22:00:13 -0000 1.2
+++ SFAudioReader.cpp 8 Jul 2007 23:16:56 -0000 1.3
@@ -86,7 +86,7 @@
}
-int SFAudioReader::get_length()
+nframes_t SFAudioReader::get_length()
{
if (m_sf) {
return m_sfinfo.frames;
Index: SFAudioReader.h
===================================================================
RCS file: /sources/traverso/traverso/src/core/SFAudioReader.h,v
retrieving revision 1.2
retrieving revision 1.3
diff -u -b -r1.2 -r1.3
--- SFAudioReader.h 7 Jul 2007 22:00:13 -0000 1.2
+++ SFAudioReader.h 8 Jul 2007 23:16:56 -0000 1.3
@@ -33,7 +33,7 @@
~SFAudioReader();
int get_num_channels();
- int get_length();
+ nframes_t get_length();
int get_rate();
bool is_compressed();
bool seek(nframes_t start);
Index: VorbisAudioReader.cpp
===================================================================
RCS file: /sources/traverso/traverso/src/core/VorbisAudioReader.cpp,v
retrieving revision 1.2
retrieving revision 1.3
diff -u -b -r1.2 -r1.3
--- VorbisAudioReader.cpp 7 Jul 2007 22:00:13 -0000 1.2
+++ VorbisAudioReader.cpp 8 Jul 2007 23:16:56 -0000 1.3
@@ -92,7 +92,7 @@
}
-int VorbisAudioReader::get_length()
+nframes_t VorbisAudioReader::get_length()
{
if (m_file) {
return ov_pcm_total(&m_vf, -1);
Index: VorbisAudioReader.h
===================================================================
RCS file: /sources/traverso/traverso/src/core/VorbisAudioReader.h,v
retrieving revision 1.2
retrieving revision 1.3
diff -u -b -r1.2 -r1.3
--- VorbisAudioReader.h 7 Jul 2007 22:00:13 -0000 1.2
+++ VorbisAudioReader.h 8 Jul 2007 23:16:57 -0000 1.3
@@ -35,7 +35,7 @@
~VorbisAudioReader();
int get_num_channels();
- int get_length();
+ nframes_t get_length();
int get_rate();
bool is_compressed();
bool seek(nframes_t start);
Index: core.pro
===================================================================
RCS file: /sources/traverso/traverso/src/core/core.pro,v
retrieving revision 1.30
retrieving revision 1.31
diff -u -b -r1.30 -r1.31
--- core.pro 6 Jul 2007 22:13:33 -0000 1.30
+++ core.pro 8 Jul 2007 23:16:57 -0000 1.31
@@ -54,7 +54,8 @@
AbstractAudioReader.cpp \
SFAudioReader.cpp \
ResampleAudioReader.cpp \
- VorbisAudioReader.cpp
+ VorbisAudioReader.cpp \
+ MadAudioReader.cpp
HEADERS = precompile.h \
AudioClip.h \
AudioClipList.h \
@@ -102,7 +103,8 @@
AbstractAudioReader.h \
SFAudioReader.h \
ResampleAudioReader.h \
- VorbisAudioReader.h
+ VorbisAudioReader.h \
+ MadAudioReader.h
macx{
QMAKE_LIBDIR += /usr/local/qt/lib
}
Index: gdither.cpp
===================================================================
RCS file: /sources/traverso/traverso/src/core/gdither.cpp,v
retrieving revision 1.2
retrieving revision 1.3
diff -u -b -r1.2 -r1.3
--- gdither.cpp 25 Aug 2006 11:25:08 -0000 1.2
+++ gdither.cpp 8 Jul 2007 23:16:57 -0000 1.3
@@ -15,7 +15,7 @@
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
*
- * $Id: gdither.cpp,v 1.2 2006/08/25 11:25:08 r_sijrier Exp $
+ * $Id: gdither.cpp,v 1.3 2007/07/08 23:16:57 benjie Exp $
*/
#include <gdither_types_internal.h>
@@ -240,7 +240,7 @@
}
}
-/* floating pint version of the inner loop function */
+/* floating point version of the inner loop function */
inline static void gdither_innner_loop_fp(const GDitherType dt,
const uint32_t stride, const float bias, const float scale,
Index: MadAudioReader.cpp
===================================================================
RCS file: MadAudioReader.cpp
diff -N MadAudioReader.cpp
--- /dev/null 1 Jan 1970 00:00:00 -0000
+++ MadAudioReader.cpp 8 Jul 2007 23:16:56 -0000 1.1
@@ -0,0 +1,859 @@
+/*
+Copyright (C) 2007 Ben Levitt
+
+This file is part of Traverso
+(Most of this Mp3 Reading code borrowed from K3b...)
