part of refactor the process

@@ -402,18 +402,44 @@ int merge_audio_file(vector<string> & files, const char * dest)
 	return 0;
 }
 
-int process_files(const char * output_dest_file)
-{
-	vector<fileinfo>  & filesaudio = media_files[type_audio];
-	vector<fileinfo>  & filesvideo = media_files[type_video];
+class MergeProcess{
+public:
+	MergeProcess(vector<fileinfo>  & a, vector<fileinfo>  & v);
+	int process_files(const char * output_dest_file);
+
+protected:
+		void init();
+		void adjust_va_timestamp();
+		void merge_left_audio();
+		int process_video_ahead_of_audio();
+		int process_video_behind_audio();
+		int process_video_align_audio();
+		int process_video_loop();
+		int process_va();
+
+protected:
+	vector<fileinfo>  & filesaudio;
+	vector<fileinfo>  & filesvideo;
 	vector<string > merged_files;
 	vector<string>  tmp_files;
-	int nv = 0;
-	int nf = 0;
-	char destfile[1024],audio_file[1024],pic_file[1024];
+	int nv; // the index of processing video file
+	int nf;//the index of processing target merged ts
+	char destfile[1024], audio_file[1024], pic_file[1024];
 	char blank_pic_file[1024];
 	char silence_aac_file[1024];
+	int audio_index;
+	int audio_start;
+	fileinfo audio;
+	fileinfo video;
+};
+
+MergeProcess::MergeProcess(vector<fileinfo> & a, vector<fileinfo> & v) :filesaudio(a), filesvideo(v)
+{
+	init();
+}
 
+void MergeProcess::init()
+{
 	strcpy(blank_pic_file, cfg_path);
 	strcat(blank_pic_file, "blank.jpg");
 
@@ -422,318 +448,365 @@ int process_files(const char * output_dest_file)
 
 	check_audio_duration();
 	get_duration_from_video_file();
-	//don't split video, for a video, using merged audios to mix with it
-	//for audio, mix with video or jpg
-
-	if (filesvideo.size()) {//has video files
-		if (filesaudio.size()){
-			for (int i = 0; i < filesaudio.size(); i++){  //
-				fileinfo audio = filesaudio[i];
-				float audio_start = audio.start_time;
+	nv = 0;
+	nf = 0;
+	audio_index = 0;
+}
 
-				for (int j = i; j < filesaudio.size(); j++){
-					filesaudio[j].start_time -= audio_start;
-					filesaudio[j].end_time -= audio_start;
-				}
+void MergeProcess::adjust_va_timestamp()
+{
+	fileinfo audio = filesaudio[audio_index];
+	float audio_start = audio.start_time;
 
-				for (int j = nv; j < filesvideo.size(); j++) {
-					filesvideo[j].start_time -= audio_start;
-					filesvideo[j].end_time -= audio_start;
-				}
+	for (int j = audio_index; j < filesaudio.size(); j++){
+		filesaudio[j].start_time -= audio_start;
+		filesaudio[j].end_time -= audio_start;
+	}
 
-				audio = filesaudio[i];
+	for (int j = nv; j < filesvideo.size(); j++) {
+		filesvideo[j].start_time -= audio_start;
+		filesvideo[j].end_time -= audio_start;
+	}
+}
 
-				audio_start = 0;//for a new processing audio,the start is 0
 
-				for (; nv < filesvideo.size(); nv++) {
-					fileinfo video = filesvideo[nv];
+void MergeProcess::merge_left_audio()
+{
+	if (audio_start < audio.end_time){
+		sprintf(destfile, "%d_%s", nf, audio.name.c_str());
+		split_audio(audio.name.c_str(), audio_start, audio.end_time - audio_start, destfile);
+		tmp_files.push_back(destfile);
+
+		sprintf(destfile, "%d.ts", nf);
+		if (nv < filesvideo.size()) {
+			fileinfo video = filesvideo[nv];
+			sprintf(pic_file, "%s.jpg", video.name.c_str());
+			get_video_first_frame_jpeg(video, pic_file);
+			tmp_files.push_back(pic_file);
+		}
+		else {
+			strcpy(pic_file, blank_pic_file);
+		}
+		merge_audio_pic(audio, nf, pic_file, destfile);
+		merged_files.push_back(destfile);
+		nf++;
+	}
+}
 
