Big change to new stream configuration. Possibility to include output channels back to input

CMakeLists.txt

@@ -76,7 +76,8 @@ else()
     set(win32 false)
 	set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fPIC -std=c11 \
 	-Werror=incompatible-pointer-types")
-		
+    include_directories(/usr/local/include/rtaudio)
+    link_directories(/usr/local/lib)
 
 	
 endif(CMAKE_SYSTEM_NAME STREQUAL "Windows")

lasp/__init__.py

@@ -6,3 +6,4 @@ from .lasp_octavefilter import *
 from .lasp_slm import *
 from .lasp_weighcal import *
 from .wrappers import *
+from .device import AvType

lasp/c/lasp_pyarray.h

@@ -0,0 +1,76 @@
+// ascee_python.h
+//
+// Author: J.A. de Jong - ASCEE
+//
+// Description:
+// Routine to generate a Numpy array from an arbitrary buffer. Careful, this
+// code should both be C and C++ compatible!!
+//////////////////////////////////////////////////////////////////////
+#pragma once
+#ifndef LASP_PYARRAY_H
+#define LASP_PYARRAY_H
+
+#define TRACERPLUS (-10)
+#include <numpy/ndarrayobject.h>
+#ifdef LASP_DOUBLE_PRECISION
+#define LASP_NUMPY_FLOAT_TYPE NPY_FLOAT64
+#define LASP_NUMPY_COMPLEX_TYPE NPY_COMPLEX128
+#else
+#define LASP_NUMPY_FLOAT_TYPE NPY_FLOAT32
+#endif
+
+#ifdef MS_WIN64
+/**
+ * Function passed to Python to use for cleanup of
+ * foreignly obtained data.
+ **/
+static inline void capsule_cleanup(void *capsule) {
+    void *memory = PyCapsule_GetPointer(capsule, NULL);
+    free(memory);
+}
+
+#endif
+
+static inline PyObject *data_to_ndarray(void *data, int n_rows, int n_cols,
+        int typenum, bool transfer_ownership,
+        bool F_contiguous) {
+
+    /* fprintf(stderr, "Enter data_to_ndarray\n"); */
+    assert(data);
+    import_array();
+
+    npy_intp dims[2] = {n_rows, n_cols};
+    assert(n_rows > 0);
+    assert(n_cols > 0);
+
+    PyArrayObject *arr =
+        (PyArrayObject *)PyArray_SimpleNewFromData(2, dims, typenum, data);
+
+    if (!arr) {
+        return NULL;
+    }
+    if (F_contiguous) {
+        PyArray_ENABLEFLAGS(arr, NPY_ARRAY_F_CONTIGUOUS);
+    }
+
+    if (transfer_ownership == true) {
+        #ifdef MS_WIN64
+            // The default destructor of Python cannot free the data, as it is allocated
+            // with malloc. Therefore, with this code, we tell Numpy/Python to use
+            // the capsule_cleanup constructor. See:
+            // https://stackoverflow.com/questions/54269956/crash-of-jupyter-due-to-the-use-of-pyarray-enableflags/54278170#54278170
+            // Note that in general it was disadvised to build all C code with MinGW on
+            // Windows. We do it anyway, see if we find any problems on the way.
+            void *capsule = PyCapsule_New(mat->_data, NULL, capsule_cleanup);
+            PyArray_SetBaseObject(arr, capsule);
+        #endif
+        /* fprintf(stderr, "============Ownership transfer================\n"); */
+        PyArray_ENABLEFLAGS(arr, NPY_OWNDATA);
+    }
+    /* fprintf(stderr, "Exit data_to_ndarray\n"); */
+
+    return (PyObject *) arr;
+}
+
+#endif // LASP_PYARRAY_H
+//////////////////////////////////////////////////////////////////////

lasp/c/lasp_python.h

@@ -9,25 +9,8 @@
 #ifndef LASP_PYTHON_H
 #define LASP_PYTHON_H
 #define TRACERPLUS (-10)
-#include <numpy/ndarrayobject.h>
+#include "lasp_pyarray.h"
-#ifdef LASP_DOUBLE_PRECISION
-#define LASP_NUMPY_FLOAT_TYPE NPY_FLOAT64
-#define LASP_NUMPY_COMPLEX_TYPE NPY_COMPLEX128
-#else
-#define LASP_NUMPY_FLOAT_TYPE NPY_FLOAT32
-#endif
 
-#ifdef MS_WIN64
-/** 
- * Function passed to Python to use for cleanup of
- * foreignly obtained data.
- **/
-static inline void capsule_cleanup(void* capsule) {
-	void *memory = PyCapsule_GetPointer(capsule, NULL);
-    free(memory);
-	}
-
-#endif
 /** 
  * Create a numpy array from an existing dmat.
  *
@@ -38,50 +21,31 @@ static inline void capsule_cleanup(void* capsule) {
  * @return Numpy array
  */
 static inline PyObject* dmat_to_ndarray(dmat* mat,bool transfer_ownership) {
-    fsTRACE(15);
     dbgassert(mat,NULLPTRDEREF);
+    dbgassert(mat->_data,NULLPTRDEREF);
+    /* fprintf(stderr, "Enter dmat_to_ndarray\n"); */
 
-    import_array();
-    
     // Dimensions given in wrong order, as mat is
     // Fortran-contiguous. Later on we transpose the result. This is
     // more easy than using the PyArray_New syntax.
-    npy_intp dims[] = {mat->n_cols,mat->n_rows};
+    PyObject* arr = data_to_ndarray(mat->_data,
-    PyObject* arr_t = PyArray_SimpleNewFromData(2,dims,
+            mat->n_rows,
-                                              LASP_NUMPY_FLOAT_TYPE,
+            mat->n_cols,
-                                              mat->_data);
+            LASP_NUMPY_FLOAT_TYPE,
-    if(!arr_t) {
+            transfer_ownership,
+            true); // Fortran-contiguous
+
+    if(transfer_ownership) {
+        mat->_foreign_data = true;
+    }
+
+    if(!arr) {
         WARN("Array creation failure");
         feTRACE(15);
         return NULL;
     }
+    /* fprintf(stderr, "Exit dmat_to_ndarray\n"); */
 
