RIFF/WAVE support, using ID3v2 #321

Borewit · 2017-08-17T18:14:35Z

Initial implementation of RIFF/WAVE format.

Currently using ID3v2 tag chunk to store metadata.

Storing metadata in the RIFF/INFO tag is still something which needs to be done.

Related issues:

Support RIFF INFO chunk metadata (AVI, WAV, XMA, xWMA, RMI, DLS) #207: Support RIFF INFO chunk metadata (AVI, WAV, XMA, xWMA, RMI, DLS)
PICARD-1128: Support Microsoft WAVE format (RIFF/WAVE) tagging
Add RIFF/WAVE support Borewit/music-metadata#19

lazka · 2017-08-20T18:05:07Z

mutagen/wave.py

+        except KeyError as e:
+            raise error(str(e))
+
+        data = waveFormatChunk.read()


check if 16 bytes are read so the unpack() below doesn't fail for truncated files

lazka · 2017-08-20T18:07:05Z

mutagen/wave.py

+        self.bitrate = self.channels * block_align * self.sample_rate
+
+        try:
+            waveDataChunk = waveFile[u'data']


Does checking for "data" add anything useful?

It is not checking, it is accessing the data chunk length to calculate the duration (length in seconds) of the audio. I put comment on top of it; removed throwing exception in following line with assigning 0 to number_of_samples.

ah, right, I missed that

lazka · 2017-08-20T18:07:33Z

mutagen/wave.py

+            waveDataChunk = waveFile[u'data']
+        except KeyError as e:
+            raise error(str(e))
+


Check that block_align and self.sample_rate > 0 or we get a division by zero

lazka · 2017-08-20T18:09:37Z

mutagen/wave.py

+    """Completely removes the ID3 chunk from the AIFF file"""
+
+    try:
+        del RiffFile(filething.fileobj)[u'ID3']


Is "ID3" correct here?

If the ID3v2 chunk should be considered as a multi purpose chunk (like RIFF and LIST) it should be in capitals everywhere. But that is a different discussion. ✔

lazka · 2017-08-20T18:11:15Z

mutagen/_riff.py

+            raise KeyError("First chunk should be a RIFF chunk.")
+
+        # Read the RIFF file Type
+        self.fileType = fileobj.read(4).decode('ascii')


decode can raise

lazka · 2017-08-20T18:12:11Z

mutagen/_riff.py

+
+            # Calculate the location of the next chunk,
+            # considering the pad byte
+            # self.__next_offset += self.__next_offset % 2


What's this about?

lazka · 2017-08-20T18:13:59Z

mutagen/wave.py

+    # looks like this is failing if python is not started with -bb in TravisCI:
+    # assert isinstance(id, text_type)
+
+    return ((len(id) <= 4) and (min(id) >= u' ') and


That has been derived from aiff.py

If the chunks are indexed including the space padding, the key should be 4 bytes long indeed

Do you have any idea why assert isinstance(id, text_type) is failing in some circumstances?
What does option -bb do?

-bb makes text_type(bytes()) fail (and comparisons between bytes/text iirc)

lazka · 2017-08-20T18:15:05Z

mutagen/wave.py

+
+def assert_valid_chunk_id(id):
+    if not is_valid_chunk_id(id):
+        raise KeyError("RIFF/WAVE-chunk-Id must be four ASCII characters.")


nitpick: I'd prefer ValueError

✔ moved to: _riff.py

lazka · 2017-08-20T18:18:41Z

mutagen/_riff.py

+def is_valid_chunk_id(id):
+    assert isinstance(id, text_type)
+
+    return ((len(id) <= 4) and (min(id) >= u' ') and


The entire check is odd, why only the first character for the lower limit and the last char for the upper limit.

if len(id) != 4: return False for i in range(0, 3): if id[i] < u' ' or id[i] > u'~': return False return True

✔

tests/test_wave.py

lazka · 2017-08-20T18:23:59Z

mutagen/wave.py

+            self.tags.filename = self.filename
+
+        fileobj.seek(0, 0)
+        self.info = WaveStreamInfo(fileobj)


Can we load WaveStreamInfo() before the ID3 tags? Might make it fail earlier in case of a wrong format.

lazka · 2017-08-20T18:25:23Z

mutagen/wave.py

+
+        if 'id3 ' in waveFile:
+            waveFile['id3 '].delete()
+        if 'ID3 ' in waveFile:


Is that needed with the lower() stuff in WaveFile?

lazka · 2017-08-20T18:32:02Z

mutagen/wave.py

+        self._update_size(new_data_size)
+
+
+class WaveFile(RiffFile):


Why does this inherit from RiffFile if it reimplements everything?

