path: root/libs/appleutility/CoreAudio/PublicUtility/CAStreamBasicDescription.cpp

/*
     File: CAStreamBasicDescription.cpp
 Abstract: CAStreamBasicDescription.h
  Version: 1.1
 
 Disclaimer: IMPORTANT:  This Apple software is supplied to you by Apple
 Inc. ("Apple") in consideration of your agreement to the following
 terms, and your use, installation, modification or redistribution of
 this Apple software constitutes acceptance of these terms.  If you do
 not agree with these terms, please do not use, install, modify or
 redistribute this Apple software.
 
 In consideration of your agreement to abide by the following terms, and
 subject to these terms, Apple grants you a personal, non-exclusive
 license, under Apple's copyrights in this original Apple software (the
 "Apple Software"), to use, reproduce, modify and redistribute the Apple
 Software, with or without modifications, in source and/or binary forms;
 provided that if you redistribute the Apple Software in its entirety and
 without modifications, you must retain this notice and the following
 text and disclaimers in all such redistributions of the Apple Software.
 Neither the name, trademarks, service marks or logos of Apple Inc. may
 be used to endorse or promote products derived from the Apple Software
 without specific prior written permission from Apple.  Except as
 expressly stated in this notice, no other rights or licenses, express or
 implied, are granted by Apple herein, including but not limited to any
 patent rights that may be infringed by your derivative works or by other
 works in which the Apple Software may be incorporated.
 
 The Apple Software is provided by Apple on an "AS IS" basis.  APPLE
 MAKES NO WARRANTIES, EXPRESS OR IMPLIED, INCLUDING WITHOUT LIMITATION
 THE IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY AND FITNESS
 FOR A PARTICULAR PURPOSE, REGARDING THE APPLE SOFTWARE OR ITS USE AND
 OPERATION ALONE OR IN COMBINATION WITH YOUR PRODUCTS.
 
 IN NO EVENT SHALL APPLE BE LIABLE FOR ANY SPECIAL, INDIRECT, INCIDENTAL
 OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 INTERRUPTION) ARISING IN ANY WAY OUT OF THE USE, REPRODUCTION,
 MODIFICATION AND/OR DISTRIBUTION OF THE APPLE SOFTWARE, HOWEVER CAUSED
 AND WHETHER UNDER THEORY OF CONTRACT, TORT (INCLUDING NEGLIGENCE),
 STRICT LIABILITY OR OTHERWISE, EVEN IF APPLE HAS BEEN ADVISED OF THE
 POSSIBILITY OF SUCH DAMAGE.
 
 Copyright (C) 2014 Apple Inc. All Rights Reserved.
 
*/
#include "CAStreamBasicDescription.h"
#include "CAMath.h"

#if !defined(__COREAUDIO_USE_FLAT_INCLUDES__)
	#include <CoreFoundation/CFByteOrder.h>
#else
	#include <CFByteOrder.h>
#endif

#pragma mark	This file needs to compile on earlier versions of the OS, so please keep that in mind when editing it

char *CAStringForOSType (OSType t, char *writeLocation, size_t bufsize)
{
	if (bufsize > 0) {
		char *p = writeLocation, *pend = writeLocation + bufsize;
		union { UInt32 i; unsigned char str[4]; } u;
		unsigned char *q = u.str;
		u.i = CFSwapInt32HostToBig(t);

		bool hasNonPrint = false;
		for (int i = 0; i < 4; ++i) {
			if (!(isprint(*q) && *q != '\\')) {
				hasNonPrint = true;
				break;
			}
            q++;
		}
        q = u.str;
		
		if (hasNonPrint)
			p += snprintf (p, pend - p, "0x");
		else if (p < pend)
			*p++ = '\'';
			
		for (int i = 0; i < 4 && p < pend; ++i) {
			if (hasNonPrint) {
				p += snprintf(p, pend - p, "%02X", *q++);
			} else {
				*p++ = *q++;
			}
		}
		if (!hasNonPrint && p < pend)
			*p++ = '\'';
		if (p >= pend) p -= 1;
		*p = '\0';
	}
	return writeLocation;
}


const AudioStreamBasicDescription	CAStreamBasicDescription::sEmpty = { 0.0, 0, 0, 0, 0, 0, 0, 0, 0 };

CAStreamBasicDescription::CAStreamBasicDescription() 
{ 
	memset (this, 0, sizeof(AudioStreamBasicDescription)); 
}
	
CAStreamBasicDescription::CAStreamBasicDescription(const AudioStreamBasicDescription &desc)
{
	SetFrom(desc);
}


