| 1 |
--- |
| 2 |
layout: default |
| 3 |
title: libsndfile : Frequently Asked Questions. |
| 4 |
--- |
| 5 |
|
| 6 |
# libsndfile : Frequently Asked Questions |
| 7 |
|
| 8 |
1. [Do you plan to support XYZ codec in libsndfile?](#Q001) |
| 9 |
2. [In version 0 the SF\_INFO struct had a pcmbitwidth field but version 1 does not. Why?](#Q002) |
| 10 |
3. [Compiling is really slow on MacOS X. Why?](#Q003) |
| 11 |
4. [When trying to compile libsndfile on Solaris I get a "bad substitution" error during linking. What can I do to fix this?](#Q004) |
| 12 |
5. [Why doesn't libsndfile do interleaving/de-interleaving?](#Q005) |
| 13 |
6. [What's the best format for storing temporary files?](#Q006) |
| 14 |
7. [On Linux/Unix/MacOS X, what's the best way of detecting the presence of libsndfile?](#Q007) |
| 15 |
8. [I have libsndfile installed and now I want to use it. I just want a simple Makefile\! What do I do?](#Q008) |
| 16 |
9. [How about adding the ability to write/read sound files to/from memory buffers?](#Q009) |
| 17 |
10. [Reading a 16 bit PCM file as normalised floats and then writing them back changes some sample values. Why?](#Q010) |
| 18 |
11. [I'm having problems with u-law encoded WAV files generated by libsndfile in Winamp. Why?](#Q011) |
| 19 |
12. [I'm looking at sf\_read\*. What are items? What are frames?](#Q012) |
| 20 |
13. [Why can't libsndfile open this Sound Designer II (SD2) file?](#Q013) |
| 21 |
14. [I'd like to statically link libsndfile to my closed source application. Can I buy a license so that this is possible?](#Q014) |
| 22 |
15. [My program is crashing during a call to a function in libsndfile. Is this a bug in libsndfile?](#Q015) |
| 23 |
16. [Will you accept a fix for compiling libsndfile with compiler X?](#Q016) |
| 24 |
17. [Can libsndfile read/write files from/to UNIX pipes?](#Q017) |
| 25 |
18. [Is it possible to build a Universal Binary on Mac OS X?](#Q018) |
| 26 |
19. [I have project files for Visual Studio / XCode / Whatever. Why don't you distribute them with libsndfile?](#Q019) |
| 27 |
20. [Why doesn't libsndfile support MP3?](#Q020) |
| 28 |
21. [How do I use libsndfile in a closed source or commercial program and comply with the license?](#Q021) |
| 29 |
22. [What versions of windows does libsndfile work on?](#Q022) |
| 30 |
23. [I'm cross compiling libsndfile for another platform. How can I run the test suite?](#Q023) |
| 31 |
|
| 32 |
----- |
| 33 |
|
| 34 |
## Q1 : Do you plan to support XYZ codec in libsndfile? {#Q001} |
| 35 |
|
| 36 |
If source code for XYZ codec is available under a suitable license (LGPL, BSD, |
| 37 |
MIT etc) then yes, I'd like to add it. |
| 38 |
|
| 39 |
If suitable documentation is available on how to decode and encode the format |
| 40 |
then maybe, depending on how much work is involved. |
| 41 |
|
| 42 |
If XYZ is some proprietary codec where no source code or documentation is |
| 43 |
available then no. |
| 44 |
|
| 45 |
So if you want support for XYZ codec, first find existing source code or |
| 46 |
documentation. If you can't find either then the answer is no. |
| 47 |
|
| 48 |
## Q2 : In version 0 the SF\_INFO struct had a pcmbitwidth field but version 1 does not. Why? {#Q002} |
| 49 |
|
| 50 |
This was dropped for a number of reasons: |
| 51 |
|
| 52 |
- pcmbitwidth makes little sense on compressed or floating point formats |
| 53 |
- with the new API you really don't need to know it |
| 54 |
|
| 55 |
As documented [here](api.md#note-1) there is now a well defined behaviour which |
| 56 |
ensures that no matter what the bit width of the source file, the scaling always |
| 57 |
does something sensible. This makes it safe to read 8, 16, 24 and 32 bit PCM |
| 58 |
files using `sf_read_short()` and always have the optimal behaviour. |
| 59 |
|
| 60 |
## Q3 : Compiling is really slow on MacOS X. Why? {#Q003} |
| 61 |
|
| 62 |
When you configure and compile libsndfile, it uses the /bin/sh shell for a |
| 63 |
number of tasks (ie configure script and libtool). Older versions of OS X |
| 64 |