-					if (video.start_time < audio_start - 0.1)
-					{//video is much more ahead of audio,try padding silence first
-						if (video.end_time < audio_start + 0.1) {
-							sprintf(destfile, "%d_silence.aac", nf);
-							split_audio(silence_aac_file, 0, video.end_time - video.start_time, destfile);
-							tmp_files.push_back(destfile);
-							sprintf(destfile, "%d.ts", nf);
-							merge_audio_video(destfile, video.name.c_str(), destfile);
-							merged_files.push_back(destfile);
-							nf++;
-							continue; //for next video
-						}
-						else {
-							// combine a audio file for the video
-							double silence_audio_start = audio.end_time;//maybe need append silence 
-							double silence_audio_end = video.end_time;
-
-							bool need_append_silence = true;
-							bool to_next_video = false;
-
-							vector<std::string > merge_audio_files;
-							sprintf(destfile, "%d_0_silence.aac", nf);//a duration of silence
-							split_audio(silence_aac_file, 0, audio_start - video.start_time, destfile);
-							merge_audio_files.push_back(destfile);
-							tmp_files.push_back(destfile);
-
-							if (audio.end_time < video.end_time + 0.1 && audio.end_time > video.end_time - 0.1) {
-								merge_audio_files.push_back(audio.name); //whole audio file,just fit
-								audio_start = audio.end_time + 0.1;
-								need_append_silence = false;
-							}
-							else if (audio.end_time > video.end_time){ //split part of audio file
-								sprintf(destfile, "%d_%s", nf, audio.name.c_str());
-								merge_audio_files.push_back(destfile);
-								tmp_files.push_back(destfile);
-								split_audio(audio.name.c_str(), audio_start, video.end_time - audio_start, destfile);
-								audio_start = video.end_time;
-								need_append_silence = false;
-							}
-							else {
-								merge_audio_files.push_back(audio.name);
-
-								for (; i + 1 < filesaudio.size(); i++){//since video is not finished,try find next audio
-									audio = filesaudio[i + 1];
-									if (audio.start_time < video.end_time) {//next audio should split to fit the video
-										silence_audio_end = audio.start_time;
-
-										sprintf(destfile, "%d_%d_silence.aac", nf, i);
-										split_audio(silence_aac_file, 0, silence_audio_end - silence_audio_start, destfile);
-										merge_audio_files.push_back(destfile);
-										tmp_files.push_back(destfile);
-
-										if (audio.end_time > video.end_time - 0.1 && audio.end_time < video.end_time + 0.1) {//just match
-											merge_audio_files.push_back(audio.name);
-											need_append_silence = false;
-											audio_start = audio.end_time + 0.1;
-											i++;//this audio is used
-											break;
-										}
-										if (audio.end_time > video.end_time){
-											sprintf(destfile, "%d_%s", nf, audio.name.c_str());
-											merge_audio_files.push_back(destfile);
-											tmp_files.push_back(destfile);
-											split_audio(audio.name.c_str(), 0, video.end_time - audio.start_time, destfile);
-											need_append_silence = false;
-											//adjust timecode for the audio is part left
-											float cur_audio_start = video.end_time - audio.start_time;
-
-											audio_start = audio.start_time;
-
-											for (int j = i + 1; j < filesaudio.size(); j++){
-												filesaudio[j].start_time -= audio_start;
-												filesaudio[j].end_time -= audio_start;
-											}
-
-											for (int j = nv; j < filesvideo.size(); j++) {