+
+Traverso is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program; if not, write to the Free Software
+Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+
+*/
+
+#include "MadAudioReader.h"
+#include <QFile>
+#include <QString>
+#include <QVector>
+#include "Utils.h"
+
+// Always put me below _all_ includes, this is needed
+// in case we run with memory leak detection enabled!
+#include "Debugger.h"
+
+
+
+
+static const int INPUT_BUFFER_SIZE = 5*8192;
+
+class K3bMad
+{
+public:
+ K3bMad();
+ ~K3bMad();
+
+ bool open( const QString& filename );
+
+ /**
+ * @return true if the mad stream contains data
+ * false if there is no data left or an error occurred.
+ * In the latter case inputError() returns true.
+ */
+ bool fillStreamBuffer();
+
+ /**
+ * Skip id3 tags.
+ *
+ * This will reset the input file.
+ */
+ bool skipTag();
+
+ /**
+ * Find first frame and seek to the beginning of that frame.
+ * This is used to skip the junk that many mp3 files start with.
+ */
+ bool seekFirstHeader();
+
+ bool eof() const;
+ bool inputError() const;
+
+ /**
+ * Current position in theinput file. This does NOT
+ * care about the status of the mad stream. Use streamPos()
+ * in that case.
+ */
+ qint64 inputPos() const;
+
+ /**
+ * Current absolut position of the decoder stream.
+ */
+ qint64 streamPos() const;
+ bool inputSeek(qint64 pos);
+
+ void initMad();
+ void cleanup();
+
+ bool decodeNextFrame();
+ bool findNextHeader();
+ bool checkFrameHeader(mad_header* header) const;
+
+ mad_stream* madStream;
+ mad_frame* madFrame;
+ mad_synth* madSynth;
+ mad_timer_t* madTimer;
+
+private:
+ QFile m_inputFile;
+ bool m_madStructuresInitialized;
+ unsigned char* m_inputBuffer;
+ bool m_bInputError;
+
+ int m_channels;
+ int m_sampleRate;
+};
+
+
+K3bMad::K3bMad()
+ : m_madStructuresInitialized(false),
+ m_bInputError(false)
+{
+ madStream = new mad_stream;
+ madFrame = new mad_frame;
+ madSynth = new mad_synth;
+ madTimer = new mad_timer_t;
+
+ //
+ // we allocate additional MAD_BUFFER_GUARD bytes to always be able to
append the
+ // zero bytes needed for decoding the last frame.
+ //
+ m_inputBuffer = new unsigned char[INPUT_BUFFER_SIZE+MAD_BUFFER_GUARD];
+}
+
+
+K3bMad::~K3bMad()
+{
+ cleanup();
+
+ delete madStream;
+ delete madFrame;
+ delete madSynth;
+ delete madTimer;
+
+ delete [] m_inputBuffer;
+}
+
+
+bool K3bMad::open(const QString& filename)
+{
+ cleanup();
+
+ m_bInputError = false;
+ m_channels = m_sampleRate = 0;
+
+ m_inputFile.setFileName(filename);
+
+ if (!m_inputFile.open(QIODevice::ReadOnly)) {
+ PERROR("could not open file %s", QS_C(m_inputFile.fileName()));
+ return false;
+ }
+
+ initMad();
+
+ memset(m_inputBuffer, 0, INPUT_BUFFER_SIZE+MAD_BUFFER_GUARD);
+
+ return true;
+}
+
+
+bool K3bMad::inputError() const
+{
+ return m_bInputError;
+}
+
+
+bool K3bMad::fillStreamBuffer()
+{
+ /* The input bucket must be filled if it becomes empty or if
+ * it's the first execution of the loop.