(10.2?) shipped a really crappy Bourne shell as /bin/sh which resulted in |
| 65 |
**really** slow compiles. Newer version of OS X ship GNU Bash as /bin/sh and |
| 66 |
this answer doesn't apply in that case. |
| 67 |
|
| 68 |
To fix this I suggest that you install the GNU Bash shell, rename /bin/sh to |
| 69 |
/bin/sh.old and make a symlink from /bin/sh to the bash shell. Bash is designed |
| 70 |
to behave as a Bourne shell when it is called as /bin/sh. |
| 71 |
|
| 72 |
When I did this on my iBook running MacOS X, compile times dropped from 13 |
| 73 |
minutes to 3 minutes. |
| 74 |
|
| 75 |
## Q4 : When trying to compile libsndfile on Solaris I get a "bad substitution" error on linking. Why? {#Q004} |
| 76 |
|
| 77 |
It seems that the Solaris Bourne shell disagrees with GNU libtool. |
| 78 |
|
| 79 |
To fix this I suggest that you install the GNU Bash shell, rename /bin/sh to |
| 80 |
/bin/sh.old and make a symlink from /bin/sh to the bash shell. Bash is designed |
| 81 |
to behave as a Bourne shell when it is called as /bin/sh. |
| 82 |
|
| 83 |
## Q5 : Why doesn't libsndfile do interleaving/de-interleaving? {#Q005} |
| 84 |
|
| 85 |
This problem is bigger than it may seem at first. |
| 86 |
|
| 87 |
For a stereo file, it is a pretty safe bet that a simple interleaving/ |
| 88 |
de-interleaving could satisfy most users. However, for files with more than 2 |
| 89 |
channels this is unlikely to be the case. If the user has a 4 channel file and |
| 90 |
want to play that file on a stereo output sound card they either want the first |
| 91 |
2 channels or they want some mixed combination of the 4 channels. |
| 92 |
|
| 93 |
When you add more channels, the combinations grow exponentially and it becomes |
| 94 |
increasingly difficult to cover even a sensible subset of the possible |
| 95 |
combinations. On top of that, coding any one style of interleaver/de-interleaver |
| 96 |
is trivial, while coding one that can cover all combinations is far from |
| 97 |
trivial. This means that this feature will not be added any time soon. |
| 98 |
|
| 99 |
## Q6 : What's the best format for storing temporary files? {#Q006} |
| 100 |
|
| 101 |
When you want to store temporary data there are a number of requirements: |
| 102 |
|
| 103 |
- A simple, easy to parse header. |
| 104 |
- The format must provide the fastest possible read and write rates (ie avoid |
| 105 |
conversions and encoding/decoding). |
| 106 |
- The file format must be reasonably common and playable by most players. |
| 107 |
- Able to store data in either endian-ness. |
| 108 |
|
| 109 |
The format which best meets these requirements is AU, which allows data to be |
| 110 |
stored in any one of short, int, float and double (among others) formats. |
| 111 |
|
| 112 |
For instance, if an application uses float data internally, its temporary files |
| 113 |
should use a format of (SF_ENDIAN_CPU | SF_FORMAT_AU | SF_FORMAT_FLOAT) which |
| 114 |
will store big endian float data in big endian CPUs and little endian float data |
| 115 |
on little endian CPUs. Reading and writing this format will not require any |
| 116 |
conversions or byte swapping regardless of the host CPU. |
| 117 |
|
| 118 |
## Q7 : On Linux/Unix/MaxOS X, what's the best way of detecting the presence of libsndfile using autoconf? {#Q007} |
| 119 |
|
| 120 |
libsndfile uses the pkg-config (man pkg-config) method of registering itself |
| 121 |
with the host system. The best way of detecting its presence is using something |
| 122 |
like this in configure.ac (or configure.in): |
| 123 |
|
| 124 |
PKG_CHECK_MODULES(SNDFILE, sndfile >= 1.0.2, ac_cv_sndfile=1, ac_cv_sndfile=0) |
| 125 |
|
| 126 |
AC_DEFINE_UNQUOTED([HAVE_SNDFILE],${ac_cv_sndfile}, |
| 127 |
[Set to 1 if you have libsndfile.]) |
| 128 |
|
| 129 |
AC_SUBST(SNDFILE_CFLAGS) |
| 130 |
AC_SUBST(SNDFILE_LIBS) |
| 131 |
|
| 132 |
This will automatically set the **SNDFILE_CFLAGS** and **SNDFILE_LIBS** |
| 133 |
variables which can be used in Makefile.am like this: |
| 134 |
|
| 135 |
SNDFILE_CFLAGS = @SNDFILE_CFLAGS@ |
| 136 |