-    if(transfer_ownership) {
-        mat->_foreign_data = true;
-#ifdef MS_WIN64
-		// The default destructor of Python cannot free the data, as it is allocated
-		// with malloc. Therefore, with this code, we tell Numpy/Python to use
-		// the capsule_cleanup constructor. See:
-		// https://stackoverflow.com/questions/54269956/crash-of-jupyter-due-to-the-use-of-pyarray-enableflags/54278170#54278170
-		// Note that in general it was disadvised to build all C code with MinGW on Windows.
-		// We do it anyway, see if we find any problems on the way.
-		void* capsule = PyCapsule_New(mat->_data, NULL, capsule_cleanup);
-		PyArray_SetBaseObject( arr_t, capsule);
-#else		
-        PyArray_ENABLEFLAGS(arr_t, NPY_OWNDATA);
-#endif
-    }
-
-    // Transpose the array
-    PyObject* arr = PyArray_Transpose((PyArrayObject*) arr_t,NULL);
-    if(!arr) {
-        WARN("Array transpose failure");
-        feTRACE(15);
-        return NULL;
-    }
-    Py_DECREF(arr_t);
- 
-    feTRACE(15);
     return arr;
 }
 

lasp/device/__init__.py

@@ -1,6 +1,3 @@
 #!/usr/bin/python3
-__all__ = ['DAQConfiguration']
+from .lasp_rtaudio import *
-from .lasp_daqconfig import DAQConfiguration, DAQInputChannel, DeviceInfo
+from .lasp_daqconfig import *
-from .lasp_rtaudio import (RtAudio, 
-                           get_numpy_dtype_from_format_string,
-                           get_sampwidth_from_format_string)

lasp/device/lasp_daqconfig.py

@@ -9,11 +9,14 @@ Data Acquistiion (DAQ) device descriptors, and the DAQ devices themselves
 
 """
 from dataclasses import dataclass, field
+from dataclasses_json import dataclass_json
 import numpy as np
+from ..lasp_common import lasp_shelve, Qty
 
 
 @dataclass
 class DeviceInfo:
+    api: int
     index: int
     probed: bool
     name: str
@@ -25,17 +28,17 @@ class DeviceInfo:
     prefsamplerate: int
 
 
-from ..lasp_common import lasp_shelve
 
+@dataclass_json
 @dataclass
-class DAQInputChannel:
+class DAQChannel:
     channel_enabled: bool
     channel_name: str
     sensitivity: float
+    unit: Qty
 
 
-
+@dataclass_json
-
 @dataclass
 class DAQConfiguration:
     """
@@ -44,6 +47,9 @@ class DAQConfiguration:
     Args:
         duplex_mode: Set device to duplex mode, if possible
         monitor_gen: If set to true, add monitor channel to recording.
+        outputDelayBlocks: number of blocks to delay output stream when added
+        to input for monitoring the output synchronously with input.
+
         input_device_name: ASCII name with which to open the device when connected
         outut_device_name: ASCII name with which to open the device when connected
 
@@ -66,6 +72,7 @@ class DAQConfiguration:
 
 
     """
+    api: int
     duplex_mode: bool
 
     input_device_name: str
@@ -76,28 +83,12 @@ class DAQConfiguration:
     en_input_rate: int
     en_output_rate: int
 
-    input_channel_configs: list = None
+    input_channel_configs: list
-    monitor_gen: bool = False
+    output_channel_configs: list
+    monitor_gen: bool
 
-
+    outputDelayBlocks: int
-    def __post_init__(self):
+    nFramesPerBlock: int
-        """
-        We do a check here to see whether the list of enabled channels is
-        contiguous. Non-contiguous is not yet implemented in RtAudio backend.
-        """
-        en_input = self.input_channel_configs
-        first_ch_enabled_found = False
-        ch_disabled_found_after = False
-        for ch in en_input:
-            if ch.channel_enabled:
-                first_ch_enabled_found = True
-                if ch_disabled_found_after:
-                    raise ValueError('Error: non-contiguous array of channels'
-                                    ' found. This is not yet implemented in'
-                                     ' backend')
-            else:
-                if first_ch_enabled_found:
-                    ch_disabled_found_after = True
 
     def firstEnabledInputChannelNumber(self):
         """
@@ -109,18 +100,55 @@ class DAQConfiguration:
                 return i
         return -1
 
+    def firstEnabledOutputChannelNumber(self):
+        """
+        Returns the channel number of the first enabled output channel. Returns -1 if
+        no channels are enabled.
+        """
+        for i, ch in enumerate(self.output_channel_configs):
+            if ch.channel_enabled:
+                return i
+        return -1
 
-    def getEnabledChannels(self):
+    def lastEnabledInputChannelNumber(self):
+        last = -1
+        for i, ch in enumerate(self.input_channel_configs):
+            if ch.channel_enabled:
+                last = i
+        return last
+
+    def lastEnabledOutputChannelNumber(self):
+        last = -1
+        for i, ch in enumerate(self.output_channel_configs):
+            print(ch)
+            if ch.channel_enabled:
+                last = i
+        return last
+
+    def getEnabledInputChannels(self):
         en_channels = []
         for chan in self.input_channel_configs:
             if chan.channel_enabled:
                 en_channels.append(chan)
         return en_channels
 
-    def getEnabledChannelSensitivities(self):
+    def getEnabledInputChannelNames(self):
-        return np.array(
+        return [ch.channel_name for ch in self.getEnabledInputChannels()]
-                [float(channel.sensitivity) for channel in
+
-                    self.getEnabledChannels()])
+    def getEnabledInputChannelSensitivities(self):
+        return [float(channel.sensitivity) for channel in
+                    self.getEnabledInputChannels()]
+
+    def getEnabledOutputChannels(self):
+        en_channels = []
+        for chan in self.output_channel_configs:
+            if chan.channel_enabled:
+                en_channels.append(chan)
+        return en_channels
+
+    def getEnabledOutputChannelSensitivities(self):
+        return [float(channel.sensitivity) for channel in
+                    self.getEnabledOutputChannels()]
 
     @staticmethod
     def loadConfigs():

lasp/device/lasp_rtaudio.pyx

@@ -9,6 +9,15 @@ from libc.stdio cimport printf, fprintf, stderr
 from libc.string cimport memcpy, memset
 from cpython.ref cimport PyObject,Py_INCREF, Py_DECREF
 
+__all__ = ['AvType', 'RtAudio', 'get_numpy_dtype_from_format_string',
+          'get_sampwidth_from_format_string']
+
+class AvType:
+    """Specificying the type of data, for adding and removing callbacks from
+    the stream."""
+    audio_input = 1
+    audio_output = 2
+    video = 4
 
 cdef extern from "RtAudio.h" nogil:
     ctypedef unsigned long RtAudioStreamStatus
@@ -47,7 +56,7 @@ cdef extern from "RtAudio.h" nogil:
 
     ctypedef int (*RtAudioCallback)(void* outputBuffer,
                                     void* inputBuffer,
-                                    unsigned int nFrames,
+                                    unsigned int nFramesPerBlock,
                                     double streamTime,
                                     RtAudioStreamStatus status,
                                     void* userData)
@@ -56,6 +65,15 @@ cdef extern from "RtAudio.h" nogil:
                                           const string& errortxt)
 
     cdef cppclass cppRtAudio "RtAudio":
+        enum Api:
+            UNSPECIFIED
+            LINUX_ALSA
+            LINUX_PULSE
+            MACOSX_CORE
+            WINDOWS_WASAPI
+            WINDOWS_ASIO
+            WINDOWS_DS
+
         cppclass DeviceInfo:
             bool probed
             string name
@@ -69,17 +87,35 @@ cdef extern from "RtAudio.h" nogil:
             RtAudioFormat nativeFormats
 
         cppclass StreamOptions:
+            StreamOptions()
             RtAudioStreamFlags flags
             unsigned int numberOfBuffers
             string streamName
             int priority
         cppclass StreamParameters:
+            StreamParameters()
             unsigned int deviceId
             unsigned int nChannels
             unsigned int firstChannel
 