-												filesvideo[j].start_time -= audio_start;
-												filesvideo[j].end_time -= audio_start;
-											}
-											i++;
-											audio = filesaudio[i];
-											audio_start = cur_audio_start;
-
-											to_next_video = true;
-
-											break;
-										}
-										merge_audio_files.push_back(audio.name);//whole audio should be appended
-										silence_audio_start = audio.end_time; //adjust the silence start
-									}
-									else {
-										break;//no need for next audio
-									}
-								}//end audio find for the video
-							}//end else
-
-							if (need_append_silence) {
-								sprintf(destfile, "%d_silence.aac", nf);
-								split_audio(silence_aac_file, 0, silence_audio_end - silence_audio_start, destfile);
-								merge_audio_files.push_back(destfile);
-								tmp_files.push_back(destfile);
-							}
-
-							sprintf(audio_file, "%d_merged.aac", nf);
-							merge_audio_file(merge_audio_files, audio_file);
-
-
-							sprintf(destfile, "%d.ts", nf);
-							merge_audio_video(audio_file, video.name.c_str(), destfile);
-							merged_files.push_back(destfile);
-							nf++;
-
-							if (!to_next_video){
-								nv++;//this video is processed
-								break;
-							}
-						}//end need combine
-					}//end video is ahead of audio
-					//-----VS-----
-					//AS----------
-					if (video.start_time - audio_start > 0.1) {//video is behind audio too much
-						sprintf(audio_file, "%d_%s", nf, audio.name.c_str());
-						if (video.start_time < audio.end_time - 0.1){
-							split_audio(audio.name.c_str(), audio_start, video.start_time - audio_start, audio_file);
-							audio_start = video.start_time;
-						}
-						else {
-							split_audio(audio.name.c_str(), audio_start, audio.end_time - audio_start, audio_file);
-						}
-						tmp_files.push_back(audio_file);
+int  MergeProcess::process_video_ahead_of_audio()
+{
+	if (video.end_time < audio_start + 0.1) {
+		sprintf(destfile, "%d_silence.aac", nf);
+		split_audio(silence_aac_file, 0, video.end_time - video.start_time, destfile);
+		tmp_files.push_back(destfile);
+		sprintf(destfile, "%d.ts", nf);
+		merge_audio_video(destfile, video.name.c_str(), destfile);
+		merged_files.push_back(destfile);
+		nf++;
+		return 0; //for next video
+	}
+	else {
+		// combine a audio file for the video
+		double silence_audio_start = audio.end_time;//maybe need append silence 
+		double silence_audio_end = video.end_time;
+
+		bool need_append_silence = true;
+		bool to_next_video = false;
+
+		vector<std::string > merge_audio_files;
+		sprintf(destfile, "%d_0_silence.aac", nf);//a duration of silence
+		split_audio(silence_aac_file, 0, audio_start - video.start_time, destfile);
+		merge_audio_files.push_back(destfile);
+		tmp_files.push_back(destfile);
+
+		if (audio.end_time < video.end_time + 0.1 && audio.end_time > video.end_time - 0.1) {
+			merge_audio_files.push_back(audio.name); //whole audio file,just fit
+			audio_start = audio.end_time + 0.1;
+			need_append_silence = false;
+		}
+		else if (audio.end_time > video.end_time){ //split part of audio file
+			sprintf(destfile, "%d_%s", nf, audio.name.c_str());
+			merge_audio_files.push_back(destfile);
+			tmp_files.push_back(destfile);
+			split_audio(audio.name.c_str(), audio_start, video.end_time - audio_start, destfile);
+			audio_start = video.end_time;
+			need_append_silence = false;
+		}
+		else {
+			merge_audio_files.push_back(audio.name);
 