SNDFILE_LIBS = @SNDFILE_LIBS@ |
| 137 |
|
| 138 |
If you install libsndfile from source, you will probably need to set the |
| 139 |
**PKG_CONFIG_PATH** environment variable as suggested at the end of the |
| 140 |
libsndfile configure process. For instance on my system I get this: |
| 141 |
|
| 142 |
-=-=-=-=-=-=-=-=-=-= Configuration Complete =-=-=-=-=-=-=-=-=-=- |
| 143 |
|
| 144 |
Configuration summary : |
| 145 |
|
| 146 |
Version : ..................... 1.0.5 |
| 147 |
Experimental code : ........... no |
| 148 |
|
| 149 |
Tools : |
| 150 |
|
| 151 |
Compiler is GCC : ............. yes |
| 152 |
GCC major version : ........... 3 |
| 153 |
|
| 154 |
Installation directories : |
| 155 |
|
| 156 |
Library directory : ........... /usr/local/lib |
| 157 |
Program directory : ........... /usr/local/bin |
| 158 |
Pkgconfig directory : ......... /usr/local/lib/pkgconfig |
| 159 |
|
| 160 |
Compiling some other packages against libsndfile may require |
| 161 |
the addition of "/usr/local/lib/pkgconfig" to the |
| 162 |
PKG_CONFIG_PATH environment variable. |
| 163 |
|
| 164 |
## Q8 : I have libsndfile installed and now I want to use it. I just want a simple Makefile\! What do I do? {#Q008} |
| 165 |
|
| 166 |
The **pkg-config** program makes finding the correct compiler flag values and |
| 167 |
library location far easier. During the installation of libsndfile, a file named |
| 168 |
**sndfile.pc** is installed in the directory **${libdir}/pkgconfig** (ie if |
| 169 |
libsndfile is installed in **/usr/local/lib**, **sndfile.pc** will be installed |
| 170 |
in **/usr/local/lib/pkgconfig/**). |
| 171 |
|
| 172 |
In order for pkg-config to find sndfile.pc it may be necessary to point the |
| 173 |
environment variable **PKG_CONFIG_PATH** in the right direction. |
| 174 |
|
| 175 |
export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig |
| 176 |
|
| 177 |
Then, to compile a C file into an object file, the command would be: |
| 178 |
|
| 179 |
gcc `pkg-config --cflags sndfile` -c somefile.c |
| 180 |
|
| 181 |
and to link a number of objects into an executable that links against |
| 182 |
libsndfile, the command would be: |
| 183 |
|
| 184 |
gcc `pkg-config --libs sndfile` obj1.o obj2.o -o program |
| 185 |
|
| 186 |
## Q9 : How about adding the ability to write/read sound files to/from memory buffers? {#Q009} |
| 187 |
|
| 188 |
This has been [added](api.md#open_virtual) for version 1.0.12. |
| 189 |
|
| 190 |
## Q10 : Reading a 16 bit PCM file as normalised floats and then writing them back changes some sample values. Why? {#Q010} |
| 191 |
|
| 192 |
This is caused by the fact that the conversion from 16 bit short to float is |
| 193 |
done by dividing by 32768 (0x8000 in hexadecimal) while the conversion from |
| 194 |
float to 16 bit short is done by multiplying by 32767 (0x7FFF in hex). So for |
| 195 |
instance, a value in a 16 bit PCM file of 20000 gets read as a floating point |
| 196 |
number of 0.6103515625 (20000.0 / 0x8000). Converting that back to a 16 bit |
| 197 |
short results in a value of 19999.3896484375 (0.6103515625 \* 0x7FFF) which then |
| 198 |
gets rounded down to 19999. |
| 199 |
|
| 200 |
You will notice that for this particular case, the error is 1 in 20000 or |
| 201 |
0.005%. Interestingly, for values of less than 16369, dividing by 0x8000 |
| 202 |
followed by multiplying by 0x7FFF and then rounding the result, gives back the |
| 203 |
original value. It turns out that as long as the host operating system supplies |
| 204 |
the 1999 ISO C Standard functions **lrintf** and **lrint** (or a replacement has |
| 205 |
been supplied) then the maximum possible error is 1 in 16369 or about 0.006%. |
| 206 |
|
| 207 |
Regardless of the size of the error, the reason why this is done is rather |
| 208 |
subtle. |
| 209 |
|
| 210 |
In a file containing 16 bit PCM samples, the values are restricted to the range |
| 211 |
[-32768, 32767] while we want floating point values in the range [-1.0, 1.0]. |
| 212 |
The only way to do this conversion is to do a floating point division by a value |