+ */
+ if (madStream->buffer == 0 || madStream->error == MAD_ERROR_BUFLEN) {
+ if (eof()) {
+ return false;
+ }
+
+ long readSize, remaining;
+ unsigned char* readStart;
+
+ if (madStream->next_frame != 0) {
+ remaining = madStream->bufend - madStream->next_frame;
+ memmove(m_inputBuffer, madStream->next_frame,
remaining);
+ readStart = m_inputBuffer + remaining;
+ readSize = INPUT_BUFFER_SIZE - remaining;
+ }
+ else {
+ readSize = INPUT_BUFFER_SIZE;
+ readStart = m_inputBuffer;
+ remaining = 0;
+ }
+
+ // Fill-in the buffer.
+ long result = m_inputFile.read((char*)readStart, readSize);
+ if (result < 0) {
+ //kdDebug() << "(K3bMad) read error on bitstream)" <<
endl;
+ m_bInputError = true;
+ return false;
+ }
+ else if (result == 0) {
+ //kdDebug() << "(K3bMad) end of input stream" << endl;
+ return false;
+ }
+ else {
+ readStart += result;
+
+ if (eof()) {
+ //kdDebug() << "(K3bMad::fillStreamBuffer)
MAD_BUFFER_GUARD" << endl;
+ memset(readStart, 0, MAD_BUFFER_GUARD);
+ result += MAD_BUFFER_GUARD;
+ }
+
+ // Pipe the new buffer content to libmad's stream
decoder facility.
+ mad_stream_buffer(madStream, m_inputBuffer, result +
remaining);
+ madStream->error = MAD_ERROR_NONE;
+ }
+ }
+
+ return true;
+}
+
+
+bool K3bMad::skipTag()
+{
+ // skip the tag at the beginning of the file
+ m_inputFile.seek(0);
+
+ //
+ // now check if the file starts with an id3 tag and skip it if so
+ //
+ char buf[4096];
+ int bufLen = 4096;
+ if (m_inputFile.read(buf, bufLen) < bufLen) {
+ //kdDebug() << "(K3bMad) unable to read " << bufLen << " bytes
from "
+ // << m_inputFile.name() << endl;
+ return false;
+ }
+
+ if ((buf[0] == 'I' && buf[1] == 'D' && buf[2] == '3') &&
+ ((unsigned short)buf[3] < 0xff && (unsigned short)buf[4] < 0xff)) {
+ // do we have a footer?
+ bool footer = (buf[5] & 0x10);
+
+ // the size is saved as a synched int meaning bit 7 is always
cleared to 0
+ unsigned int size =
+ ( (buf[6] & 0x7f) << 21 ) |
+ ( (buf[7] & 0x7f) << 14 ) |
+ ( (buf[8] & 0x7f) << 7) |
+ (buf[9] & 0x7f);
+ unsigned int offset = size + 10;
+
+ if (footer) {
+ offset += 10;
+ }
+
+ // skip the id3 tag
+ if (!m_inputFile.seek(offset)) {
+ PERROR("Couldn't seek to %u in %s", offset,
m_inputFile.fileName());
+ return false;
+ }
+ }
+ else {
+ // reset file
+ return m_inputFile.seek(0);
+ }
+
+ return true;
+}
+
+
+bool K3bMad::seekFirstHeader()
+{
+ //
+ // A lot of mp3 files start with a lot of junk which confuses mad.
+ // We "allow" an mp3 file to start with at most 1 KB of junk. This is
just
+ // some random value since we do not want to search the hole file. That
would
+ // take way to long for non-mp3 files.