-						sprintf(pic_file, "%s.jpg", video.name.c_str());
-						get_video_first_frame_jpeg(video, pic_file);
-						tmp_files.push_back(pic_file);
+			for (; audio_index + 1 < filesaudio.size(); audio_index++){//since video is not finished,try find next audio
+				audio = filesaudio[audio_index + 1];
+				if (audio.start_time < video.end_time) {//next audio should split to fit the video
+					silence_audio_end = audio.start_time;
 
-						sprintf(destfile, "%d.ts", nf);
-						merge_audio_pic(audio_file, pic_file, destfile);
-						merged_files.push_back(destfile);
-						nf++;
+					sprintf(destfile, "%d_%d_silence.aac", nf, audio_index);
+					split_audio(silence_aac_file, 0, silence_audio_end - silence_audio_start, destfile);
+					merge_audio_files.push_back(destfile);
+					tmp_files.push_back(destfile);
 
-						if (video.start_time >= audio.end_time - 0.1){//all audio file no video, to next audio
-							audio_start = audio.end_time + 0.1;//no audio left
-							break;
-						}
+					if (audio.end_time > video.end_time - 0.1 && audio.end_time < video.end_time + 0.1) {//just match
+						merge_audio_files.push_back(audio.name);
+						need_append_silence = false;
+						audio_start = audio.end_time + 0.1;
+						audio_index++;//this audio is used
+						break;
 					}
-					//----AS--------
-					//----VS--------
-					else if (audio_start - video.start_time < 0.1){
-
-						if (audio.end_time > video.end_time){ //this video finish, to next video
-							sprintf(destfile, "%d_%s", nf, audio.name.c_str());
-							split_audio(audio.name.c_str(), video.start_time, video.end_time - video.start_time, destfile);
-							tmp_files.push_back(destfile);
-
-							audio_start = video.end_time;
-							sprintf(destfile, "%d.ts", nf);
-							merge_audio_video(audio, nf, video, destfile);
-							merged_files.push_back(destfile);
-							nf++;
-						}
-						else if (video.end_time - audio.end_time < 0.1){//just fine, this audio file finish
-							sprintf(destfile, "%d_%s", nf, audio.name.c_str());
-							split_audio(audio.name.c_str(), video.start_time, audio.end_time - video.start_time, destfile);
-							tmp_files.push_back(destfile);
-
-							sprintf(destfile, "%d.ts", nf);
-							merge_audio_video(audio, nf, video, destfile);
-							merged_files.push_back(destfile);
-							audio_start = audio.end_time + 0.1;//no audio left
-							nf++;
-							nv++;//this video is used
-							break;
+					if (audio.end_time > video.end_time){
+						sprintf(destfile, "%d_%s", nf, audio.name.c_str());
+						merge_audio_files.push_back(destfile);
+						tmp_files.push_back(destfile);
+						split_audio(audio.name.c_str(), 0, video.end_time - audio.start_time, destfile);
+						need_append_silence = false;
+						//adjust timecode for the audio is part left
+						float cur_audio_start = video.end_time - audio.start_time;
+
+						audio_start = audio.start_time;
+
+						for (int j = audio_index + 1; j < filesaudio.size(); j++){
+							filesaudio[j].start_time -= audio_start;
+							filesaudio[j].end_time -= audio_start;
 						}
-						else { // this audio finish,add silence and/or next audio
-							sprintf(destfile, "%d_%s", nf, audio.name.c_str());
-							split_audio(audio.name.c_str(), video.start_time, audio.end_time - video.start_time, destfile);
-							vector<std::string > merge_audio_files;
-							merge_audio_files.push_back(destfile);
-							tmp_files.push_back(destfile);
-
-							audio_start = audio.end_time + 0.1;
-
-							double silence_audio_start = audio.end_time;
-							double silence_audio_end = video.end_time;
-
-							bool need_silence = true;
-							bool to_next_video = false;
-							for (; i + 1 < filesaudio.size(); i++){//since video is not finished,try find next audio
-								audio = filesaudio[i + 1];
-								if (audio.start_time < video.end_time) {//next audio should split to fit the video
-									silence_audio_end = audio.start_time;
-
-									sprintf(destfile, "%d_%d_silence.aac", nf, i);
-									split_audio(silence_aac_file, 0, silence_audio_end - silence_audio_start, destfile);