| 213 |
of 0x8000. Converting the other way, the only way to ensure that floating point |
| 214 |
values in the range [-1.0, 1.0] are within the valid range allowed by a 16 bit |
| 215 |
short is to multiply by 0x7FFF. |
| 216 |
|
| 217 |
Some people would say that this is a severe short-coming of libsndfile. I would |
| 218 |
counter that anybody who is constantly converting back and forth between 16 bit |
| 219 |
shorts and normalised floats is going to suffer other losses in audio quality |
| 220 |
that they should also be concerned about. |
| 221 |
|
| 222 |
Since this problem only occurs when converting between integer data on disk and |
| 223 |
normalized floats in the application, it can be avoided by using something other |
| 224 |
than normalized floats in the application. Alternatives to normalized floats are |
| 225 |
the **short** and **int** data types (ie using sf_read_short or sf_read_int) or |
| 226 |
using un-normalized floats (see |
| 227 |
[SFC_SET_NORM_FLOAT](command.html#sfc_set_norm_float)). |
| 228 |
|
| 229 |
Another way to deal with this problem is to consider 16 bit short data as a |
| 230 |
final destination format only, not as an intermediate storage format. All |
| 231 |
intermediate data (ie which is going to be processed further) should be stored |
| 232 |
in floating point format which is supported by all of the most common file |
| 233 |
formats. If floating point files are considered too large (2 times the size of a |
| 234 |
16 bit PCM file), it would also be possible to use 24 bit PCM as an intermediate |
| 235 |
storage format (and which is also supported by most common file types). |
| 236 |
|
| 237 |
## Q11 : I'm having problems with u-law encoded WAV files generated by libsndfile in Winamp. Why? {#Q011} |
| 238 |
|
| 239 |
This is actually a Winamp problem. The official Microsoft spec suggests that the |
| 240 |
'fmt ' chunk should be 18 bytes. Unfortunately at least one of Microsoft's own |
| 241 |
applications (Sound Recorder on Win98 I believe) did not accept 18 bytes 'fmt ' |
| 242 |
chunks. |
| 243 |
|
| 244 |
Michael Lee did some experimenting and found that: |
| 245 |
|
| 246 |
> I have checked that Windows Media Player 9, QuickTime Player 6.4, RealOne |
| 247 |
> Player 2.0 and GoldWave 5.06 can all play u-law files with 16-byte or 18-byte |
| 248 |
> 'fmt ' chunk. Only Winamp (2.91) and foobar2000 are unable to play u-law files |
| 249 |
> with 16-byte 'fmt ' chunk. |
| 250 |
|
| 251 |
Even this is a very small sampling of all the players out there. For that reason |
| 252 |
it is probably not a good idea to change this now because there is the risk of |
| 253 |
breaking something that currently works. |
| 254 |
|
| 255 |
## Q12 : I'm looking at sf_read*. What are items? What are frames? {#Q012} |
| 256 |
|
| 257 |
An `item` is a single sample of the data type you are reading; ie a single |
| 258 |
`short` value for `sf_read_short` or a single `float` for `sf_read_float`. |
| 259 |
|
| 260 |
For a sound file with only one channel, a frame is the same as a item (ie a |
| 261 |
single sample) while for multi channel sound files, a single frame contains a |
| 262 |
single item for each channel. |
| 263 |
|
| 264 |
Here are two simple, correct examples, both of which are assumed to be working |
| 265 |
on a stereo file, first using items: |
| 266 |
|
| 267 |
```c |
| 268 |
#define CHANNELS 2 |
| 269 |
short data [CHANNELS * 100] ; |
| 270 |
sf_count items_read = sf_read_short (file, data, 200) ; |
| 271 |
assert (items_read == 200) ; |
| 272 |
``` |
| 273 |
|
| 274 |
and now reading the exact same amount of data using frames: |
| 275 |
|
| 276 |
```c |
| 277 |
#define CHANNELS 2 |
| 278 |
short data [CHANNELS * 100] ; |
| 279 |
sf_count frames_read = sf_readf_short (file, data, 100) ; |
| 280 |
assert (frames_read == 100) ; |
| 281 |
``` |
| 282 |
|
| 283 |
## Q13 : Why can't libsndfile open this Sound Designer II (SD2) file? {#Q013} |
| 284 |
|
| 285 |
This is somewhat complicated. First some background. |
| 286 |
|
| 287 |