CAStreamBasicDescription::CAStreamBasicDescription(double inSampleRate,		UInt32 inFormatID,
									UInt32 inBytesPerPacket,	UInt32 inFramesPerPacket,
									UInt32 inBytesPerFrame,		UInt32 inChannelsPerFrame,
									UInt32 inBitsPerChannel,	UInt32 inFormatFlags)
{
	mSampleRate = inSampleRate;
	mFormatID = inFormatID;
	mBytesPerPacket = inBytesPerPacket;
	mFramesPerPacket = inFramesPerPacket;
	mBytesPerFrame = inBytesPerFrame;
	mChannelsPerFrame = inChannelsPerFrame;
	mBitsPerChannel = inBitsPerChannel;
	mFormatFlags = inFormatFlags;
	mReserved = 0;
}

char *CAStreamBasicDescription::AsString(char *buf, size_t _bufsize, bool brief /*=false*/) const
{
	int bufsize = (int)_bufsize;	// must be signed to protect against overflow
	char *theBuffer = buf;
	int nc;
	char formatID[24];
	CAStringForOSType(mFormatID, formatID, sizeof(formatID));
	if (brief) {
		CommonPCMFormat com;
		bool interleaved;
		if (IdentifyCommonPCMFormat(com, &interleaved) && com != kPCMFormatOther) {
			const char *desc;
			switch (com) {
			case kPCMFormatInt16:
				desc = "Int16";
				break;
			case kPCMFormatFixed824:
				desc = "Int8.24";
				break;
			case kPCMFormatFloat32:
				desc = "Float32";
				break;
			case kPCMFormatFloat64:
				desc = "Float64";
				break;
			default:
				desc = NULL;
				break;
			}
			if (desc) {
				const char *inter ="";
				if (mChannelsPerFrame > 1)
					inter = !interleaved ? ", non-inter" : ", inter";
				snprintf(buf, static_cast<size_t>(bufsize), "%2d ch, %6.0f Hz, %s%s", (int)mChannelsPerFrame, mSampleRate, desc, inter);
				return theBuffer;
			}
		}
		if (mChannelsPerFrame == 0 && mSampleRate == 0.0 && mFormatID == 0) {
			snprintf(buf, static_cast<size_t>(bufsize), "%2d ch, %6.0f Hz", (int)mChannelsPerFrame, mSampleRate);
			return theBuffer;
		}
	}
	
	nc = snprintf(buf, static_cast<size_t>(bufsize), "%2d ch, %6.0f Hz, %s (0x%08X) ", (int)NumberChannels(), mSampleRate, formatID, (int)mFormatFlags);
	buf += nc; if ((bufsize -= nc) <= 0) goto exit;
	if (mFormatID == kAudioFormatLinearPCM) {
		bool isInt = !(mFormatFlags & kLinearPCMFormatFlagIsFloat);
		int wordSize = static_cast<int>(SampleWordSize());
		const char *endian = (wordSize > 1) ? 
			((mFormatFlags & kLinearPCMFormatFlagIsBigEndian) ? " big-endian" : " little-endian" ) : "";
		const char *sign = isInt ? 
			((mFormatFlags & kLinearPCMFormatFlagIsSignedInteger) ? " signed" : " unsigned") : "";
		const char *floatInt = isInt ? "integer" : "float";
		char packed[32];
		if (wordSize > 0 && PackednessIsSignificant()) {
			if (mFormatFlags & kLinearPCMFormatFlagIsPacked)
				snprintf(packed, sizeof(packed), "packed in %d bytes", wordSize);
			else
				snprintf(packed, sizeof(packed), "unpacked in %d bytes", wordSize);
		} else
			packed[0] = '\0';
		const char *align = (wordSize > 0 && AlignmentIsSignificant()) ?
			((mFormatFlags & kLinearPCMFormatFlagIsAlignedHigh) ? " high-aligned" : " low-aligned") : "";
		const char *deinter = (mFormatFlags & kAudioFormatFlagIsNonInterleaved) ? ", deinterleaved" : "";
		const char *commaSpace = (packed[0]!='\0') || (align[0]!='\0') ? ", " : "";
		char bitdepth[20];

		int fracbits = (mFormatFlags & kLinearPCMFormatFlagsSampleFractionMask) >> kLinearPCMFormatFlagsSampleFractionShift;
		if (fracbits > 0)
			snprintf(bitdepth, sizeof(bitdepth), "%d.%d", (int)mBitsPerChannel - fracbits, fracbits);
		else
			snprintf(bitdepth, sizeof(bitdepth), "%d", (int)mBitsPerChannel);
		