-        RtAudio() except +
+        @staticmethod
+        void getCompiledApi(vector[cppRtAudio.Api]& apis)
+
+        @staticmethod
+        cppRtAudio.Api getCompiledApiByName(string& name)
+
+        @staticmethod
+        string getApiDisplayName(cppRtAudio.Api api) 
+
+        @staticmethod
+        string getApiName(cppRtAudio.Api api) 
+
+        # RtAudio() except +
+        cppRtAudio() except +
+        cppRtAudio(cppRtAudio.Api api) except +
+
         # ~RtAudio() Destructors should not be listed
+
         unsigned int getDeviceCount()
         DeviceInfo getDeviceInfo(unsigned int device)
         unsigned int getDefaultOutputDevice()
@@ -126,6 +162,13 @@ cdef extern from "atomic" namespace "std" nogil:
         T load()
         void store(T)
 
+cdef extern from "lasp_pyarray.h":
+    PyObject* data_to_ndarray(void* data,
+                              int n_rows,int n_cols,
+                              int typenum,
+                              bool transfer_ownership,
+                              bool F_contiguous)
+
 _formats_strkey = {
         '8-bit integers': (RTAUDIO_SINT8,   1, np.int8),
         '16-bit integers': (RTAUDIO_SINT16, 2, np.int16),
@@ -135,10 +178,10 @@ _formats_strkey = {
         '64-bit floats': (RTAUDIO_FLOAT64,  8, np.float64),
 }
 _formats_rtkey = {
-        RTAUDIO_SINT8: ('8-bit integers', 1, cnp.NPY_INT8),
+        RTAUDIO_SINT8: ('8-bit integers',   1, cnp.NPY_INT8),
-        RTAUDIO_SINT16: ('16-bit integers',2, cnp.NPY_INT16),
+        RTAUDIO_SINT16: ('16-bit integers', 2, cnp.NPY_INT16),
-        RTAUDIO_SINT24: ('24-bit integers',3),
+        RTAUDIO_SINT24: ('24-bit integers', 3),
-        RTAUDIO_SINT32: ('32-bit integers',4, cnp.NPY_INT32),
+        RTAUDIO_SINT32: ('32-bit integers', 4, cnp.NPY_INT32),
         RTAUDIO_FLOAT32: ('32-bit floats',  4, cnp.NPY_FLOAT32),
         RTAUDIO_FLOAT64: ('64-bit floats',  8, cnp.NPY_FLOAT64),
 }
@@ -153,52 +196,60 @@ def get_sampwidth_from_format_string(format_string):
 # It took me quite a long time to fully understand Cython's idiosyncrasies
 # concerning C(++) callbacks, the GIL and passing Python objects as pointers
 # into C(++) functions. But finally, here it is!
-cdef object fromBufferToNPYNoCopy(
-                             cnp.NPY_TYPES buffer_format_type,
-                             void* buf,
-                             size_t nchannels,
-                             size_t nframes):
-    cdef cnp.npy_intp[2] dims = [nframes, nchannels]
- 
-    # Interleaved data is C-style contiguous. Therefore, we can directly use
-    # SimpleNewFromData()
-    array = cnp.PyArray_SimpleNewFromData(2, &dims[0], buffer_format_type,
-            buf)
 
-    return array
+# cdef void fromNPYToBuffer(cnp.ndarray arr,
+#                           void* buf):
+#     """
+#     Copy a Python numpy array over to a buffer
+#     No checks, just memcpy! Careful!
+#     """
+#     memcpy(buf, arr.data, arr.size*arr.itemsize)
 
+cdef void copyChannel(void* to, void* from_,
+                      unsigned bytesperchan,
+                      unsigned toindex,
+                      unsigned fromindex) nogil:
+    memcpy(<void*> &((<char*> to)[bytesperchan*toindex]),
+           <void*> &((<char*> from_)[bytesperchan*fromindex]),
+           bytesperchan)
 
-cdef void fromNPYToBuffer(cnp.ndarray arr,
-                          void* buf):
-    """
-    Copy a Python numpy array over to a buffer
-    No checks, just memcpy! Careful!
-    """
-    memcpy(buf, arr.data, arr.size*arr.itemsize)
 
 ctypedef struct _Stream:
     PyObject* pyCallback
     RtAudioFormat sampleformat
 
+    # Flag used to pass the stopThread.
     atomic[bool] stopThread
 
-    unsigned int nFrames
+    # Number of frames per block
+    unsigned nFramesPerBlock
+    # Number of bytes per channel
+    unsigned int nBytesPerChan
 
+    # Number of blocks to delay the output before adding to the input
+    unsigned int outputDelayBlocks
+
+    # The structures as used by RtAudio
     cppRtAudio.StreamParameters inputParams
     cppRtAudio.StreamParameters outputParams
 
+    bool* inputChannelsEnabled
+    bool* outputChannelsEnabled
+
+    unsigned ninputchannels_forwarded
+    unsigned noutputchannels_forwarded
+
     # If these queue pointers are NULL, it means the stream does not have an
     # input, or output.
+    SafeQueue[void*] *outputDelayQueue
     SafeQueue[void*] *inputQueue
     SafeQueue[void*] *outputQueue
-    size_t inputbuffersize  # Full size of the output buffer, in BYTES
-    size_t outputbuffersize # Full size of the output buffer, in BYTES
     CPPThread[void*, void (*)(void*)] *thread
 
-    
+
 cdef int audioCallback(void* outputbuffer,
                        void* inputbuffer,
-                       unsigned int nFrames,
+                       unsigned int nFramesPerBlock,
                        double streamTime,
                        RtAudioStreamStatus status,
                        void* userData) nogil:
@@ -211,8 +262,22 @@ cdef int audioCallback(void* outputbuffer,
         _Stream* stream
         void* outputbuffercpy = NULL
         void* inputbuffercpy = NULL
+        unsigned j, i
+        unsigned bytesperchan
+        unsigned noutputchannels_forwarded, ninputchannels_forwarded
+        bint ch_en
 
     stream = <_Stream*>(userData)
+    bytesperchan = stream.nBytesPerChan
+    ninputchannels_forwarded = stream.ninputchannels_forwarded
+    noutputchannels_forwarded = stream.noutputchannels_forwarded
+
+    # with gil:
+    #     print(f'bytesperchan: {bytesperchan}')
+    #     print(f'ninputchannels_forwarded:: {ninputchannels_forwarded}')
+    #     print(f'noutputchannels_forwarded:: {noutputchannels_forwarded}')
+    # fprintf(stderr, "Stream heartbeat...\n")
+
 