+ //
+ bool headerFound = findNextHeader();
+ qint64 inputPos = streamPos();
+ while (!headerFound &&
+ !m_inputFile.atEnd() &&
+ streamPos() <= inputPos+1024) {
+ headerFound = findNextHeader();
+ }
+
+ // seek back to the begin of the frame
+ if (headerFound) {
+ int streamSize = madStream->bufend - madStream->buffer;
+ int bytesToFrame = madStream->this_frame - madStream->buffer;
+ m_inputFile.seek(m_inputFile.pos() - streamSize + bytesToFrame);
+
+ //kdDebug() << "(K3bMad) found first header at " <<
m_inputFile.pos() << endl;
+ }
+
+ // reset the stream to make sure mad really starts decoding at out seek
position
+ mad_stream_finish(madStream);
+ mad_stream_init(madStream);
+
+ return headerFound;
+}
+
+
+bool K3bMad::eof() const
+{
+ return m_inputFile.atEnd();
+}
+
+
+qint64 K3bMad::inputPos() const
+{
+ return m_inputFile.pos();
+}
+
+
+qint64 K3bMad::streamPos() const
+{
+ return inputPos() - (madStream->bufend - madStream->this_frame + 1);
+}
+
+
+bool K3bMad::inputSeek(qint64 pos)
+{
+ return m_inputFile.seek(pos);
+}
+
+
+void K3bMad::initMad()
+{
+ if (!m_madStructuresInitialized) {
+ mad_stream_init(madStream);
+ mad_timer_reset(madTimer);
+ mad_frame_init(madFrame);
+ mad_synth_init(madSynth);
+
+ m_madStructuresInitialized = true;
+ }
+}
+
+
+void K3bMad::cleanup()
+{
+ if (m_inputFile.isOpen()) {
+ //kdDebug() << "(K3bMad) cleanup at offset: "
+ // << "Input file at: " << m_inputFile.pos() << " "
+ // << "Input file size: " << m_inputFile.size() << "
"
+ // << "stream pos: "
+ // << ( m_inputFile.pos() - (madStream->bufend -
madStream->this_frame + 1) )
+ // << endl;
+ m_inputFile.close();
+ }
+
+ if (m_madStructuresInitialized) {
+ mad_frame_finish(madFrame);
+ mad_synth_finish(madSynth);
+ mad_stream_finish(madStream);
+ }
+
+ m_madStructuresInitialized = false;
+}
+
+
+//
+// LOSTSYNC could happen when mad encounters the id3 tag...
+//
+bool K3bMad::findNextHeader()
+{
+ if (!fillStreamBuffer()) {
+ return false;
+ }
+
+ //
+ // MAD_RECOVERABLE == true: frame was read, decoding failed (about to
skip frame)
+ // MAD_RECOVERABLE == false: frame was not read, need data
+ //
+
+ if (mad_header_decode( &madFrame->header, madStream ) < 0) {
+ if (MAD_RECOVERABLE(madStream->error) ||
+ madStream->error == MAD_ERROR_BUFLEN) {
+ return findNextHeader();
+ }
+ else
+ // kdDebug() << "(K3bMad::findNextHeader) error: " <<
mad_stream_errorstr( madStream ) << endl;
+
+ // FIXME probably we should not do this here since we don't do
it
+ // in the frame decoding
+ // if(!checkFrameHeader(&madFrame->header))
+ // return findNextHeader();
+
+ return false;
+ }
+
+ if (!m_channels) {
+ m_channels = MAD_NCHANNELS(&madFrame->header);
+ m_sampleRate = madFrame->header.samplerate;
+ }
+
+ mad_timer_add(madTimer, madFrame->header.duration);
+
+ return true;
+}
+
+
+bool K3bMad::decodeNextFrame()
+{
+ if (!fillStreamBuffer()) {
+ return false;
+ }
+
+ if (mad_frame_decode(madFrame, madStream) < 0) {
+ if (MAD_RECOVERABLE(madStream->error) ||