		/*nc =*/ snprintf(buf, static_cast<size_t>(bufsize), "%s-bit%s%s %s%s%s%s%s",
			bitdepth, endian, sign, floatInt, 
			commaSpace, packed, align, deinter);
		// buf += nc; if ((bufsize -= nc) <= 0) goto exit;
	} else if (mFormatID == kAudioFormatAppleLossless) {
		int sourceBits = 0;
		switch (mFormatFlags)
		{
			case 1:	//	kAppleLosslessFormatFlag_16BitSourceData
				sourceBits = 16;
				break;
    		case 2:	//	kAppleLosslessFormatFlag_20BitSourceData
    			sourceBits = 20;
    			break;
    		case 3:	//	kAppleLosslessFormatFlag_24BitSourceData
    			sourceBits = 24;
    			break;
    		case 4:	//	kAppleLosslessFormatFlag_32BitSourceData
    			sourceBits = 32;
    			break;
		}
		if (sourceBits)
			nc = snprintf(buf, static_cast<size_t>(bufsize), "from %d-bit source, ", sourceBits);
		else
			nc = snprintf(buf, static_cast<size_t>(bufsize), "from UNKNOWN source bit depth, ");
		buf += nc; if ((bufsize -= nc) <= 0) goto exit;
		/*nc =*/ snprintf(buf, static_cast<size_t>(bufsize), "%d frames/packet", (int)mFramesPerPacket);
		//	buf += nc; if ((bufsize -= nc) <= 0) goto exit;
	}
	else
		/*nc =*/ snprintf(buf, static_cast<size_t>(bufsize), "%d bits/channel, %d bytes/packet, %d frames/packet, %d bytes/frame", 
			(int)mBitsPerChannel, (int)mBytesPerPacket, (int)mFramesPerPacket, (int)mBytesPerFrame);
exit:
	return theBuffer;
}

void	CAStreamBasicDescription::NormalizeLinearPCMFormat(AudioStreamBasicDescription& ioDescription)
{
	//  the only thing that changes is to make mixable linear PCM into the canonical linear PCM format
	if((ioDescription.mFormatID == kAudioFormatLinearPCM) && ((ioDescription.mFormatFlags & kIsNonMixableFlag) == 0))
	{
		//  the canonical linear PCM format
		ioDescription.mFormatFlags = kAudioFormatFlagsCanonical;
		ioDescription.mBytesPerPacket = SizeOf32(AudioSampleType) * ioDescription.mChannelsPerFrame;
		ioDescription.mFramesPerPacket = 1;
		ioDescription.mBytesPerFrame = SizeOf32(AudioSampleType) * ioDescription.mChannelsPerFrame;
		ioDescription.mBitsPerChannel = 8 * SizeOf32(AudioSampleType);
	}
}

void	CAStreamBasicDescription::NormalizeLinearPCMFormat(bool inNativeEndian, AudioStreamBasicDescription& ioDescription)
{
	//  the only thing that changes is to make mixable linear PCM into the canonical linear PCM format
	if((ioDescription.mFormatID == kAudioFormatLinearPCM) && ((ioDescription.mFormatFlags & kIsNonMixableFlag) == 0))
	{
		//  the canonical linear PCM format
		ioDescription.mFormatFlags = kAudioFormatFlagIsFloat | kAudioFormatFlagIsPacked;
		if(inNativeEndian)
		{
#if TARGET_RT_BIG_ENDIAN
			ioDescription.mFormatFlags |= kAudioFormatFlagIsBigEndian;
#endif
		}
		else
		{
#if TARGET_RT_LITTLE_ENDIAN
			ioDescription.mFormatFlags |= kAudioFormatFlagIsBigEndian;
#endif
		}
		ioDescription.mBytesPerPacket = SizeOf32(AudioSampleType) * ioDescription.mChannelsPerFrame;
		ioDescription.mFramesPerPacket = 1;
		ioDescription.mBytesPerFrame = SizeOf32(AudioSampleType) * ioDescription.mChannelsPerFrame;
		ioDescription.mBitsPerChannel = 8 * SizeOf32(AudioSampleType);
	}
}

void	CAStreamBasicDescription::ResetFormat(AudioStreamBasicDescription& ioDescription)
{
	ioDescription.mSampleRate = 0;
	ioDescription.mFormatID = 0;
	ioDescription.mBytesPerPacket = 0;
	ioDescription.mFramesPerPacket = 0;
	ioDescription.mBytesPerFrame = 0;
	ioDescription.mChannelsPerFrame = 0;
	ioDescription.mBitsPerChannel = 0;
	ioDescription.mFormatFlags = 0;
}

void	CAStreamBasicDescription::FillOutFormat(AudioStreamBasicDescription& ioDescription, const AudioStreamBasicDescription& inTemplateDescription)
{
	if(fiszero(ioDescription.mSampleRate))
	{
		ioDescription.mSampleRate = inTemplateDescription.mSampleRate;
	}
	if(ioDescription.mFormatID == 0)
	{
		ioDescription.mFormatID = inTemplateDescription.mFormatID;
	}
	if(ioDescription.mFormatFlags == 0)
	{
		ioDescription.mFormatFlags = inTemplateDescription.mFormatFlags;
	}
	if(ioDescription.mBytesPerPacket == 0)
	{
		ioDescription.mBytesPerPacket = inTemplateDescription.mBytesPerPacket;
	}
	if(ioDescription.mFramesPerPacket == 0)
	{
		ioDescription.mFramesPerPacket = inTemplateDescription.mFramesPerPacket;
	}
	if(ioDescription.mBytesPerFrame == 0)
	{
		ioDescription.mBytesPerFrame = inTemplateDescription.mBytesPerFrame;
	}
	if(ioDescription.mChannelsPerFrame == 0)
	{
		ioDescription.mChannelsPerFrame = inTemplateDescription.mChannelsPerFrame;
	}
	if(ioDescription.mBitsPerChannel == 0)
	{
		ioDescription.mBitsPerChannel = inTemplateDescription.mBitsPerChannel;
	}
}

