[DEV] add ogg decoder basis

ewolsa/LoadedFile.cpp

@@ -11,17 +11,27 @@
 #include <ewolsa/debug.h>
 #include <ewolsa/LoadedFile.h>
 #include <ewolsa/decWav.h>
+#include <ewolsa/decOgg.h>
 
 
-ewolsa::LoadedFile::LoadedFile(const std::string& _file, int8_t _nbChanRequested) :
-  m_file(_file),
+ewolsa::LoadedFile::LoadedFile(const std::string& _fileName, int8_t _nbChanRequested) :
+  m_file(_fileName),
   m_nbSamples(0),
   m_requestedTime(1),
   m_data(NULL){
-	m_data = ewolsa::loadAudioFile(_file, _nbChanRequested, m_nbSamples);
+	std::string tmpName = std::tolower(_fileName);
+	// select the corect Loader :
+	if (end_with(tmpName, ".wav") == true) {
+		m_data = ewolsa::wav::loadAudioFile(_fileName, _nbChanRequested, m_nbSamples);
+	} else if (end_with(tmpName, ".ogg") == true) {
+		m_data = ewolsa::ogg::loadAudioFile(_fileName, _nbChanRequested, m_nbSamples);
+	} else {
+		EWOLSA_ERROR("Extention not managed '" << _fileName << "' Sopported extention : .wav / .ogg");
+		return;
+	}
 	if (m_data == NULL) {
 		// write an error ...
-		EWOLSA_ERROR("Can not open file : " << _file);
+		EWOLSA_ERROR("Can not open file : " << _fileName);
 	}
 }
 

ewolsa/LoadedFile.h

@@ -15,7 +15,7 @@
 namespace ewolsa {
 	class LoadedFile {
 		public :
-			LoadedFile(const std::string& _file, int8_t _nbChanRequested=1);
+			LoadedFile(const std::string& _fileName, int8_t _nbChanRequested=1);
 			~LoadedFile(void);
 			std::string m_file;
 			int32_t m_nbSamples;

ewolsa/decOgg.cpp

@@ -0,0 +1,144 @@
+/**
+ * @author Edouard DUPIN
+ * 
+ * @copyright 2011, Edouard DUPIN, all right reserved
+ * 
+ * @license BSD 3 clauses (see license file)
+ */
+
+#include <etk/types.h>
+#include <etk/os/FSNode.h>
+#include <ewolsa/debug.h>
+#include <ewolsa/decOgg.h>
+#include <tremor/ivorbiscodec.h>
+#include <tremor/ivorbisfile.h>
+
+
+static size_t LocalReadFunc(void *ptr, size_t size, size_t nmemb, void *datasource) {
+	etk::FSNode* file = static_cast<etk::FSNode*>(datasource);
+	return file->fileRead(ptr, size, nmemb);
+}
+
+static int localSeekFunc(void *datasource, ogg_int64_t offset, int whence) {
+	etk::FSNode* file = static_cast<etk::FSNode*>(datasource);
+	enum etk::seekNode mode = etk::FSN_SEEK_START;
+	if (whence == SEEK_SET) {
+		mode = etk::FSN_SEEK_START;
+	} else if (whence == SEEK_END) {
+		mode = etk::FSN_SEEK_END;
+	} else if (whence == SEEK_CUR) {
+		mode = etk::FSN_SEEK_CURRENT;
+	}
+	if (file->fileSeek(offset, mode) == true) {
+		return 0;
+	} else {
+		return -1;
+	}
+}
+
+static int localCloseFunc(void *datasource) {
+	etk::FSNode* file = static_cast<etk::FSNode*>(datasource);
+	file->fileClose();
+}
+
+static long localTellFunc(void *datasource) {
+	etk::FSNode* file = static_cast<etk::FSNode*>(datasource);
+	return file->fileTell();
+}
+
+int16_t* ewolsa::ogg::loadAudioFile(const std::string& _filename, int8_t _nbChan, int32_t& _nbSampleOut) {
+	_nbSampleOut = 0;
+	OggVorbis_File vf;
+	int32_t eof=0;
+	int32_t current_section;
+	ov_callbacks tmpCallback = {
+		LocalReadFunc,
+		localSeekFunc,
+		localCloseFunc,
+		localTellFunc
+	};
+	etk::FSNode fileAccess(_filename);
+	// Start loading the XML : 
+	EWOLSA_DEBUG("open file (OGG) \"" << fileAccess << "\"");
+	if (false == fileAccess.exist()) {
+		EWOLSA_ERROR("File Does not exist : \"" << fileAccess << "\"");
+		return NULL;
+	}
+	int32_t fileSize = fileAccess.fileSize();
+	if (0 == fileSize) {
+		EWOLSA_ERROR("This file is empty : \"" << fileAccess << "\"");
+		return NULL;
+	}
+	if (false == fileAccess.fileOpenRead()) {
+		EWOLSA_ERROR("Can not open the file : \"" << fileAccess << "\"");
+		return NULL;
+	}
+	if (ov_open_callbacks(&fileAccess, &vf, NULL, 0, tmpCallback) < 0) {
+		EWOLSA_ERROR("Input does not appear to be an Ogg bitstream.");
+		return NULL;
+	}
+	vorbis_info *vi=ov_info(&vf,-1);
+	_nbSampleOut = ov_pcm_total(&vf,-1) / vi->channels;
+	{
+		char **ptr=ov_comment(&vf,-1)->user_comments;
+		while(*ptr){
+			EWOLSA_DEBUG("comment : " << *ptr);
+			++ptr;
+		}
+		EWOLSA_DEBUG("Bitstream is " << vi->channels << " channel, " << vi->rate << "Hz");
+		EWOLSA_DEBUG("Decoded length: " << _nbSampleOut << " samples");
+		EWOLSA_DEBUG("Encoded by: " << ov_comment(&vf,-1)->vendor);
+		EWOLSA_DEBUG("time: " << ((float)_nbSampleOut/(float)vi->rate)/60.0);
+	}
+	int16_t* outputData = new int16_t[_nbSampleOut*_nbChan*sizeof(int16_t)];
+	if (NULL == outputData) {
+		EWOLSA_ERROR("Allocation ERROR try to allocate " << (int32_t)(_nbSampleOut*_nbChan*sizeof(int16_t) ) << "bytes");
+		return NULL;
+	}
+	int32_t pos = 0;
+	char pcmout[4096];
+	while(!eof){
+		long ret=ov_read(&vf, pcmout, sizeof(pcmout), &current_section);
+		if (ret == 0) {
+			/* EOF */
+			eof=1;
+		} else if (ret < 0) {
+			if(ret==OV_EBADLINK){
+				EWOLSA_ERROR("Corrupt bitstream section! Exiting.");
+				// TODO : Remove pointer data ...
+				return NULL;
+			}
+		} else {
+			int16_t* pointerIn = (int16_t*)pcmout;
+			int16_t* pointerOut = outputData+pos;
+			if (_nbChan == vi->channels) {
+				memcpy(pointerOut, pointerIn, ret);
+				pos += ret/2;
+			} else {
+				if (    _nbChan == 1
+				     && vi->channels == 2) {
+					for (int32_t iii=0; iii<ret/4 ; ++iii) {
+						pointerOut[iii] = pointerIn[iii*2];
+					}
+					pos += ret/4;
+				} else if (    _nbChan == 2
+				            && vi->channels == 1) {
+					for (int32_t iii=0; iii<ret/2 ; ++iii) {
+						pointerOut[iii*2]   = pointerIn[iii];
+						pointerOut[iii*2+1] = pointerIn[iii];
+					}
+					pos += ret;
+				} else {
+					EWOLSA_ERROR("Can not convert " << vi->channels << " channels in " << _nbChan << " channels");
+				}
+			}
+			// fwrite(pcmout,1,ret,stdout);
+			//EWOLSA_DEBUG("get data : " << ret << " Bytes");
+		}
+	}
+	/* cleanup */
+	ov_clear(&vf);
+	EWOLSA_CRITICAL("Done.");
+	return outputData;
+}
+