     # Returning 2 means aborting the stream immediately
     if status == RTAUDIO_INPUT_OVERFLOW:
@@ -224,35 +289,78 @@ cdef int audioCallback(void* outputbuffer,
         # stream.stopThread.store(True)
         return 0
 
-    if nFrames != stream.nFrames:
+    if nFramesPerBlock != stream.nFramesPerBlock:
         printf('Number of frames mismath in callback data!\n')
         stream.stopThread.store(True)
         return 2
 
     if inputbuffer:
-        # assert stream.inputQueue is not NULL
+        # fprintf(stderr, "enter copying input buffer code\n")
-        inputbuffercpy = malloc(stream.inputbuffersize)
+        # with gil:
-        memcpy(inputbuffercpy, inputbuffer,
+        #     assert stream.inputQueue is not NULL
-               stream.inputbuffersize)
+        inputbuffercpy = malloc(bytesperchan*ninputchannels_forwarded)
+        if not inputbuffercpy:
+            fprintf(stderr, "Error allocating buffer\n")
+            return 2
+
+        if stream.outputDelayQueue:
+            if stream.outputDelayQueue.size() > stream.outputDelayBlocks:
+                outputbuffercpy = stream.outputDelayQueue.dequeue()
+                memcpy(inputbuffercpy, outputbuffercpy,
+                       bytesperchan*noutputchannels_forwarded)
+
+                # Cleanup buffer
+                free(outputbuffercpy)
+                outputbuffercpy = NULL
+            else:
+                memset(inputbuffercpy, 0,
+                       bytesperchan*noutputchannels_forwarded)
+
+            # Starting channel for copying input channels according to the channel
+            # map
+            j = stream.noutputchannels_forwarded
+        else:
+            j = 0
+        i = 0
+        for i in range(stream.inputParams.nChannels):
+            ch_en = stream.inputChannelsEnabled[i]
+            if ch_en:
+                copyChannel(inputbuffercpy, inputbuffer, bytesperchan, j, i)
+                j+=1
+            i+=1
+
         stream.inputQueue.enqueue(inputbuffercpy)
 
     if outputbuffer:
-        # assert stream.outputQueue is not NULL
+        # with gil:
+        #     assert stream.outputQueue
+        # fprintf(stderr, "enter copying output buffer code\n")
         if stream.outputQueue.empty():
             fprintf(stderr, 'Stream output buffer underflow, zero-ing buffer...\n')
-            # Pre-stack three empty output buffers
+            memset(outputbuffer, 0, stream.outputParams.nChannels*bytesperchan)
-            # printf('Pre-stacking\n')
+        else:
-            #     outputbuffer = malloc(stream.outputbuffersize)
+            outputbuffercpy = stream.outputQueue.dequeue()
-            memset(outputbuffer, 0, stream.outputbuffersize)
+            # fprintf(stderr, 'Copying data to stream output buffer...\n')
-            if stream.inputQueue:
+            j = 0
-                stream.inputQueue.enqueue(NULL)
+            i = 0
-            return 0
+            for i in range(stream.outputParams.nChannels):
-        
+                ch_en = stream.outputChannelsEnabled[i]
-        outputbuffercpy = stream.outputQueue.dequeue()
+                if ch_en:
-        memcpy(outputbuffer, outputbuffercpy,
+                    copyChannel(outputbuffer, outputbuffercpy, bytesperchan, i, j)
-               stream.outputbuffersize)
+                    j+=1
-        free(outputbuffercpy)
+                else:
+                    # If channel is not enabled, we set the data to zero
+                    memset(<void*> &((<char*> outputbuffer)[bytesperchan*i]), 0, bytesperchan)
+                    pass
+                i+=1
 
+            if stream.outputDelayQueue:
+                # fprintf(stderr, "Adding to delay queue\n")
+                stream.outputDelayQueue.enqueue(outputbuffercpy)
+            else:
+                free(outputbuffercpy)
+
+            outputbuffercpy = NULL
 
     return 0
 
@@ -263,69 +371,104 @@ cdef void audioCallbackPythonThreadFunction(void* voidstream) nogil:
         cnp.NPY_TYPES npy_format
         void* inputbuffer = NULL
         void* outputbuffer = NULL
+        unsigned noutputchannels_forwarded
+        unsigned ninputchannels_forwarded
+        unsigned nBytesPerChan
+        unsigned nFramesPerBlock
+
+        void* _TESTDATA 
 
     stream = <_Stream*> voidstream
-    printf('Thread started\n')
+    ninputchannels_forwarded = stream.ninputchannels_forwarded
+    noutputchannels_forwarded = stream.noutputchannels_forwarded
+    nBytesPerChan = stream.nBytesPerChan
+    nFramesPerBlock = stream.nFramesPerBlock
 
     with gil:
         npy_format = _formats_rtkey[stream.sampleformat][2]
         callback = <object> stream.pyCallback
+        print(f'noutputchannels_forwarded: {noutputchannels_forwarded}')
+        print(f'ninputchannels_forwarded: {ninputchannels_forwarded}')
+        print(f'nBytesPerChan: {nBytesPerChan}')
+        print(f'nFramesPerBlock: {nFramesPerBlock}')
+
     while True:
         if stream.stopThread.load() == True:
             printf('Stopping thread...\n')
             return
 
         if stream.inputQueue:
+            # fprintf(stderr, "Waiting on input queue\n")
             inputbuffer = stream.inputQueue.dequeue()
-            # if inputbuffer == NULL:
+            if not inputbuffer:
-            #     continue
+                printf('Stopping thread...\n')
+                return
 
         if stream.outputQueue:
-            outputbuffer = malloc(stream.outputbuffersize)
+            # fprintf(stderr, 'Allocating output buffer...\n')
+            outputbuffer = malloc(nBytesPerChan*noutputchannels_forwarded)
+            # memset(outputbuffer, 0, nBytesPerChan*noutputchannels_forwarded)
 
         with gil:
             
             # Obtain stream information
             npy_input = None
             npy_output = None
-
+            
-            if stream.inputQueue and inputbuffer: 
+            if stream.inputQueue: 
+                # assert(inputbuffer)
                 try:
-                    npy_input = fromBufferToNPYNoCopy(
+                    # print(f'========ninputchannels_forwarded: {ninputchannels_forwarded}')
-                            npy_format,
+                    # print(f'========nFramesPerBlock: {nFramesPerBlock}')
-                            inputbuffer,
+                    # print(f'========npy_format: {npy_format}')
-                            stream.inputParams.nChannels,
+
-                            stream.nFrames)
+                    npy_input = <object> data_to_ndarray(
+                        inputbuffer,
+                        nFramesPerBlock,
+                        ninputchannels_forwarded,
+                        npy_format,
+                        True, # Do transfer ownership
+                        True) # F-contiguous is True: data is Fortran-cont.
+                    # fprintf(stderr, "Copying array...\n")
+                    # fprintf(stderr, "End Copying array...\n")
 