void	CAStreamBasicDescription::GetSimpleName(const AudioStreamBasicDescription& inDescription, char* outName, UInt32 inMaxNameLength, bool inAbbreviate, bool inIncludeSampleRate)
{
	if(inIncludeSampleRate)
	{
		int theCharactersWritten = snprintf(outName, inMaxNameLength, "%.0f ", inDescription.mSampleRate);
		outName += theCharactersWritten;
		inMaxNameLength -= static_cast<UInt32>(theCharactersWritten);
	}
	
	switch(inDescription.mFormatID)
	{
		case kAudioFormatLinearPCM:
			{
				const char* theEndianString = NULL;
				if((inDescription.mFormatFlags & kAudioFormatFlagIsBigEndian) != 0)
				{
					#if	TARGET_RT_LITTLE_ENDIAN
						theEndianString = "Big Endian";
					#endif
				}
				else
				{
					#if	TARGET_RT_BIG_ENDIAN
						theEndianString = "Little Endian";
					#endif
				}
				
				const char* theKindString = NULL;
				if((inDescription.mFormatFlags & kAudioFormatFlagIsFloat) != 0)
				{
					theKindString = (inAbbreviate ? "Float" : "Floating Point");
				}
				else if((inDescription.mFormatFlags & kAudioFormatFlagIsSignedInteger) != 0)
				{
					theKindString = (inAbbreviate ? "SInt" : "Signed Integer");
				}
				else
				{
					theKindString = (inAbbreviate ? "UInt" : "Unsigned Integer");
				}
				
				const char* thePackingString = NULL;
				if((inDescription.mFormatFlags & kAudioFormatFlagIsPacked) == 0)
				{
					if((inDescription.mFormatFlags & kAudioFormatFlagIsAlignedHigh) != 0)
					{
						thePackingString = "High";
					}
					else
					{
						thePackingString = "Low";
					}
				}
				
				const char* theMixabilityString = NULL;
				if((inDescription.mFormatFlags & kIsNonMixableFlag) == 0)
				{
					theMixabilityString = "Mixable";
				}
				else
				{
					theMixabilityString = "Unmixable";
				}
				
				if(inAbbreviate)
				{
					if(theEndianString != NULL)
					{
						if(thePackingString != NULL)
						{
							snprintf(outName, inMaxNameLength, "%s %d Ch %s %s %s%d/%s%d", theMixabilityString, (int)inDescription.mChannelsPerFrame, theEndianString, thePackingString, theKindString, (int)inDescription.mBitsPerChannel, theKindString, (int)(inDescription.mBytesPerFrame / inDescription.mChannelsPerFrame) * 8);
						}
						else
						{
							snprintf(outName, inMaxNameLength, "%s %d Ch %s %s%d", theMixabilityString, (int)inDescription.mChannelsPerFrame, theEndianString, theKindString, (int)inDescription.mBitsPerChannel);
						}
					}
					else
					{
						if(thePackingString != NULL)
						{
							snprintf(outName, inMaxNameLength, "%s %d Ch %s %s%d/%s%d", theMixabilityString, (int)inDescription.mChannelsPerFrame, thePackingString, theKindString, (int)inDescription.mBitsPerChannel, theKindString, (int)((inDescription.mBytesPerFrame / inDescription.mChannelsPerFrame) * 8));
						}
						else
						{
							snprintf(outName, inMaxNameLength, "%s %d Ch %s%d", theMixabilityString, (int)inDescription.mChannelsPerFrame, theKindString, (int)inDescription.mBitsPerChannel);
						}
					}
				}
				else
				{
					if(theEndianString != NULL)
					{
						if(thePackingString != NULL)
						{
							snprintf(outName, inMaxNameLength, "%s %d Channel %d Bit %s %s Aligned %s in %d Bits", theMixabilityString, (int)inDescription.mChannelsPerFrame, (int)inDescription.mBitsPerChannel, theEndianString, theKindString, thePackingString, (int)(inDescription.mBytesPerFrame / inDescription.mChannelsPerFrame) * 8);
						}
						else
						{
							snprintf(outName, inMaxNameLength, "%s %d Channel %d Bit %s %s", theMixabilityString, (int)inDescription.mChannelsPerFrame, (int)inDescription.mBitsPerChannel, theEndianString, theKindString);
						}
					}
					else
					{
						if(thePackingString != NULL)
						{
							snprintf(outName, inMaxNameLength, "%s %d Channel %d Bit %s Aligned %s in %d Bits", theMixabilityString, (int)inDescription.mChannelsPerFrame, (int)inDescription.mBitsPerChannel, theKindString, thePackingString, (int)(inDescription.mBytesPerFrame / inDescription.mChannelsPerFrame) * 8);
						}
						else
						{
							snprintf(outName, inMaxNameLength, "%s %d Channel %d Bit %s", theMixabilityString, (int)inDescription.mChannelsPerFrame, (int)inDescription.mBitsPerChannel, theKindString);
						}
					}
				}
			}
			break;
		