+ madStream->error == MAD_ERROR_BUFLEN) {
+ return decodeNextFrame();
+ }
+
+ return false;
+ }
+
+ if (!m_channels) {
+ m_channels = MAD_NCHANNELS(&madFrame->header);
+ m_sampleRate = madFrame->header.samplerate;
+ }
+
+ mad_timer_add(madTimer, madFrame->header.duration);
+
+ return true;
+}
+
+
+//
+// This is from the arts mad decoder
+//
+bool K3bMad::checkFrameHeader(mad_header* header) const
+{
+ int frameSize = MAD_NSBSAMPLES(header) * 32;
+
+ if (frameSize <= 0) {
+ return false;
+ }
+
+ if (m_channels && m_channels != MAD_NCHANNELS(header)) {
+ return false;
+ }
+
+ return true;
+}
+
+
+
+class MadAudioReader::MadDecoderPrivate
+{
+public:
+ MadDecoderPrivate()
+ : outputBuffer(0),
+ outputPointer(0),
+ outputBufferEnd(0)
+ {
+ mad_header_init( &firstHeader );
+ }
+
+ K3bMad* handle;
+
+ QVector<unsigned long long> seekPositions;
+
+ bool bOutputFinished;
+
+ audio_sample_t* outputBuffer;
+ audio_sample_t* outputPointer;
+ audio_sample_t* outputBufferEnd;
+
+ audio_sample_t* overflowBuffer;
+ int overflowSize;
+ int overflowStart;
+
+ // the first frame header for technical info
+ mad_header firstHeader;
+ bool vbr;
+};
+
+
+MadAudioReader::MadAudioReader(QString filename)
+ : AbstractAudioReader(filename)
+{
+ d = new MadDecoderPrivate();
+ d->handle = new K3bMad();
+
+ initDecoderInternal();
+
+ m_channels = 0;
+ m_frames = countFrames();
+
+ if (m_frames <= 0) {
+ d->handle->cleanup();
+ delete d->handle;
+ delete d;
+ d = 0;
+ return;
+ }
+
+ switch( d->firstHeader.mode ) {
+ case MAD_MODE_SINGLE_CHANNEL:
+ m_channels = 1;
+ case MAD_MODE_DUAL_CHANNEL:
+ case MAD_MODE_JOINT_STEREO:
+ case MAD_MODE_STEREO:
+ m_channels = 2;
+ }
+
+ d->overflowBuffer = new audio_sample_t[1152 * get_num_channels()];
+ d->overflowSize = 0;
+ d->overflowStart = 0;
+
+ seek(0);
+}
+
+
+MadAudioReader::~MadAudioReader()
+{
+ if (d) {
+ d->handle->cleanup();
+ delete d->handle;
+ delete d->overflowBuffer;
+ delete d;
+ }
+}
+
+
+bool MadAudioReader::can_decode(QString filename)
+{
+ //
+ // HACK:
+ //
+ // I am simply no good at this and this detection code is no good as
well
+ // It always takes waves for mp3 files so we introduce this hack to
+ // filter out wave files. :(
+ //
+ QFile f( filename );
+ if( !f.open( QIODevice::ReadOnly ) )
+ return false;
+ char buffer[12];
+ if( f.read( buffer, 12 ) != 12 )
+ return false;
+ if( !qstrncmp( buffer, "RIFF", 4 ) && !qstrncmp( buffer + 8, "WAVE", 4
) )
+ return false;
+ f.close();
+
+
+ K3bMad handle;
+ if( !handle.open( filename ) )
+ return false;
+
+ handle.skipTag();
+ if( !handle.seekFirstHeader() )
+ return false;
+
+ if( handle.findNextHeader() ) {
+ int c = MAD_NCHANNELS( &handle.madFrame->header );
+ int layer = handle.madFrame->header.layer;
+ unsigned int s = handle.madFrame->header.samplerate;
+
+ //
+ // find 4 more mp3 headers (random value since 2 was not enough)
+ // This way we get most of the mp3 files while sorting out
+ // for example wave files.