SD2 files are native to the Apple Macintosh platform and use features of the Mac |
| 288 |
filesystem (file resource forks) to store the file's sample rate, number of |
| 289 |
channels, sample width and more. When you look at a file and its resource fork |
| 290 |
on Mac OS X it looks like this: |
| 291 |
|
| 292 |
-rw-r--r-- 1 erikd erikd 46512 Oct 18 22:57 file.sd2 |
| 293 |
-rw-r--r-- 1 erikd erikd 538 Oct 18 22:57 file.sd2/rsrc |
| 294 |
|
| 295 |
Notice how the file itself looks like a directory containing a single file named |
| 296 |
**rsrc**. When libsndfile is compiled for MacOS X, it should open (for write and |
| 297 |
read) SD2 file with resource forks like this without any problems. It will also |
| 298 |
handle files with the resource fork in a separate file as described below. |
| 299 |
|
| 300 |
When SD2 files are moved to other platforms, the resource fork of the file can |
| 301 |
sometimes be dropped altogether. All that remains is the raw audio data and no |
| 302 |
information about the number of channels, sample rate or bit width which makes |
| 303 |
it a little difficult for libsndfile to open the file. |
| 304 |
|
| 305 |
However, it is possible to safely move an SD2 file to a Linux or Windows |
| 306 |
machine. For instance, when an SD2 file is copied from inside MacOS X to a |
| 307 |
windows shared directory or a Samba share (ie Linux), MacOS X is clever enough |
| 308 |
to store the resource fork of the file in a separate hidden file in the same |
| 309 |
directory like this: |
| 310 |
|
| 311 |
-rw-r--r-- 1 erikd erikd 538 Oct 18 22:57 ._file.sd2 |
| 312 |
-rw-r--r-- 1 erikd erikd 46512 Oct 18 22:57 file.sd2 |
| 313 |
|
| 314 |
Regardless of what platform it is running on, when libsndfile is asked to open a |
| 315 |
file named **"foo"** and it can't recognize the file type from the data in the |
| 316 |
file, it will attempt to open the resource fork and if that fails, it then tries |
| 317 |
to open a file named **"._foo"** to see if the file has a valid resource fork. |
| 318 |
This is the same regardless of whether the file is being opened for read or |
| 319 |
write. |
| 320 |
|
| 321 |
In short, libsndfile should open SD2 files with a valid resource fork on all of |
| 322 |
the platforms that libsndfile supports. If a file has lost its resource fork, |
| 323 |
the only option is the open the file using the SF_FORMAT_RAW option and guessing |
| 324 |
its sample rate, channel count and bit width. |
| 325 |
|
| 326 |
Occasionally, when SD2 files are moved to other systems, the file is |
| 327 |
[BinHexed](http://www.macdisk.com/binhexen.php3) which wraps the resource fork |
| 328 |
and the data fork together. For these files, it would be possible to write a |
| 329 |
BinHex parser but there is not a lot to gain considering how rare these BinHexed |
| 330 |
SD2 files are. |
| 331 |
|
| 332 |
## Q14 : I'd like to statically link libsndfile to my closed source application. Can I buy a license so that this is possible? {#Q014} |
| 333 |
|
| 334 |
Unfortunately no. libsndfile contains code written by other people who have |
| 335 |
agreed that their code be used under the GNU LGPL but no more. Even if they were |
| 336 |
to agree, there would be significant difficulties in dividing up the payments |
| 337 |
fairly. |
| 338 |
|
| 339 |
The **only** way you can legally use libsndfile as a statically linked library |
| 340 |
is if your application is released under the GNU GPL or LGPL. |
| 341 |
|
| 342 |
## Q15 : My program is crashing during a call to a function in libsndfile. Is this a bug in libsndfile? {#Q015} |
| 343 |
|
| 344 |
libsndfile is being used by large numbers of people all over the world without |
| 345 |
any problems like this. That means that it is much more likely that your code |
| 346 |
has a bug than libsndfile. However, it is still possible that there is a bug in |
| 347 |
libsndfile. |
| 348 |
|
| 349 |
To figure out whether it is your code or libsndfile you should do the following: |
| 350 |
|
| 351 |