		case kAudioFormatAC3:
			strlcpy(outName, "AC-3", sizeof(outName));
			break;
		
		case kAudioFormat60958AC3:
			strlcpy(outName, "AC-3 for SPDIF", sizeof(outName));
			break;
		
		default:
			CACopy4CCToCString(outName, inDescription.mFormatID);
			break;
	};
}

#if CoreAudio_Debug
#include "CALogMacros.h"

void	CAStreamBasicDescription::PrintToLog(const AudioStreamBasicDescription& inDesc)
{
	PrintFloat		("  Sample Rate:        ", inDesc.mSampleRate);
	Print4CharCode	("  Format ID:          ", inDesc.mFormatID);
	PrintHex		("  Format Flags:       ", inDesc.mFormatFlags);
	PrintInt		("  Bytes per Packet:   ", inDesc.mBytesPerPacket);
	PrintInt		("  Frames per Packet:  ", inDesc.mFramesPerPacket);
	PrintInt		("  Bytes per Frame:    ", inDesc.mBytesPerFrame);
	PrintInt		("  Channels per Frame: ", inDesc.mChannelsPerFrame);
	PrintInt		("  Bits per Channel:   ", inDesc.mBitsPerChannel);
}
#endif

bool	operator<(const AudioStreamBasicDescription& x, const AudioStreamBasicDescription& y)
{
	bool theAnswer = false;
	bool isDone = false;
	
	//	note that if either side is 0, that field is skipped
	
	//	format ID is the first order sort
	if((!isDone) && ((x.mFormatID != 0) && (y.mFormatID != 0)))
	{
		if(x.mFormatID != y.mFormatID)
		{
			//	formats are sorted numerically except that linear
			//	PCM is always first
			if(x.mFormatID == kAudioFormatLinearPCM)
			{
				theAnswer = true;
			}
			else if(y.mFormatID == kAudioFormatLinearPCM)
			{
				theAnswer = false;
			}
			else
			{
				theAnswer = x.mFormatID < y.mFormatID;
			}
			isDone = true;
		}
	}
	
	
	//  mixable is always better than non-mixable for linear PCM and should be the second order sort item
	if((!isDone) && ((x.mFormatID == kAudioFormatLinearPCM) && (y.mFormatID == kAudioFormatLinearPCM)))
	{
		if(((x.mFormatFlags & kIsNonMixableFlag) == 0) && ((y.mFormatFlags & kIsNonMixableFlag) != 0))
		{
			theAnswer = true;
			isDone = true;
		}
		else if(((x.mFormatFlags & kIsNonMixableFlag) != 0) && ((y.mFormatFlags & kIsNonMixableFlag) == 0))
		{
			theAnswer = false;
			isDone = true;
		}
	}
	
	//	floating point vs integer for linear PCM only
	if((!isDone) && ((x.mFormatID == kAudioFormatLinearPCM) && (y.mFormatID == kAudioFormatLinearPCM)))
	{
		if((x.mFormatFlags & kAudioFormatFlagIsFloat) != (y.mFormatFlags & kAudioFormatFlagIsFloat))
		{
			//	floating point is better than integer
			theAnswer = y.mFormatFlags & kAudioFormatFlagIsFloat;
			isDone = true;
		}
	}
	
	//	bit depth
	if((!isDone) && ((x.mBitsPerChannel != 0) && (y.mBitsPerChannel != 0)))
	{
		if(x.mBitsPerChannel != y.mBitsPerChannel)
		{
			//	deeper bit depths are higher quality
			theAnswer = x.mBitsPerChannel < y.mBitsPerChannel;
			isDone = true;
		}
	}
	
	//	sample rate
	if((!isDone) && fnonzero(x.mSampleRate) && fnonzero(y.mSampleRate))
	{
		if(fnotequal(x.mSampleRate, y.mSampleRate))
		{
			//	higher sample rates are higher quality
			theAnswer = x.mSampleRate < y.mSampleRate;
			isDone = true;
		}
	}
	