                 except Exception as e:
                     print('exception in cython callback for audio input: ', str(e))
                     return
 
             if stream.outputQueue: 
+                # fprintf(stderr, 'Copying output buffer to Numpy...\n')
                 try:
-                    assert outputbuffer != NULL
+                    npy_output = <object> data_to_ndarray(
-                    npy_output = fromBufferToNPYNoCopy(
+                        outputbuffer,
-                            npy_format,
+                        nFramesPerBlock,
-                            outputbuffer,
+                        noutputchannels_forwarded,
-                            stream.outputParams.nChannels,
+                        npy_format,
-                            stream.nFrames)
+                        False, # Do not transfer ownership
+                        True) # F-contiguous
 
                 except Exception as e:
                     print('exception in Cython callback for audio output: ', str(e))
                     return
+
             try:
+                # fprintf(stderr, "Python callback...\n")
                 rval = callback(npy_input,
                                 npy_output,
-                                stream.nFrames,
+                                nFramesPerBlock,
                                )
+                # fprintf(stderr, "Return from Python callback...\n")
 
             except Exception as e:
                 print('Exception in Cython callback: ', str(e))
                 return
 
-        if stream.outputQueue:
+        if stream.outputQueue: 
+            # fprintf(stderr, 'Enqueuing output buffer...\n')
+
             stream.outputQueue.enqueue(outputbuffer)
             if not stream.inputQueue:
+                # fprintf(stderr, 'No input queue!\n')
                 while stream.outputQueue.size() > 10 and not stream.stopThread.load():
                     # printf('Sleeping...\n')
                     # No input queue to wait on, so we relax a bit here.
@@ -334,30 +477,50 @@ cdef void audioCallbackPythonThreadFunction(void* voidstream) nogil:
         # Outputbuffer is free'ed by the audiothread, so should not be touched
         # here.
         outputbuffer = NULL
+
         # Inputbuffer memory is owned by Numpy, so should not be free'ed
         inputbuffer = NULL
 
 
 cdef void errorCallback(RtAudioError.Type _type,const string& errortxt) nogil:
-    printf('RtAudio error callback called: ')
+    fprintf(stderr, 'RtAudio error callback called: ')
-    printf(errortxt.c_str())
+    fprintf(stderr, errortxt.c_str())
-    printf('\n')
+    fprintf(stderr, '\n')
 
 
 cdef class RtAudio:
     cdef:
-        cppRtAudio _rtaudio
+        cppRtAudio* _rtaudio
         _Stream* _stream
+        int api
 
-    def __cinit__(self):
+    def __cinit__(self, unsigned int iapi):
+        cdef:
+            cppRtAudio.Api api = <cppRtAudio.Api> iapi
+        self._rtaudio = new cppRtAudio(api)
         self._stream = NULL
         self._rtaudio.showWarnings(True)
+        self.api = api
 
     def __dealloc__(self):
         if self._stream is not NULL:
-            fprintf(stderr, 'Force closing stream...')
+            # fprintf(stderr, 'Force closing stream...')
             self._rtaudio.closeStream()
             self.cleanupStream(self._stream)
+        del self._rtaudio
+
+    @staticmethod
+    def getApi():
+        cdef:
+            vector[cppRtAudio.Api] apis
+        cppRtAudio.getCompiledApi(apis)
+        apidict = {}
+        for api in apis:
+            apidict[<int> api] = {
+                'displayname': cppRtAudio.getApiDisplayName(api).decode('utf-8'),
+                 'name': cppRtAudio.getApiName(api).decode('utf-8')
+            }
+        return apidict
 
     cpdef unsigned int getDeviceCount(self):
         return self._rtaudio.getDeviceCount()
@@ -372,14 +535,17 @@ cdef class RtAudio:
         """
         Return device information of the current device
         """
-        cdef cppRtAudio.DeviceInfo devinfo = self._rtaudio.getDeviceInfo(device)
         sampleformats = []
+        cdef:
+            cppRtAudio.DeviceInfo devinfo
+        devinfo = self._rtaudio.getDeviceInfo(device)
         nf = devinfo.nativeFormats
         for format_ in [ RTAUDIO_SINT8, RTAUDIO_SINT16, RTAUDIO_SINT24,
                 RTAUDIO_SINT32, RTAUDIO_FLOAT32, RTAUDIO_FLOAT64]:
             if nf & format_:
                 sampleformats.append(_formats_rtkey[format_][0])
         return DeviceInfo(
+                api = self.api,
                 index = device,
                 probed = devinfo.probed,
                 name = devinfo.name.decode('utf-8'),
@@ -391,109 +557,204 @@ cdef class RtAudio:
                 prefsamplerate = devinfo.preferredSampleRate)
 
     @cython.nonecheck(True)
-    def openStream(self,object outputParams,
+    def openStream(self,
-                        object inputParams,
+                   avstream
-                        str sampleformat,
+                  ):
-                        unsigned int sampleRate,
-                        unsigned int bufferFrames,
-                        object pyCallback,
-                        object options = None,
-                        object pyErrorCallback = None):
         """
         Opening a stream with specified parameters
 
         Args:
-            outputParams: dictionary of stream outputParameters, set to None
+            avstream: AvStream instance
-            if no outputPararms are specified
-            inputParams: dictionary of stream inputParameters, set to None
-            if no inputPararms are specified
-            sampleRate: desired sample rate.
-            bufferFrames: the amount of frames in a callback buffer
-            callback: callback to call. Note: this callback is called on a
-            different thread!
-            options: A dictionary of optional additional stream options
-            errorCallback: client-defined function that will be invoked when an
-            error has occured.
 
         Returns: None
         """
+
         if self._stream is not NULL:
             raise RuntimeError('Stream is already opened.')
 
+        daqconfig = avstream.daqconfig
+        avtype = avstream.avtype
+        device = avstream.device
+
         cdef:
+            bint duplex_mode = daqconfig.duplex_mode
+            bint monitorOutput = daqconfig.monitor_gen
+            unsigned int outputDelayBlocks = daqconfig.outputDelayBlocks
             cppRtAudio.StreamParameters *rtOutputParams_ptr = NULL
             cppRtAudio.StreamParameters *rtInputParams_ptr = NULL
             cppRtAudio.StreamOptions streamoptions
-            size_t bytespersample
+            size_t sw
-            unsigned int bufferFrames_local
+            unsigned int nFramesPerBlock = int(daqconfig.nFramesPerBlock)
+            int firstinputchannel, firstoutputchannel
+            int lastinputchannel, lastoutputchannel
+            unsigned int ninputchannels_forwarded=0
+            unsigned int ninputchannels_rtaudio=0
+            unsigned int noutputchannels_rtaudio=0
+            unsigned int noutputchannels_forwarded=0
+            unsigned int samplerate
+            int i
+            bint input_stream=False, output_stream=False
+            bool* inputChannelsEnabled
+            bool* outputChannelsEnabled
 