- Make sure you are compiling your code with warnings switched on and that you |
| 352 |
fix as many warnings as possible. With the GNU compiler (gcc) I would |
| 353 |
recommend at least **-W -Wall -Werror** which will force you to fix all |
| 354 |
warnings before you can run the code. |
| 355 |
- Try using a memory debugger. [Valgrind](http://valgrind.kde.org/) on x86 Linux |
| 356 |
is excellent. [Purify](http://www.ibm.com/software/awdtools/purify/) also has |
| 357 |
a good reputation. |
| 358 |
- If the code is clean after the above two steps and you still get a crash in |
| 359 |
libsndfile, then send me a small snippet of code (no more than 30-40 lines) |
| 360 |
which includes the call to sf_open() and also shows how all variables passed |
| 361 |
to/returned from sf_open() are defined. |
| 362 |
|
| 363 |
## Q16 : Will you accept a fix for compiling libsndfile with compiler X? {#Q016} |
| 364 |
|
| 365 |
If compiler X is a C++ compiler then no. C and C++ are different enough to make |
| 366 |
writing code that compiles as valid C and valid C++ too difficult. I would |
| 367 |
rather spend my time fixing bugs and adding features. |
| 368 |
|
| 369 |
If compiler X is a C compiler then I will do what I can as long as that does not |
| 370 |
hamper the correctness, portability and maintainability of the existing code. It |
| 371 |
should be noted however that libsndfile uses features specified by the 1999 ISO |
| 372 |
C Standard. This can make compiling libsndfile with some older compilers |
| 373 |
difficult. |
| 374 |
|
| 375 |
## Q17 : Can libsndfile read/write files from/to UNIX pipes? {#Q017} |
| 376 |
|
| 377 |
Yes, libsndfile can read files from pipes. Unfortunately, the write case is much |
| 378 |
more complicated. |
| 379 |
|
| 380 |
File formats like AIFF and WAV have information at the start of the file (the |
| 381 |
file header) which states the length of the file, the number of sample frames |
| 382 |
etc. This information must be filled in correctly when the file header is |
| 383 |
written, but this information is not reliably known until the file is closed. |
| 384 |
This means that libsndfile cannot write AIFF, WAV and many other file types to a |
| 385 |
pipe. |
| 386 |
|
| 387 |
However, there is at least one file format (AU) which is specifically designed |
| 388 |
to be written to a pipe. Like AIFF and WAV, AU has a header with a sample frames |
| 389 |
field, but it is specifically allowable to set that frames field to 0x7FFFFFFF |
| 390 |
if the file length is not known when the header is written. The AU file format |
| 391 |
can also hold data in many of the standard formats (ie SF_FORMAT_PCM_16, |
| 392 |
SF_FORMAT_PCM_24, SF_FORMAT_FLOAT etc) as well as allowing data in both big and |
| 393 |
little endian format. |
| 394 |
|
| 395 |
See also [FAQ Q6](#Q006). |
| 396 |
|
| 397 |
## Q18 : Is it possible to build a Universal Binary on Mac OS X? {#Q018} |
| 398 |
|
| 399 |
Yes, but you must do two separate configure/build/test runs; one on PowerPC and |
| 400 |
one on Intel. It is then possible to merge the binaries into a single universal |
| 401 |
binary using one of the programs in the Apple tool chain. |
| 402 |
|
| 403 |
It is **not** possible to build a working universal binary via a single |
| 404 |
compile/build run on a single CPU. |
| 405 |
|
| 406 |
The problem is that the libsndfile build process detects features of the CPU its |
| 407 |
being built for during the configure process and when building a universal |
| 408 |
binary, configure is only run once and that data is then used for both CPUs. |
| 409 |
That configure data will be wrong for one of those CPUs. You will still be able |
| 410 |
to compile libsndfile, and the test suite will pass on the machine you compiled |
| 411 |
it on. However, if you take the universal binary test suite programs compiled on |
| 412 |
one CPU and run them on the other, the test suite will fail. |
| 413 |
|
| 414 |
Part of the problem is that the CPU endian-ness is detected at configure time. |
| 415 |