	//	number of channels
	if((!isDone) && ((x.mChannelsPerFrame != 0) && (y.mChannelsPerFrame != 0)))
	{
		if(x.mChannelsPerFrame != y.mChannelsPerFrame)
		{
			//	more channels is higher quality
			theAnswer = x.mChannelsPerFrame < y.mChannelsPerFrame;
			//isDone = true;
		}
	}
	
	return theAnswer;
}

void CAStreamBasicDescription::ModifyFormatFlagsForMatching(const AudioStreamBasicDescription& x, const AudioStreamBasicDescription& y, UInt32& xFlags, UInt32& yFlags, bool converterOnly )
{
    // match wildcards
	if (x.mFormatID == 0 || y.mFormatID == 0 || xFlags == 0 || yFlags == 0)
    {
		// Obliterate all flags.
        xFlags = yFlags = 0;
        return;
    }
    
    if (x.mFormatID == kAudioFormatLinearPCM) {
        // knock off the all clear flag
        xFlags = xFlags & ~kAudioFormatFlagsAreAllClear;
        yFlags = yFlags & ~kAudioFormatFlagsAreAllClear;
        
        // if both kAudioFormatFlagIsPacked bits are set, then we don't care about the kAudioFormatFlagIsAlignedHigh bit.
        if (xFlags & yFlags & kAudioFormatFlagIsPacked) {
            xFlags = xFlags & ~static_cast<UInt32>(kAudioFormatFlagIsAlignedHigh);
            yFlags = yFlags & ~static_cast<UInt32>(kAudioFormatFlagIsAlignedHigh);
        }
        
        // if both kAudioFormatFlagIsFloat bits are set, then we don't care about the kAudioFormatFlagIsSignedInteger bit.
        if (xFlags & yFlags & kAudioFormatFlagIsFloat) {
            xFlags = xFlags & ~static_cast<UInt32>(kAudioFormatFlagIsSignedInteger);
            yFlags = yFlags & ~static_cast<UInt32>(kAudioFormatFlagIsSignedInteger);
        }
        
        //	if the bit depth is 8 bits or less and the format is packed, we don't care about endianness
        if((x.mBitsPerChannel <= 8) && ((xFlags & kAudioFormatFlagIsPacked) == kAudioFormatFlagIsPacked))
        {
            xFlags = xFlags & ~static_cast<UInt32>(kAudioFormatFlagIsBigEndian);
        }
        if((y.mBitsPerChannel <= 8) && ((yFlags & kAudioFormatFlagIsPacked) == kAudioFormatFlagIsPacked))
        {
            yFlags = yFlags & ~static_cast<UInt32>(kAudioFormatFlagIsBigEndian);
        }
        
        //	if the number of channels is 1, we don't care about non-interleavedness
        if (x.mChannelsPerFrame == 1 && y.mChannelsPerFrame == 1) {
            xFlags &= ~static_cast<UInt32>(kLinearPCMFormatFlagIsNonInterleaved);
            yFlags &= ~static_cast<UInt32>(kLinearPCMFormatFlagIsNonInterleaved);
        }
        
        if (converterOnly) {
            CAStreamBasicDescription cas_x = CAStreamBasicDescription(x);
            CAStreamBasicDescription cas_y = CAStreamBasicDescription(y);
            if (!cas_x.PackednessIsSignificant() && !cas_y.PackednessIsSignificant()) {
                xFlags &= ~static_cast<UInt32>(kAudioFormatFlagIsPacked);
                yFlags &= ~static_cast<UInt32>(kAudioFormatFlagIsPacked);
            }
            if (!cas_x.AlignmentIsSignificant() && !cas_y.AlignmentIsSignificant()) {
                xFlags &= ~static_cast<UInt32>(kAudioFormatFlagIsAlignedHigh);
                yFlags &= ~static_cast<UInt32>(kAudioFormatFlagIsAlignedHigh);
            }
            // We don't care about whether the streams are mixable in this case
            xFlags &= ~static_cast<UInt32>(kAudioFormatFlagIsNonMixable);
            yFlags &= ~static_cast<UInt32>(kAudioFormatFlagIsNonMixable);
        }
    }
}

static bool MatchFormatFlags(const AudioStreamBasicDescription& x, const AudioStreamBasicDescription& y)
{
	UInt32 xFlags = x.mFormatFlags;
	UInt32 yFlags = y.mFormatFlags;
	
    CAStreamBasicDescription::ModifyFormatFlagsForMatching(x, y, xFlags, yFlags, false);
	return xFlags == yFlags;
}

bool	operator==(const AudioStreamBasicDescription& x, const AudioStreamBasicDescription& y)
{
	//	the semantics for equality are:
	//		1) Values must match exactly -- except for PCM format flags, see above.
	//		2) wildcard's are ignored in the comparison
	