-        streamoptions.flags = RTAUDIO_HOG_DEVICE
+        if daqconfig.nFramesPerBlock > 8192 or daqconfig.nFramesPerBlock < 512:
-        streamoptions.numberOfBuffers = 4
+            raise ValueError('Invalid number of nFramesPerBlock')
-        bufferFrames_local = bufferFrames
 
-        self._stream = <_Stream*> malloc(sizeof(_Stream))
+        if daqconfig.outputDelayBlocks < 0 or daqconfig.outputDelayBlocks > 10:
-        if self._stream == NULL:
+            raise ValueError('Invalid number of outputDelayBlocks')
-            raise MemoryError()
-
-        self._stream.pyCallback = <PyObject*> pyCallback
-        Py_INCREF(pyCallback)
-        self._stream.sampleformat = _formats_strkey[sampleformat][0]
-        self._stream.inputQueue = NULL
-        self._stream.outputQueue = NULL
-        self._stream.outputbuffersize = 0
-        self._stream.inputbuffersize = 0
-        self._stream.stopThread.store(False)
-        self._stream.thread = NULL
-
-        bytespersample = get_sampwidth_from_format_string(sampleformat)
-
-        if outputParams is not None:
-            rtOutputParams_ptr = &self._stream.outputParams
-            rtOutputParams_ptr.deviceId = outputParams['deviceid']
-            rtOutputParams_ptr.nChannels = outputParams['nchannels']
-            rtOutputParams_ptr.firstChannel = outputParams['firstchannel']
-            self._stream.outputQueue = new SafeQueue[void*]()
-
-        if inputParams is not None:
-            rtInputParams_ptr = &self._stream.inputParams
-            rtInputParams_ptr.deviceId = inputParams['deviceid']
-            rtInputParams_ptr.nChannels = inputParams['nchannels']
-            rtInputParams_ptr.firstChannel = inputParams['firstchannel']
-            self._stream.inputQueue = new SafeQueue[void*]()
 
         try:
+
+            # Determine sample rate and sample format, determine whether we are an
+            # input or an output stream, or both
+            if avtype == AvType.audio_input or duplex_mode:
+                # Here, we override the sample format in case of duplex mode.
+                sampleformat = daqconfig.en_input_sample_format
+                samplerate = int(daqconfig.en_input_rate)
+                input_stream = True
+                if duplex_mode:
+                    output_stream = True
+            else:
+                sampleformat = daqconfig.en_output_sample_format
+                samplerate = int(daqconfig.en_output_rate)
+                output_stream = True
+
+            print(f'Is input stream: {input_stream}')
+            print(f'Is output stream: {output_stream}')
+
+            sw = get_sampwidth_from_format_string(sampleformat)
+            print(f'samplewidth: {sw}')
+
+            # All set, allocate the stream!
+            self._stream = <_Stream*> malloc(sizeof(_Stream))
+            if self._stream == NULL:
+                raise MemoryError('Could not allocate stream: memory error.')
+
+            self._stream.pyCallback = <PyObject*> avstream._audioCallback
+            # Increase reference count to the callback
+            Py_INCREF(<object> avstream._audioCallback)
+
+            self._stream.sampleformat = _formats_strkey[sampleformat][0]
+            self._stream.stopThread.store(False)
+            self._stream.inputQueue = NULL
+            self._stream.outputQueue = NULL
+            self._stream.outputDelayQueue = NULL
+
+            self._stream.thread = NULL
+
+            self._stream.outputDelayBlocks = outputDelayBlocks
+            self._stream.ninputchannels_forwarded = 0
+            self._stream.noutputchannels_forwarded = 0
+            self._stream.inputChannelsEnabled = NULL
+            self._stream.outputChannelsEnabled = NULL
+
+            # Create channel maps for input channels, set RtAudio input stream
+            # parameters
+            if input_stream:
+                firstinputchannel = daqconfig.firstEnabledInputChannelNumber()
+                lastinputchannel = daqconfig.lastEnabledInputChannelNumber()
+                ninputchannels_rtaudio = lastinputchannel-firstinputchannel+1
+
+                # print(firstinputchannel)
+                # print(lastinputchannel)
+                # print(ninputchannels_rtaudio)
+
+                if lastinputchannel < 0 or ninputchannels_rtaudio < 1:
+                    raise ValueError('Not enough input channels selected')
+                input_ch = daqconfig.input_channel_configs 
+
+                inputChannelsEnabled = <bool*> malloc(sizeof(bool)*ninputchannels_rtaudio)
+                self._stream.inputChannelsEnabled = inputChannelsEnabled
+
+                for i in range(firstinputchannel, lastinputchannel+1):
+                    ch_en = input_ch[i].channel_enabled
+                    if ch_en:
+                        ninputchannels_forwarded += 1
+                    inputChannelsEnabled[i] = ch_en
+
+                rtInputParams_ptr = &self._stream.inputParams
+                rtInputParams_ptr.deviceId = device.index
+                rtInputParams_ptr.nChannels = ninputchannels_rtaudio
+                rtInputParams_ptr.firstChannel = firstinputchannel
+
+                self._stream.inputQueue = new SafeQueue[void*]()
+                self._stream.ninputchannels_forwarded = ninputchannels_forwarded
+
+
+            # Create channel maps for output channels
+            if output_stream:
+                firstoutputchannel = daqconfig.firstEnabledOutputChannelNumber()
+                lastoutputchannel = daqconfig.lastEnabledOutputChannelNumber()
+                noutputchannels_rtaudio = lastoutputchannel-firstoutputchannel+1
+
+                # print(firstoutputchannel)
+                # print(lastoutputchannel)
+                # print(noutputchannels_rtaudio)
+
+                if lastoutputchannel < 0 or noutputchannels_rtaudio < 1:
+                    raise ValueError('Not enough output channels selected')
+                output_ch = daqconfig.output_channel_configs 
+
+                outputChannelsEnabled = <bool*> malloc(sizeof(bool)*noutputchannels_rtaudio)
+                self._stream.outputChannelsEnabled = outputChannelsEnabled
+                for i in range(firstoutputchannel, lastoutputchannel+1):
+                    ch_en = output_ch[i].channel_enabled
+                    if ch_en:
+                        noutputchannels_forwarded += 1
+                    outputChannelsEnabled[i] = ch_en
+
+                rtOutputParams_ptr = &self._stream.outputParams
+                rtOutputParams_ptr.deviceId = device.index
+                rtOutputParams_ptr.nChannels = noutputchannels_rtaudio
+                rtOutputParams_ptr.firstChannel = firstoutputchannel
+
+                self._stream.outputQueue = new SafeQueue[void*]()
+                self._stream.noutputchannels_forwarded = noutputchannels_forwarded
+
+                if monitorOutput and duplex_mode:
+                    self._stream.outputDelayQueue = new SafeQueue[void*]()
+                    self._stream.ninputchannels_forwarded += noutputchannels_forwarded
+
+
+            streamoptions.flags = RTAUDIO_HOG_DEVICE
+            streamoptions.flags |= RTAUDIO_NONINTERLEAVED
+            streamoptions.numberOfBuffers = 4
+            streamoptions.streamName = "LASP Audio stream".encode('utf-8')
+            streamoptions.priority = 1
+
             self._rtaudio.openStream(rtOutputParams_ptr,
                                      rtInputParams_ptr,
                                      _formats_strkey[sampleformat][0],
-                                     sampleRate,
+                                     samplerate,
-                                     &bufferFrames_local,
+                                     &nFramesPerBlock,
                                      audioCallback,
                                      <void*> self._stream,
                                      &streamoptions, # Stream options
                                      errorCallback  # Error callback
                                      ) 
-            self._stream.nFrames = bufferFrames_local
+
+            self._stream.nBytesPerChan = nFramesPerBlock*sw
+            self._stream.nFramesPerBlock = nFramesPerBlock
 