ewolsa/decOgg.h

@@ -0,0 +1,20 @@
+/**
+ * @author Edouard DUPIN
+ * 
+ * @copyright 2011, Edouard DUPIN, all right reserved
+ * 
+ * @license BSD 3 clauses (see license file)
+ */
+
+#ifndef __EWOLSA_DEC_OGG_H__
+#define __EWOLSA_DEC_OGG_H__
+
+#include <etk/types.h>
+
+namespace ewolsa {
+	namespace ogg {
+		int16_t* loadAudioFile(const std::string& _filename, int8_t _nbChan, int32_t& _nbSampleOut);
+	}
+};
+
+#endif

ewolsa/decWav.cpp

@@ -59,7 +59,7 @@ typedef struct {
 #define COMPR_G721   (64)
 #define COMPR_MPEG   (80)
 
-int16_t* ewolsa::loadAudioFile(const std::string& _filename, int8_t _nbChan, int32_t& _nbSampleOut) {
+int16_t* ewolsa::wav::loadAudioFile(const std::string& _filename, int8_t _nbChan, int32_t& _nbSampleOut) {
 	_nbSampleOut = 0;
 	waveHeader myHeader;
 	memset(&myHeader, 0, sizeof(waveHeader));

ewolsa/decWav.h

@@ -12,7 +12,9 @@
 #include <etk/types.h>
 
 namespace ewolsa {
-	int16_t* loadAudioFile(const std::string& _filename, int8_t _nbChan, int32_t& _nbSampleOut);
+	namespace wav {
+		int16_t* loadAudioFile(const std::string& _filename, int8_t _nbChan, int32_t& _nbSampleOut);
+	}
 };
 
 #endif

lutin_ewolsa.py

@@ -13,6 +13,7 @@ def create(target):
 	myModule.add_src_file([
 		'ewolsa/debug.cpp',
 		'ewolsa/decWav.cpp',
+		'ewolsa/decOgg.cpp',
 		'ewolsa/effects.cpp',
 		'ewolsa/ewolsa.cpp',
 		'ewolsa/music.cpp',
@@ -20,7 +21,7 @@ def create(target):
 		])
 	
 	# name of the dependency
-	myModule.add_module_depend(['ewol', 'airtaudio'])
+	myModule.add_module_depend(['ewol', 'airtaudio', 'ogg'])
 	
 	myModule.add_export_path(tools.get_current_path(__file__))