#define MATCH(name) ((x.name) == 0 || (y.name) == 0 || (x.name) == (y.name))
	
	return 
    //	check all but the format flags
    CAStreamBasicDescription::FlagIndependentEquivalence(x, y)
    //	check the format flags
    && MatchFormatFlags(x, y);  
}

bool    CAStreamBasicDescription::FlagIndependentEquivalence(const AudioStreamBasicDescription &x, const AudioStreamBasicDescription &y)
{
    return 
    //	check the sample rate
    (fiszero(x.mSampleRate) || fiszero(y.mSampleRate) || fequal(x.mSampleRate, y.mSampleRate))
    
    //	check the format ids
    && MATCH(mFormatID)
        
    //	check the bytes per packet
    && MATCH(mBytesPerPacket) 
    
    //	check the frames per packet
    && MATCH(mFramesPerPacket) 
    
    //	check the bytes per frame
    && MATCH(mBytesPerFrame) 
    
    //	check the channels per frame
    && MATCH(mChannelsPerFrame) 
    
    //	check the channels per frame
    && MATCH(mBitsPerChannel) ;
}

bool	CAStreamBasicDescription::IsEqual(const AudioStreamBasicDescription &other, bool interpretingWildcards) const
{
	if (interpretingWildcards)
		return *this == other;
	return memcmp(this, &other, offsetof(AudioStreamBasicDescription, mReserved)) == 0;
}

bool    CAStreamBasicDescription::IsFunctionallyEquivalent(const AudioStreamBasicDescription &x, const AudioStreamBasicDescription &y)
{
    UInt32 xFlags = x.mFormatFlags, yFlags = y.mFormatFlags;    
    CAStreamBasicDescription::ModifyFormatFlagsForMatching(x, y, xFlags, yFlags, true);
    
    return
    // check all but the format flags
    CAStreamBasicDescription::FlagIndependentEquivalence(x, y)
    // check the format flags with converter focus
    && (xFlags == yFlags);
    
}

bool SanityCheck(const AudioStreamBasicDescription& x)
{
	// This function returns false if there are sufficiently insane values in any field.
	// It is very conservative so even some very unlikely values will pass.
	// This is just meant to catch the case where the data from a file is corrupted.
	
	return 
		(x.mSampleRate >= 0.)	
		&& (x.mSampleRate < 3e6)	// SACD sample rate is 2.8224 MHz
		&& (x.mBytesPerPacket < 1000000)
		&& (x.mFramesPerPacket < 1000000)
		&& (x.mBytesPerFrame < 1000000)
		&& (x.mChannelsPerFrame <= 1024)
		&& (x.mBitsPerChannel <= 1024)
		&& (x.mFormatID != 0)
		&& !(x.mFormatID == kAudioFormatLinearPCM && (x.mFramesPerPacket != 1 || x.mBytesPerPacket != x.mBytesPerFrame));
}

bool CAStreamBasicDescription::FromText(const char *inTextDesc, AudioStreamBasicDescription &fmt)
{
	const char *p = inTextDesc;
	
	memset(&fmt, 0, sizeof(fmt));

	bool isPCM = true;	// until proven otherwise
	UInt32 pcmFlags = kAudioFormatFlagIsPacked | kAudioFormatFlagIsSignedInteger;

	if (p[0] == '-')	// previously we required a leading dash on PCM formats
		++p;

	if (p[0] == 'B' && p[1] == 'E') {
		pcmFlags |= kLinearPCMFormatFlagIsBigEndian;
		p += 2;
	} else if (p[0] == 'L' && p[1] == 'E') {
		p += 2;
	} else {
		// default is native-endian
#if TARGET_RT_BIG_ENDIAN
		pcmFlags |= kLinearPCMFormatFlagIsBigEndian;
#endif
	}
	if (p[0] == 'F') {
		pcmFlags = (pcmFlags & ~static_cast<UInt32>(kAudioFormatFlagIsSignedInteger)) | kAudioFormatFlagIsFloat;
		++p;
	} else {
		if (p[0] == 'U') {
			pcmFlags &= ~static_cast<UInt32>(kAudioFormatFlagIsSignedInteger);
			++p;
		}
		if (p[0] == 'I')
			++p;
		else {
			// it's not PCM; presumably some other format (NOT VALIDATED; use AudioFormat for that)
			isPCM = false;
			p = inTextDesc;	// go back to the beginning
			char buf[4] = { ' ',' ',' ',' ' };
			for (int i = 0; i < 4; ++i) {
				if (*p != '\\') {
					if ((buf[i] = *p++) == '\0') {
						// special-case for 'aac'
						if (i != 3) return false;
						--p;	// keep pointing at the terminating null
						buf[i] = ' ';
						break;
					}
				} else {
					// "\xNN" is a hex byte
					if (*++p != 'x') return false;
					int x;
					if (sscanf(++p, "%02X", &x) != 1) return false;
					buf[i] = static_cast<char>(x);
					p += 2;
				}
			}
			