I agree it is messy.

I started of with a copy of aiff.py because I knew it had a pretty simular structure.

The idea is to move everything what is part of the RIFF standard into _riff.py, and WAVE specific stuff into wave.py. Which was easier said then done, because is was not always clear to me what belonged to core RIFF and what to MS-WAVE. Because the sub-chunks used a little endian size, in contradiction to the RIFF chuck using big endian, I figured the sub-chunk definition belonged to WAV spec. Looks like I was wrong. I found somewhere that chunks with capital tag-id's are common purpose chunks and lower case are RIFF-type specific.

Anyway, I will do my best do clean up the mess.

Ref: - quodlibet#321 (comment) - a5c3785ed8d6a35868bc169f07e40e889087fd2e

Borewit · 2017-09-06T19:08:16Z

Something is going wrong in those generic unit tests, I have difficulties understanding how this testing works. I see the constructor is called multiple times with one of the test files, and at a certain point it does not seem to recognize the file anymore as a RIFF/WAVE.

RIFF/WAVE_WaveID3(ID3)RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav
Falsifying example: run(t=<test___init__._TestFileObj at 0x2d8d39f64a8>)
RIFF/WAVE __init__:  C:\Users\Borewit\code\github\mutagen\tests\data\silence-2s-PCM-16000-08-ID3v23.wav

Ran 1 test in 8.758s

FAILED (errors=1)

Error
Traceback (most recent call last):
  File "C:\Program Files\Python36\lib\unittest\case.py", line 59, in testPartExecutor
    yield
  File "C:\Program Files\Python36\lib\unittest\case.py", line 601, in run
    testMethod()
  File "C:\Users\Borewit\code\github\mutagen\tests\test___init__.py", line 539, in test_mock_fileobj
    run()
  File "C:\Users\Borewit\code\github\mutagen\tests\test___init__.py", line 533, in run
    def run(t):
  File "C:\Program Files\Python36\lib\site-packages\hypothesis\core.py", line 726, in wrapped_test
    state.run()
  File "C:\Program Files\Python36\lib\site-packages\hypothesis\core.py", line 619, in run
    print_example=True, is_final=True
  File "C:\Program Files\Python36\lib\site-packages\hypothesis\executors.py", line 58, in default_new_style_executor
    return function(data)
  File "C:\Program Files\Python36\lib\site-packages\hypothesis\core.py", line 115, in run
    return test(*args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\tests\test___init__.py", line 535, in run
    File(t)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_util.py", line 138, in wrapper_func
    return func(h, *args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_file.py", line 298, in File
    return Kind(fileobj, filename=filething.filename)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_file.py", line 49, in __init__
    self.load(*args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_util.py", line 159, in wrapper
    return func(*args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_util.py", line 130, in wrapper
    return func(self, h, *args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\wave.py", line 206, in load
    self.tags = _WaveID3(fileobj, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\id3\_file.py", line 77, in __init__
    super(ID3, self).__init__(*args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\id3\_tags.py", line 177, in __init__
    super(ID3Tags, self).__init__(*args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_util.py", line 519, in __init__
    super(DictProxy, self).__init__(*args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_tags.py", line 111, in __init__
    self.load(*args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_util.py", line 159, in wrapper
    return func(*args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_util.py", line 130, in wrapper
    return func(self, h, *args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\id3\_file.py", line 147, in load
    self._pre_load_header(fileobj)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\wave.py", line 107, in _pre_load_header
    fileobj.seek(WaveFile(fileobj)[u'id3 '].data_offset)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\wave.py", line 34, in __init__
    raise ValueError("Expected RIFF/WAVE.")
ValueError: Expected RIFF/WAVE.

lazka · 2017-09-07T07:38:43Z

Something is going wrong in those generic unit tests, I have difficulties understanding how this testing works. I see the constructor is called multiple times with one of the test files, and at a certain point it does not seem to recognize the file anymore as a RIFF/WAVE.

Yeah, it gets called with variations of truncated files and expects a subclass of MutagenError to be raised or to succeed. This for uncovering wrong/missing error handling in the parsing code.