+ //
+ int cnt = 1;
+ while( handle.findNextHeader() ) {
+ // compare the found headers
+ if( MAD_NCHANNELS( &handle.madFrame->header ) == c &&
+ handle.madFrame->header.layer == layer &&
+ handle.madFrame->header.samplerate == s ) {
+ // only support layer III for now since
otherwise some wave files
+ // are taken for layer I
+ if( ++cnt >= 5 ) {
+ //stdout << "(MadDecoder) valid mpeg 1
layer " << layer
+ //<< " file with " << c << " channels
and a samplerate of "
+ //<< s << endl;
+ return ( layer == MAD_LAYER_III );
+ }
+ }
+ else
+ break;
+ }
+ }
+
+ PERROR("unsupported format: %s",QS_C(filename));
+
+ return false;
+}
+
+
+int MadAudioReader::get_num_channels()
+{
+ if (d) {
+ return m_channels;
+ }
+ return 0;
+}
+
+
+nframes_t MadAudioReader::get_length()
+{
+ if (d) {
+ return m_frames;
+ }
+ return 0;
+}
+
+
+int MadAudioReader::get_rate()
+{
+ if (d) {
+ return d->firstHeader.samplerate;
+ }
+ return 0;
+}
+
+
+// Should this exist? Should we just be smarter in MonoReader so we don't
need this?
+bool MadAudioReader::is_compressed()
+{
+ return false;
+}
+
+
+bool MadAudioReader::seek(nframes_t start)
+{
+ Q_ASSERT(d);
+
+ if (start >= get_length()) {
+ return false;
+ }
+
+ //
+ // we need to reset the complete mad stuff
+ //
+ if (!initDecoderInternal()) {
+ return false;
+ }
+
+ //
+ // search a position
+ // This is all hacking, I don't really know what I am doing here... ;)
+ //
+ double mp3FrameSecs =
static_cast<double>(d->firstHeader.duration.seconds) +
static_cast<double>(d->firstHeader.duration.fraction) /
static_cast<double>(MAD_TIMER_RESOLUTION);
+
+ double posSecs = static_cast<double>(start) / get_rate();
+
+ // seekPosition to seek after frame i
+ unsigned int frame = static_cast<unsigned int>( posSecs / mp3FrameSecs
);
+
+ // Rob said: 29 frames is the theoretically max frame reservoir limit
(whatever that means...)
+ // it seems that mad needs at most 29 frames to get ready
+ unsigned int frameReservoirProtect = ( frame > 29 ? 29 : frame );
+
+ frame -= frameReservoirProtect;
+
+ // seek in the input file behind the already decoded data
+ d->handle->inputSeek( d->seekPositions[frame] );
+
+ // decode some frames ignoring MAD_ERROR_BADDATAPTR errors
+ unsigned int i = 1;
+ while( i <= frameReservoirProtect ) {
+ d->handle->fillStreamBuffer();
+ if( mad_frame_decode( d->handle->madFrame, d->handle->madStream
) ) {
+ if( MAD_RECOVERABLE( d->handle->madStream->error ) ) {
+ if( d->handle->madStream->error ==
MAD_ERROR_BUFLEN )
+ continue;
+ else if( d->handle->madStream->error !=
MAD_ERROR_BADDATAPTR ) {
+ //kdDebug() << "(K3bMadDecoder)
Seeking: recoverable mad error ("
+ //<<
mad_stream_errorstr(d->handle->madStream) << ")" << endl;
+ continue;
+ }
+ else {
+ //kdDebug() << "(K3bMadDecoder)
Seeking: ignoring ("
+ //<<
mad_stream_errorstr(d->handle->madStream) << ")" << endl;
+ }
+ }
+ else
+ return false;
+ }
+
+ if( i == frameReservoirProtect ) // synth only the last frame
(Rob said so ;)
+ mad_synth_frame( d->handle->madSynth,
d->handle->madFrame );
+
+ ++i;
+ }
+
+ m_nextFrame = start;
+
+ return true;
+}
+
+
+bool MadAudioReader::initDecoderInternal()
+{
+ d->handle->cleanup();
+
+ d->bOutputFinished = false;
+
+ if( !d->handle->open( m_fileName ) )
+ return false;
+