-									merge_audio_files.push_back(destfile);
-									tmp_files.push_back(destfile);
-
-									if (audio.end_time > video.end_time - 0.1 && audio.end_time < video.end_time + 0.1) {//just match
-										merge_audio_files.push_back(audio.name);
-										need_silence = false;
-										audio_start = audio.end_time + 0.1;
-										i++;
-										break;
-									}
-									if (audio.end_time > video.end_time){
-										sprintf(destfile, "%d_%s", nf, audio.name.c_str());
-										merge_audio_files.push_back(destfile);
-										tmp_files.push_back(destfile);
-										split_audio(audio.name.c_str(), 0, video.end_time - audio.start_time, destfile);
-										need_silence = false;
-										//adjust timecode for the audio is part left
-										float cur_audio_start = video.end_time - audio.start_time;
-
-										audio_start = audio.start_time;
-
-										for (int j = i + 1; j < filesaudio.size(); j++){
-											filesaudio[j].start_time -= audio_start;
-											filesaudio[j].end_time -= audio_start;
-										}
-
-										for (int j = nv; j < filesvideo.size(); j++) {
-											filesvideo[j].start_time -= audio_start;
-											filesvideo[j].end_time -= audio_start;
-										}
-										i++;
-										audio = filesaudio[i];
-										audio_start = cur_audio_start;
-
-										to_next_video = true;
-
-										break;
-									}
-									merge_audio_files.push_back(audio.name);//whole audio should be appended
-									silence_audio_start = audio.end_time; //adjust the silence start
-									audio_start = audio.end_time + 0.1;
-								}
-								else {
-									break;//no need for next audio
-								}
-							}
-
-							if (need_silence) {
-								sprintf(destfile, "%d_silence.aac", nf);
-								split_audio(silence_aac_file, 0, silence_audio_end - silence_audio_start, destfile);
-								merge_audio_files.push_back(destfile);
-								tmp_files.push_back(destfile);
-							}
-
-							sprintf(audio_file, "%d_merged.aac", nf);
-							merge_audio_file(merge_audio_files, audio_file);
-
-
-							sprintf(destfile, "%d.ts", nf);
-							merge_audio_video(audio_file, video.name.c_str(), destfile);
-							merged_files.push_back(destfile);
-							nf++;
-
-							if (!to_next_video){
-								nv++;
-								break;
-							}
+
+						for (int j = nv; j < filesvideo.size(); j++) {
+							filesvideo[j].start_time -= audio_start;
+							filesvideo[j].end_time -= audio_start;
 						}
+						audio_index++;
+						audio = filesaudio[audio_index];
+						audio_start = cur_audio_start;
+
+						to_next_video = true;
+
+						break;
 					}
+					merge_audio_files.push_back(audio.name);//whole audio should be appended
+					silence_audio_start = audio.end_time; //adjust the silence start
+				}
+				else {
+					break;//no need for next audio
 				}
+			}//end audio find for the video
+		}//end else
+
+		if (need_append_silence) {
+			sprintf(destfile, "%d_silence.aac", nf);
+			split_audio(silence_aac_file, 0, silence_audio_end - silence_audio_start, destfile);
+			merge_audio_files.push_back(destfile);
+			tmp_files.push_back(destfile);
+		}
+
+		sprintf(audio_file, "%d_merged.aac", nf);
+		merge_audio_file(merge_audio_files, audio_file);
+
+
+		sprintf(destfile, "%d.ts", nf);
+		merge_audio_video(audio_file, video.name.c_str(), destfile);
+		merged_files.push_back(destfile);
+		nf++;
+
+		if (!to_next_video){
+			nv++;//this video is processed
+			return 1;
+		}
+	}//end need combine
+	return 0;
+}
+
+int MergeProcess::process_video_behind_audio()
+{
+	sprintf(audio_file, "%d_%s", nf, audio.name.c_str());
+	if (video.start_time < audio.end_time - 0.1){
+		split_audio(audio.name.c_str(), audio_start, video.start_time - audio_start, audio_file);
+		audio_start = video.start_time;
+	}
+	else {
+		split_audio(audio.name.c_str(), audio_start, audio.end_time - audio_start, audio_file);
+	}
+	tmp_files.push_back(audio_file);
+
+	sprintf(pic_file, "%s.jpg", video.name.c_str());
+	get_video_first_frame_jpeg(video, pic_file);
+	tmp_files.push_back(pic_file);
 