Yes, I know the Apple compiler defines one of the macros \_\_LITTLE\_ENDIAN\_\_ |
| 416 |
and \_\_BIG\_ENDIAN\_\_, but those macros are not part of the 1999 ISO C |
| 417 |
Standard and they are not portable. |
| 418 |
|
| 419 |
Endian issues are not the only reason why the cross compiled binary will fail. |
| 420 |
The configure script also detects other CPU specific idiosyncrasies to provide |
| 421 |
more optimized code. |
| 422 |
|
| 423 |
Finally, the real show stopper problem with universal binaries is the problem |
| 424 |
with the test suite. libsndfile contains a huge, comprehensive test suite. When |
| 425 |
you compile a universal binary and run the test suite, you only test the native |
| 426 |
compile. The cross compiled binary (the one with the much higher chance of |
| 427 |
having problems) cannot be tested. |
| 428 |
|
| 429 |
Now, if you have read this far you're probably thinking there must be a way to |
| 430 |
fix this and there probably is. The problem is that its a hell of a lot of work |
| 431 |
and would require significant changes to the configure process, the internal |
| 432 |
code and the test suite. In addition, these changes must not break compilation |
| 433 |
on any of the platforms libsndfile is currently working on. |
| 434 |
|
| 435 |
## Q19 : I have project files for Visual Studio / XCode / Whatever. Why don't you distribute them with libsndfile? {#Q019} |
| 436 |
|
| 437 |
Use CMake project. |
| 438 |
|
| 439 |
## Q20 : Why doesn't libsndfile support MP3? {#Q020} |
| 440 |
|
| 441 |
~~In the past, MP3 was not supported because the technology behind MP3 was |
| 442 |
patented. Those patents have now expired and there is an |
| 443 |
[open ticket](https://github.com/libsndfile/libsndfile/issues/258) to implement |
| 444 |
MP3 support.~~ |
| 445 |
|
| 446 |
**Update :** Starting from version 1.1.0 libsndfile supports MP3 format. |
| 447 |
|
| 448 |
## Q21 : How do I use libsndfile in a closed source or commercial program and comply with the license? {#Q021} |
| 449 |
|
| 450 |
Here is a checklist of things you need to do to make sure your use of libsndfile |
| 451 |
in a closed source or commercial project complies with the license libsndfile is |
| 452 |
released under, the GNU Lesser General Public License (LGPL): |
| 453 |
|
| 454 |
- Make sure you are linking to libsndfile as a shared library (Linux and Unix |
| 455 |
systems), Dynamic Link Library (Microsoft Windows) or dynlib (Mac OS X). If |
| 456 |
you are using some other operating system that doesn't allow dynamically |
| 457 |
linked libraries, you will not be able to use libsndfile unless you release |
| 458 |
the source code to your program. |
| 459 |
- In the licensing documentation for your program, add a statement that your |
| 460 |
software depends on libsndfile and that libsndfile is released under the GNU |
| 461 |
Lesser General Public License, either |
| 462 |
[version 2.1](http://www.gnu.org/licenses/lgpl-2.1.txt) or optionally |
| 463 |
[version 3](http://www.gnu.org/licenses/lgpl.txt). |
| 464 |
- Include the text for both versions of the license, possibly as separate files |
| 465 |
named libsndfile_lgpl_v2_1.txt and libsndfile_lgpl_v3.txt. |
| 466 |
|
| 467 |
## Q22 : What versions of Windows does libsndfile work on? {#Q022} |
| 468 |
|
| 469 |
New versions of libsndfile binary releases require Wiindows Vista. If you need |
| 470 |
Windows XP support, you can build DLL from sources, we don't use specific WinXP |
| 471 |
features. |
| 472 |
|
| 473 |
## Q23 : I'm cross compiling libsndfile for another platform. How can I run the test suite? {#Q023} |
| 474 |
|
| 475 |
Since version 1.0.21 the top level Makefile has an extra make target, |
| 476 |
'test-tarball'. Building this target creates a tarball called called: |
| 477 |
|
| 478 |
` libsndfile-testsuite-${host_triplet}-${version}.tar.gz` |
| 479 |
|
| 480 |
in the top level directory. This tarball can then be copied to the target |
| 481 |
platform. Once untarred and test script `test_wrapper.sh` can be run from the |
| 482 |
top level of the extracted tarball. |