         except Exception as e:
-            print('Exception occured in stream opening: ', str(e))
+            print('Exception occured in stream opening: ', e)
             self.cleanupStream(self._stream)
             self._stream = NULL
-            raise
+            raise e
-
-        if inputParams is not None:
-            self._stream.inputbuffersize = bufferFrames_local*bytespersample*inputParams['nchannels']
-        if outputParams is not None:
-            self._stream.outputbuffersize = bufferFrames_local*bytespersample*outputParams['nchannels']
 
         with nogil:
             self._stream.thread = new CPPThread[void*, void (*)(void*)](audioCallbackPythonThreadFunction,
                                                 <void*> self._stream)
+            # Allow it to start
             CPPsleep(500)
+            pass
 
-        return bufferFrames_local
+        return nFramesPerBlock
 
     cdef cleanupStream(self, _Stream* stream):
         # printf('Entrance function cleanupStream...\n')
+        cdef:
+            void* ptr
         if stream == NULL:
             return
 
@@ -509,17 +770,32 @@ cdef class RtAudio:
                 stream.thread.join()
                 # printf('Thread joined!\n')
                 del stream.thread
+                stream.thread = NULL
+
+            if stream.inputChannelsEnabled:
+                free(stream.inputChannelsEnabled)
+            if stream.outputChannelsEnabled:
+                free(stream.outputChannelsEnabled)
 
             if stream.outputQueue:
+                while not stream.outputQueue.empty():
+                    free(stream.outputQueue.dequeue())
                 del stream.outputQueue
             if stream.inputQueue:
+                while not stream.inputQueue.empty():
+                    free(stream.inputQueue.dequeue())
                 del stream.inputQueue
+            if stream.outputDelayQueue:
+                while not stream.outputDelayQueue.empty():
+                    free(stream.outputDelayQueue.dequeue())
+                del stream.outputDelayQueue
+            fprintf(stderr, "End cleanup stream queues...\n")
 
         if stream.pyCallback:
             Py_DECREF(<object> stream.pyCallback)
-
+            stream.pyCallback = NULL
+        # fprintf(stderr, "End cleanup callback...\n")
         free(stream)
-        # printf('Cleanup of stream is done\n')
 
     def startStream(self):
         self._rtaudio.startStream()

lasp/lasp_avstream.py

@@ -11,20 +11,13 @@ import numpy as np
 import time
 from .device import (RtAudio, DeviceInfo, DAQConfiguration,
                      get_numpy_dtype_from_format_string,
-                     get_sampwidth_from_format_string)
+                     get_sampwidth_from_format_string, AvType)
 
-__all__ = ['AvType', 'AvStream']
+__all__ = ['AvStream']
 
 video_x, video_y = 640, 480
 
 
-class AvType:
-    """Specificying the type of data, for adding and removing callbacks from
-    the stream."""
-    audio_input = 1
-    audio_output = 2
-    video = 4
-
 
 class AvStream:
     """Audio and video data stream, to which callbacks can be added for
@@ -49,59 +42,28 @@ class AvStream:
         """
 
         self.daqconfig = daqconfig
-        self._device = device
+        self.device = device
         self.avtype = avtype
         self.duplex_mode = daqconfig.duplex_mode
-        self.monitor_gen = daqconfig.monitor_gen
 
-        # Determine highest input channel number
+        self.output_channel_names = [ch.channel_name for ch in daqconfig.getEnabledOutputChannels()]
-        channelconfigs = daqconfig.input_channel_configs
-        firstchannel = daqconfig.firstEnabledInputChannelNumber()
-        if firstchannel < 0:
-            raise ValueError('No input channels enabled')
-
-        self.channel_names = []
-        self.sensitivity = self.daqconfig.getEnabledChannelSensitivities()
 
         if daqconfig.monitor_gen:
-            assert self.duplex_mode
+            self.input_channel_names = self.output_channel_names
-            self.channel_names.append('Generated signal')
+            self.input_sensitivity = daqconfig.getEnabledOutputChannelSensitivities()
-            self.sensitivity = np.concatenate([np.array([1.]),
+        else:
-                                               self.sensitivity])
+            self.input_channel_names = []
+            self.input_sensitivity = []
+
+        self.input_sensitivity += daqconfig.getEnabledInputChannelSensitivities()
+        self.input_sensitivity = np.asarray(self.input_sensitivity)
 
 
-        rtaudio_inputparams = None
+        self.input_channel_names += [
-        rtaudio_outputparams = None
+            ch.channel_name for ch in daqconfig.getEnabledInputChannels()]
 
-        self.nframes_per_block = 1024
+        self.input_samplerate = float(daqconfig.en_input_rate)
-
+        self.output_samplerate = float(daqconfig.en_output_rate)
-        if self.duplex_mode or avtype == AvType.audio_output:
-            rtaudio_outputparams = {'deviceid': device.index,
-                                    # TODO: Add option to specify the number of output channels to use
-                                    'nchannels': 1,  # device.outputchannels,
-                                    'firstchannel': 0}
-            self.sampleformat = daqconfig.en_output_sample_format
-            self.samplerate = int(daqconfig.en_output_rate)
-
-        if avtype == AvType.audio_input or self.duplex_mode:
-            for i, channelconfig in enumerate(channelconfigs):
-                if channelconfig.channel_enabled:
-                    self.nchannels = i+1
-                    self.channel_names.append(channelconfig.channel_name)
-            rtaudio_inputparams = {'deviceid': device.index,
-                                   'nchannels': self.nchannels,
-                                   'firstchannel': firstchannel}
-
-            # Here, we override the sample format in case of duplex mode.
-            self.sampleformat = daqconfig.en_input_sample_format
-            self.samplerate = int(daqconfig.en_input_rate)
-
-
-        # Fill in numpy data type, and sample width
-        self.numpy_dtype = get_numpy_dtype_from_format_string(
-            self.sampleformat)
-        self.sampwidth = get_sampwidth_from_format_string(
-            self.sampleformat)
 
         # Counters for the number of frames that have been coming in
         self._aframectr = Atomic(0)
@@ -125,19 +87,19 @@ class AvStream:
 