-				if (audio_start < audio.end_time){
+	sprintf(destfile, "%d.ts", nf);
+	merge_audio_pic(audio_file, pic_file, destfile);
+	merged_files.push_back(destfile);
+	nf++;
+
+	if (video.start_time >= audio.end_time - 0.1){//all audio file no video, to next audio
+		audio_start = audio.end_time + 0.1;//no audio left
+		return 1;
+	}
+	return 0;
+}
+
+int MergeProcess::process_video_align_audio()
+{
+	if (audio.end_time > video.end_time){ //this video finish, to next video
+		sprintf(destfile, "%d_%s", nf, audio.name.c_str());
+		split_audio(audio.name.c_str(), video.start_time, video.end_time - video.start_time, destfile);
+		tmp_files.push_back(destfile);
+
+		audio_start = video.end_time;
+		sprintf(destfile, "%d.ts", nf);
+		merge_audio_video(audio, nf, video, destfile);
+		merged_files.push_back(destfile);
+		nf++;
+	}
+	else if (video.end_time - audio.end_time < 0.1){//just fine, this audio file finish
+		sprintf(destfile, "%d_%s", nf, audio.name.c_str());
+		split_audio(audio.name.c_str(), video.start_time, audio.end_time - video.start_time, destfile);
+		tmp_files.push_back(destfile);
+
+		sprintf(destfile, "%d.ts", nf);
+		merge_audio_video(audio, nf, video, destfile);
+		merged_files.push_back(destfile);
+		audio_start = audio.end_time + 0.1;//no audio left
+		nf++;
+		nv++;//this video is used
+		return 1;
+	}
+	else { // this audio finish,add silence and/or next audio
+		sprintf(destfile, "%d_%s", nf, audio.name.c_str());
+		split_audio(audio.name.c_str(), video.start_time, audio.end_time - video.start_time, destfile);
+		vector<std::string > merge_audio_files;
+		merge_audio_files.push_back(destfile);
+		tmp_files.push_back(destfile);
+
+		audio_start = audio.end_time + 0.1;
+
+		double silence_audio_start = audio.end_time;
+		double silence_audio_end = video.end_time;
+
+		bool need_silence = true;
+		bool to_next_video = false;
+		for (; audio_index + 1 < filesaudio.size(); audio_index++){//since video is not finished,try find next audio
+			audio = filesaudio[audio_index + 1];
+			if (audio.start_time < video.end_time) {//next audio should split to fit the video
+				silence_audio_end = audio.start_time;
+
+				sprintf(destfile, "%d_%d_silence.aac", nf, audio_index);
+				split_audio(silence_aac_file, 0, silence_audio_end - silence_audio_start, destfile);
+				merge_audio_files.push_back(destfile);
+				tmp_files.push_back(destfile);
+
+				if (audio.end_time > video.end_time - 0.1 && audio.end_time < video.end_time + 0.1) {//just match
+					merge_audio_files.push_back(audio.name);
+					need_silence = false;
+					audio_start = audio.end_time + 0.1;
+					audio_index++;
+					break;
+				}
+				if (audio.end_time > video.end_time){
 					sprintf(destfile, "%d_%s", nf, audio.name.c_str());
-					split_audio(audio.name.c_str(), audio_start, audio.end_time - audio_start, destfile);
+					merge_audio_files.push_back(destfile);
 					tmp_files.push_back(destfile);
+					split_audio(audio.name.c_str(), 0, video.end_time - audio.start_time, destfile);
+					need_silence = false;
+					//adjust timecode for the audio is part left
+					float cur_audio_start = video.end_time - audio.start_time;
 