			if (strchr("-@/#", buf[3])) {
				// further special-casing for 'aac'
				buf[3] = ' ';
				--p;
			}
			
			memcpy(&fmt.mFormatID, buf, 4);
			fmt.mFormatID = CFSwapInt32BigToHost(fmt.mFormatID);
		}
	}
	
	if (isPCM) {
		fmt.mFormatID = kAudioFormatLinearPCM;
		fmt.mFormatFlags = pcmFlags;
		fmt.mFramesPerPacket = 1;
		fmt.mChannelsPerFrame = 1;
		UInt32 bitdepth = 0, fracbits = 0;
		while (isdigit(*p))
			bitdepth = 10 * bitdepth + static_cast<UInt32>(*p++ - '0');
		if (*p == '.') {
			++p;
			if (!isdigit(*p)) {
				fprintf(stderr, "Expected fractional bits following '.'\n");
				goto Bail;
			}
			while (isdigit(*p))
				fracbits = 10 * fracbits + static_cast<UInt32>(*p++ - '0');
			bitdepth += fracbits;
			fmt.mFormatFlags |= (fracbits << kLinearPCMFormatFlagsSampleFractionShift);
		}
		fmt.mBitsPerChannel = bitdepth;
		fmt.mBytesPerPacket = fmt.mBytesPerFrame = (bitdepth + 7) / 8;
		if (bitdepth & 7) {
			// assume unpacked. (packed odd bit depths are describable but not supported in AudioConverter.)
			fmt.mFormatFlags &= ~static_cast<UInt32>(kLinearPCMFormatFlagIsPacked);
			// alignment matters; default to high-aligned. use ':L_' for low.
			fmt.mFormatFlags |= kLinearPCMFormatFlagIsAlignedHigh;
		}
	}
	if (*p == '@') {
		++p;
		while (isdigit(*p))
			fmt.mSampleRate = 10 * fmt.mSampleRate + (*p++ - '0');
	}
	if (*p == '/') {
		UInt32 flags = 0;
		while (true) {
			char c = *++p;
			if (c >= '0' && c <= '9')
				flags = (flags << 4) | static_cast<UInt32>(c - '0');
			else if (c >= 'A' && c <= 'F')
				flags = (flags << 4) | static_cast<UInt32>(c - 'A' + 10);
			else if (c >= 'a' && c <= 'f')
				flags = (flags << 4) | static_cast<UInt32>(c - 'a' + 10);
			else break;
		}
		fmt.mFormatFlags = flags;
	}
	if (*p == '#') {
		++p;
		while (isdigit(*p))
			fmt.mFramesPerPacket = 10 * fmt.mFramesPerPacket + static_cast<UInt32>(*p++ - '0');
	}
	if (*p == ':') {
		++p;
		fmt.mFormatFlags &= ~static_cast<UInt32>(kLinearPCMFormatFlagIsPacked);
		if (*p == 'L')
			fmt.mFormatFlags &= ~static_cast<UInt32>(kLinearPCMFormatFlagIsAlignedHigh);
		else if (*p == 'H')
			fmt.mFormatFlags |= kLinearPCMFormatFlagIsAlignedHigh;
		else
			goto Bail;
		++p;
		UInt32 bytesPerFrame = 0;
		while (isdigit(*p))
			bytesPerFrame = 10 * bytesPerFrame + static_cast<UInt32>(*p++ - '0');
		fmt.mBytesPerFrame = fmt.mBytesPerPacket = bytesPerFrame;
	}
	if (*p == ',') {
		++p;
		int ch = 0;
		while (isdigit(*p))
			ch = 10 * ch + (*p++ - '0');
		fmt.mChannelsPerFrame = static_cast<UInt32>(ch);
		if (*p == 'D') {
			++p;
			if (fmt.mFormatID != kAudioFormatLinearPCM) {
				fprintf(stderr, "non-interleaved flag invalid for non-PCM formats\n");
				goto Bail;
			}
			fmt.mFormatFlags |= kAudioFormatFlagIsNonInterleaved;
		} else {
			if (*p == 'I') ++p;	// default
			if (fmt.mFormatID == kAudioFormatLinearPCM)
				fmt.mBytesPerPacket = fmt.mBytesPerFrame *= static_cast<UInt32>(ch);
		}
	}
	if (*p != '\0') {
		fprintf(stderr, "extra characters at end of format string: %s\n", p);
		goto Bail;
	}
	return true;

Bail:
	fprintf(stderr, "Invalid format string: %s\n", inTextDesc);
	fprintf(stderr, "Syntax of format strings is: \n");
	return false;
}

const char *CAStreamBasicDescription::sTextParsingUsageString = 
	"format[@sample_rate_hz][/format_flags][#frames_per_packet][:LHbytesPerFrame][,channelsDI].\n"
	"Format for PCM is [-][BE|LE]{F|I|UI}{bitdepth}; else a 4-char format code (e.g. aac, alac).\n";