         # Possible, but long not tested: store video
         self._videothread = None
+        self._rtaudio = RtAudio(daqconfig.api)
+        self.blocksize = self._rtaudio.openStream(self)
 
-        try:
+        # Fill in numpy data type, and sample width
-            self._rtaudio = RtAudio()
+        self.input_numpy_dtype = get_numpy_dtype_from_format_string(
-            self.blocksize = self._rtaudio.openStream(
+            daqconfig.en_input_sample_format)
-                rtaudio_outputparams,  # Outputparams
+        self.output_numpy_dtype = get_numpy_dtype_from_format_string(
-                rtaudio_inputparams,   # Inputparams
+            daqconfig.en_output_sample_format)
-                self.sampleformat,     # Sampleformat
-                self.samplerate,
-                self.nframes_per_block,  # Buffer size in frames
-                self._audioCallback)
 
-        except Exception as e:
+        self.input_sampwidth = get_sampwidth_from_format_string(
-            raise RuntimeError(f'Could not initialize DAQ device: {str(e)}')
+            daqconfig.en_input_sample_format)
+        self.output_sampwidth = get_sampwidth_from_format_string(
+            daqconfig.en_output_sample_format)
 
     def close(self):
         self._rtaudio.closeStream()
@@ -212,29 +174,30 @@ class AvStream:
         """This is called (from a separate thread) for each audio block."""
         self._aframectr += nframes
         with self._callbacklock:
-
             # Count the number of output callbacks. If no output callbacks are
             # present, and there should be output callbacks, we explicitly set
             # the output buffer to zero
             noutput_cb = len(self._callbacks[AvType.audio_output])
             shouldhaveoutput = (self.avtype == AvType.audio_output or
-                                self.duplex_mode)
+                                self.daqconfig.duplex_mode)
-            if noutput_cb == 0 and shouldhaveoutput:
+            if noutput_cb == 0 and shouldhaveoutput and outdata is not None:
                 outdata[:, :] = 0
 
             # Loop over callbacks
-            for cb in self._callbacks[AvType.audio_output]:
+            if outdata is not None:
-                try:
+                for cb in self._callbacks[AvType.audio_output]:
-                    cb(indata, outdata, self._aframectr())
+                    try:
-                except Exception as e:
+                        cb(indata, outdata, self._aframectr())
-                    print(e)
+                    except Exception as e:
-                    return 1
+                        print(e)
-            for cb in self._callbacks[AvType.audio_input]:
+                        return 2
-                try:
+            if indata is not None:
-                    cb(indata, outdata, self._aframectr())
+                for cb in self._callbacks[AvType.audio_input]:
-                except Exception as e:
+                    try:
-                    print(e)
+                        cb(indata, outdata, self._aframectr())
-                    return 1
+                    except Exception as e:
+                        print(e)
+                        return 1
 
         return 0 if self._running else 1
 

lasp/lasp_common.py

@@ -9,6 +9,8 @@ import numpy as np
 
 from .wrappers import Window as wWindow
 from collections import namedtuple
+from dataclasses import dataclass
+from dataclasses_json import dataclass_json
 
 """
 Common definitions used throughout the code.
@@ -39,9 +41,24 @@ U_REF = 5e-8 # 50 nano meter / s
 # hence this is the reference level as specified below.
 dBFS_REF = 0.5*2**0.5 # Which level would be -3.01 dBFS
 
-Qty = namedtuple('Qty', 'name unit_name unit_symb level_unit level_ref_name level_ref_value')
+@dataclass_json
+@dataclass
+class Qty:
+    name: str
+    unit_name: str
+    unit_symb: str
+    level_unit: str
+    level_ref_name: str
+    level_ref_value: str
 
 class SIQtys:
+    N = Qty(name='Number',
+            unit_name='No unit / full scale',
+            unit_symb=('-'),
+            level_unit=('dBFS',),
+            level_ref_name=('Full scale sine wave',),
+            level_ref_value=(dBFS_REF,)
+           )
     AP = Qty(name='Acoustic Pressure',
              unit_name='Pascal',
              unit_symb=('Pa', 'muPa'),
@@ -51,21 +68,14 @@ class SIQtys:
             )
 
     V = Qty(name='Voltage',
-             unit_name='volt',
+             unit_name='Volt',
              unit_symb='V',
              level_unit=('dBV',), # dBV
              level_ref_name=('1V',),
              level_ref_value=(1.0,),
            )
-    N = Qty(name='No unit',
-            unit_name='',
-            unit_symb='[-]',
-            level_unit=('dBFS',),
-            level_ref_name=('Full scale sine wave',),
-            level_ref_value=(dBFS_REF,)
-           )
     types = (AP, V, N)
-    default = AP
+    default = N
     default_index = 0
 
     @staticmethod
@@ -78,7 +88,7 @@ class SIQtys:
         """
         cb.clear()
         for ty in SIQtys.types:
-            cb.addItem(f'{ty.name} [{ty.unit_symb[0]}')
+            cb.addItem(f'{ty.unit_name}')
         cb.setCurrentIndex(SIQtys.default_index)
 
     @staticmethod

lasp/lasp_measurement.py

@@ -9,6 +9,8 @@ The ASCEE hdf5 measurement file format contains the following fields:
 
 - Attributes:
 
+'version': If not given, version 1 is assumed. For version 1, measurement data
+is assumed to be acoustic data.
 'samplerate': The audio data sample rate in Hz.
 'nchannels': The number of audio channels in the file
 'sensitivity': (Optionally) the stored sensitivity of the record channels.
@@ -42,6 +44,7 @@ from .lasp_config import LASP_NUMPY_FLOAT_TYPE
 from scipy.io import wavfile
 import os
 import time
+import wave
 
 
 class BlockIter:

lasp/lasp_record.py

@@ -78,10 +78,6 @@ class Recording:
         stream = self._stream
         f = self._f
         nchannels = stream.nchannels
-        if stream.monitor_gen:
-            nchannels += 1
-
-        self.monitor_gen = stream.monitor_gen
 
         self._ad = f.create_dataset('audio',
                                     (1, stream.blocksize, nchannels),
@@ -173,11 +169,7 @@ class Recording:
             return
 
         self._ad.resize(self._ablockno()+1, axis=0)
-        if self.monitor_gen:
+        self._ad[self._ablockno(), :, :] = indata
-            self._ad[self._ablockno(), :, 0] = outdata[:, 0]
-            self._ad[self._ablockno(), :, 1:] = indata
-        else:
-            self._ad[self._ablockno(), :, :] = indata
         self._ablockno += 1
 
     def _vCallback(self, frame, framectr):

setup.py

@@ -32,6 +32,7 @@ setup(
       author_email="j.a.dejong@ascee.nl",
       install_requires=['matplotlib>=1.0',
                         'scipy>=1.0', 'numpy>=1.0', 'h5py',
+                        'dataclasses_json',
                         ],
       license='MIT',
       description="Library for Acoustic Signal Processing",