+ if( !d->handle->skipTag() )
+ return false;
+
+ if( !d->handle->seekFirstHeader() )
+ return false;
+
+ return true;
+}
+
+
+unsigned long MadAudioReader::countFrames()
+ {
+ //kdDebug() << "(K3bMadDecoder::countFrames)" << endl;
+
+ unsigned long frames = 0;
+ bool error = false;
+ d->vbr = false;
+ bool bFirstHeaderSaved = false;
+
+ d->seekPositions.clear();
+
+ while (!error && d->handle->findNextHeader()) {
+ if (!bFirstHeaderSaved) {
+ bFirstHeaderSaved = true;
+ d->firstHeader = d->handle->madFrame->header;
+ }
+ else if (d->handle->madFrame->header.bitrate !=
d->firstHeader.bitrate) {
+ d->vbr = true;
+ }
+ //
+ // position in stream: postion in file minus the not yet used
buffer
+ //
+ unsigned long long seekPos = d->handle->inputPos() -
+ (d->handle->madStream->bufend -
d->handle->madStream->this_frame + 1);
+
+ // save the number of bytes to be read to decode i-1 frames at
position i
+ // in other words: when seeking to seekPos the next decoded
frame will be i
+ d->seekPositions.append( seekPos );
+ }
+
+ if( !d->handle->inputError() && !error ) {
+ frames = get_rate() * (d->handle->madTimer->seconds +
+
(float)d->handle->madTimer->fraction/(float)MAD_TIMER_RESOLUTION);
+ //kdDebug() << "(K3bMadDecoder) length of track " << seconds <<
endl;
+ }
+
+ d->handle->cleanup();
+
+ //kdDebug() << "(K3bMadDecoder::countFrames) end" << endl;
+
+ return frames;
+}
+
+
+int MadAudioReader::read(audio_sample_t* dst, int sampleCount)
+{
+ d->outputBuffer = dst;
+ d->outputBufferEnd = d->outputBuffer + sampleCount;
+ d->outputPointer = d->outputBuffer;
+
+ bool bOutputBufferFull = false;
+
+ // Deal with existing overflow
+ if (d->overflowSize > 0) {
+ if (d->overflowSize < sampleCount) {
+ //printf("output all %d overflow samples\n",
d->overflowSize);
+ memcpy(d->outputBuffer, d->overflowBuffer +
d->overflowStart, d->overflowSize * sizeof(audio_sample_t));
+ d->outputPointer += d->overflowSize;
+ d->overflowSize = 0;
+ d->overflowStart = 0;
+ }
+ else {
+ //printf("output %d overflow samples, returned from
overflow\n", sampleCount);
+ memcpy(d->outputBuffer, d->overflowBuffer +
d->overflowStart, sampleCount * sizeof(audio_sample_t));
+ d->overflowSize -= sampleCount;
+ d->overflowStart += sampleCount;
+ m_nextFrame += sampleCount / get_num_channels();
+ return sampleCount;
+ }
+ }
+
+ while (!bOutputBufferFull && d->handle->fillStreamBuffer()) {
+ // a mad_synth contains of the data of one mad_frame
+ // one mad_frame represents a mp3-frame which is always 1152
samples
+ // for us that means we need 1152 samples per channel of output
buffer
+ // for every frame
+ if (d->outputBufferEnd - d->outputPointer == 0) {
+ bOutputBufferFull = true;
+ }
+ else if (d->handle->decodeNextFrame()) {
+ //
+ // Once decoded the frame is synthesized to PCM
samples. No errors
+ // are reported by mad_synth_frame();
+ //
+ mad_synth_frame( d->handle->madSynth,
d->handle->madFrame );
+
+ // this fills the output buffer
+ if (!createPcmSamples(d->handle->madSynth)) {
+ return -1;
+ }
+ }
+ else if (d->handle->inputError()) {
+ return -1;
+ }
+ }
+
+ nframes_t samplesWritten = d->outputPointer - d->outputBuffer;
+
+ // Pad end with zeros if necessary
+ // FIXME: This shouldn't be necessary! :P
+ // is get_length() reporting incorrectly?