-					sprintf(destfile, "%d.ts", nf);
-					if (nv < filesvideo.size()) {
-						fileinfo video = filesvideo[nv];
-						sprintf(pic_file, "%s.jpg", video.name.c_str());
-						get_video_first_frame_jpeg(video, pic_file);
-						tmp_files.push_back(pic_file);
+					audio_start = audio.start_time;
+
+					for (int j = audio_index + 1; j < filesaudio.size(); j++){
+						filesaudio[j].start_time -= audio_start;
+						filesaudio[j].end_time -= audio_start;
 					}
-					else {
-						strcpy(pic_file, blank_pic_file);
+
+					for (int j = nv; j < filesvideo.size(); j++) {
+						filesvideo[j].start_time -= audio_start;
+						filesvideo[j].end_time -= audio_start;
 					}
-					merge_audio_pic(audio, nf, pic_file, destfile);
-					merged_files.push_back(destfile);
-					nf++;
+					audio_index++;
+					audio = filesaudio[audio_index];
+					audio_start = cur_audio_start;
+
+					to_next_video = true;
+
+					break;
 				}
+				merge_audio_files.push_back(audio.name);//whole audio should be appended
+				silence_audio_start = audio.end_time; //adjust the silence start
+				audio_start = audio.end_time + 0.1;
+			}
+			else {
+				break;//no need for next audio
 			}
 		}
+
+		if (need_silence) {
+			sprintf(destfile, "%d_silence.aac", nf);
+			split_audio(silence_aac_file, 0, silence_audio_end - silence_audio_start, destfile);
+			merge_audio_files.push_back(destfile);
+			tmp_files.push_back(destfile);
+		}
+
+		sprintf(audio_file, "%d_merged.aac", nf);
+		merge_audio_file(merge_audio_files, audio_file);
+
+
+		sprintf(destfile, "%d.ts", nf);
+		merge_audio_video(audio_file, video.name.c_str(), destfile);
+		merged_files.push_back(destfile);
+		nf++;
+
+		if (!to_next_video){
+			nv++;
+			return 1;
+		}
+	}
+	return 0;
+}
+
+int MergeProcess::process_video_loop()
+{
+	for (; nv < filesvideo.size(); nv++) {
+		video = filesvideo[nv];
+
+		if (video.start_time < audio_start - 0.1)
+		{//video is much more ahead of audio,try padding silence first
+			if (process_video_ahead_of_audio())
+				break;
+		}//end video is ahead of audio
+		if (video.start_time - audio_start > 0.1) {//video is behind audio too much
+			if (process_video_behind_audio())
+				break;
+		}
+		else if (audio_start - video.start_time < 0.1){
+			if (process_video_align_audio())
+				break;
+		}
+	}
+	return 0;
+}
+
+int MergeProcess::process_va()
+{
+	for (audio_index = 0; audio_index < filesaudio.size(); audio_index++){  //
+
+		adjust_va_timestamp();
+
+		audio = filesaudio[audio_index];
+		audio_start = 0;//for a new processing audio,the start is 0
+
+		process_video_loop();
+
+		merge_left_audio();
+	}
+	return 0;
+}
+
+
+int MergeProcess::process_files(const char * output_dest_file)
+{
+	//don't split video, for a video, using merged audios to mix with it
+	//for audio, mix with video or jpg
+
+	if (filesvideo.size()) {//has video files
+		if (filesaudio.size()){
+			process_va(); //process the case both audio and video files exist
+		}
 		else{//only video
 			if (filesvideo.size() == 1){
 				fileinfo video = filesvideo[0];
@@ -772,7 +845,6 @@ int process_files(const char * output_dest_file)
 
 	adjust_dest_timecode("m.ts", output_dest_file);
 
-
 	if (!keep_tmp_files) {
 		removefiles(tmp_files);
 		removefiles(merged_files);
@@ -866,7 +938,8 @@ int main(int argc, char * argv[])
 
 	load_codec_param();
 
-	process_files("dest.ts");
+	MergeProcess mp(media_files[type_audio], media_files[type_video]);
+	mp.process_files("dest.ts");
 
 	return 0;
 }