+ // are we not outputting the last mp3-frame for some reason?
+ int remainingSamplesRequested = sampleCount - samplesWritten;
+ int remainingSamplesInFile = get_length() * get_num_channels() -
(m_nextFrame * get_num_channels() + samplesWritten);
+ if (remainingSamplesRequested > 0 && remainingSamplesInFile > 0) {
+ int padLength = (remainingSamplesRequested >
remainingSamplesInFile) ? remainingSamplesInFile : remainingSamplesRequested;
+ memset(d->outputPointer, 0, padLength);
+ samplesWritten += padLength;
+ }
+
+ //printf("at: %lu (total: %lu), request: %d (returned: %d)\n",
m_nextFrame, m_frames, sampleCount/get_num_channels(),
(samplesWritten)/get_num_channels());
+
+ m_nextFrame += samplesWritten / get_num_channels();
+ return samplesWritten;
+}
+
+
+bool MadAudioReader::createPcmSamples( mad_synth* synth )
+{
+ unsigned short nsamples = synth->pcm.length;
+ bool overflow = false;
+
+ // now create the output
+ for (int i = 0; i < nsamples; i++) {
+ if (d->outputPointer >= d->outputBufferEnd) {
+ //printf("start overflowing\n");
+ d->outputPointer = d->overflowBuffer;
+ overflow = true;
+ }
+
+ /* Left channel */
+ *(d->outputPointer++) =
mad_f_todouble(synth->pcm.samples[0][i]);
+
+ /* Right channel. If the decoded stream is monophonic then
+ * the right output channel is the same as the left one.
+ */
+ if (synth->pcm.channels == 2) {
+ *(d->outputPointer++) =
mad_f_todouble(synth->pcm.samples[1][i]);
+ }
+ } // pcm conversion
+
+ if (overflow) {
+ d->overflowSize = d->outputPointer - d->overflowBuffer;
+ d->overflowStart = 0;
+ d->outputPointer = d->outputBufferEnd;
+ }
+
+ return true;
+}
+
+
Index: MadAudioReader.h
===================================================================
RCS file: MadAudioReader.h
diff -N MadAudioReader.h
--- /dev/null 1 Jan 1970 00:00:00 -0000
+++ MadAudioReader.h 8 Jul 2007 23:16:56 -0000 1.1
@@ -0,0 +1,60 @@
+/*
+Copyright (C) 2007 Ben Levitt
+
+This file is part of Traverso
+
+Traverso is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with this program; if not, write to the Free Software
+Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
+
+*/
+
+#ifndef MADAUDIOREADER_H
+#define MADAUDIOREADER_H
+
+#include <AbstractAudioReader.h>
+
+extern "C" {
+#include <mad.h>
+}
+
+
+class MadAudioReader : public AbstractAudioReader
+{
+public:
+ MadAudioReader(QString filename);
+ ~MadAudioReader();
+
+ int get_num_channels();
+ nframes_t get_length();
+ int get_rate();
+ bool is_compressed();
+ bool seek(nframes_t start);
+ int read(audio_sample_t* dst, int sampleCount);
+
+ static bool can_decode(QString filename);
+
+protected:
+ bool initDecoderInternal();
+ unsigned long countFrames();
+ bool createPcmSamples(mad_synth* synth);
+
+ static int MaxAllowedRecoverableErrors;
+ nframes_t m_frames;
+ int m_channels;
+
+ class MadDecoderPrivate;
+ MadDecoderPrivate* d;
+};
+
+#endif
- [Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr...,
Ben Levitt <=
- [Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr..., Ben Levitt, 2007/07/09
- [Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr..., Ben Levitt, 2007/07/13
- [Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr..., Ben Levitt, 2007/07/15
- [Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr..., Ben Levitt, 2007/07/18
- [Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr..., Ben Levitt, 2007/07/19
- [Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr..., Ben Levitt, 2007/07/20
- [Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr..., Ben Levitt, 2007/07/24
- [Traverso-commit] traverso/src/core AbstractAudioReader.cpp Abstr..., Ben Levitt, 2007/07/25