In your example traceback: load() should not raise ValueError for invalid files.

Borewit · 2018-03-01T19:05:38Z

If I have this failing test case:

Falsifying example: run(t=<test___init__._TestFileObj at 0x25595f0ac18>)

You can reproduce this example by temporarily adding @reproduce_failure('3.44.26', b'AAEA') as a decorator on your test case

Error
Traceback (most recent call last):
  File "C:\Users\Borewit\AppData\Local\Programs\Python\Python36\lib\unittest\case.py", line 59, in testPartExecutor
    yield
  File "C:\Users\Borewit\AppData\Local\Programs\Python\Python36\lib\unittest\case.py", line 605, in run
    testMethod()
  File "C:\Users\Borewit\code\github\mutagen\tests\test___init__.py", line 376, in test_test_fileobj_delete
    run()
  File "C:\Users\Borewit\code\github\mutagen\tests\test___init__.py", line 369, in run
    h, lambda t: o.delete(fileobj=t)))
  File "C:\Users\Borewit\AppData\Local\Programs\Python\Python36\lib\site-packages\hypothesis\core.py", line 1002, in wrapped_test
    state.run()
  File "C:\Users\Borewit\AppData\Local\Programs\Python\Python36\lib\site-packages\hypothesis\core.py", line 818, in run
    falsifying_example.__expected_traceback,
  File "C:\Users\Borewit\AppData\Local\Programs\Python\Python36\lib\site-packages\hypothesis\core.py", line 579, in execute
    result = self.test_runner(data, run)
  File "C:\Users\Borewit\AppData\Local\Programs\Python\Python36\lib\site-packages\hypothesis\executors.py", line 58, in default_new_style_executor
    return function(data)
  File "C:\Users\Borewit\AppData\Local\Programs\Python\Python36\lib\site-packages\hypothesis\core.py", line 571, in run
    return test(*args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\tests\test___init__.py", line 369, in run
    h, lambda t: o.delete(fileobj=t)))
  File "C:\Users\Borewit\AppData\Local\Programs\Python\Python36\lib\site-packages\hypothesis\core.py", line 518, in test
    result = self.test(*args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\tests\test___init__.py", line 372, in run
    o.delete(fileobj=t)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_util.py", line 140, in wrapper
    return func(self, h, *args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_file.py", line 120, in delete
    return self.tags.delete(filething)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_util.py", line 140, in wrapper
    return func(self, h, *args, **kwargs)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\wave.py", line 143, in delete
    waveFile = WaveFile(fileobj)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\wave.py", line 35, in __init__
    RiffFile.__init__(self, fileobj)
  File "C:\Users\Borewit\code\github\mutagen\mutagen\_riff.py", line 140, in __init__
    fileobj.seek(0)
  File "C:\Users\Borewit\code\github\mutagen\tests\test___init__.py", line 254, in seek
    self._check_fail()
  File "C:\Users\Borewit\code\github\mutagen\tests\test___init__.py", line 216, in _check_fail
    raise IOError("fail")
OSError: fail

How can I execute just this test case?

It is stating @reproduce_failure('3.44.26', b'AAEA') as a decorator on your test case, I tried to add this to a few places, but was not able to isolate the test case.

Borewit · 2018-04-18T19:41:53Z

@lazka Can you please can give some guidance?

Ref: - quodlibet#321 (comment) - a5c3785ed8d6a35868bc169f07e40e889087fd2e

phw · 2018-09-11T11:39:45Z

@Borewit I tested your code, rebased it against master and fixed the tests at https://github.com/phw/mutagen/commits/feature/riff-wave

I also have a local patch for Picard to use this and tested this with Picard, seems to work just fine.

You can either update your PR here or I can open a new PR, whatever you prefer.

regarding the code itself I'm not so much into mutagen's internals. What worries me a bit is how the code in _riff.py duplicates a lot of the code for IFF in aiff.py. But I'm also not enough into these formats to say how much this separation is warranted, all I know is that both AIFF and RIFF are based on IFF. @lazka What are your thoughts?

Borewit · 2018-09-11T18:29:37Z

@phw thank you so much for picking up this work!

It would be fantastic if the functionality ends up in Picard. Due to a lack of experience programming in Python and knowledge of Mutagen I got stuck and had to give up.

AIFF and RIFF are somewhat simular. For example the size in AIFF chunk is big endian, in RIFF the size is encoded in little endian.

I will include your changes in my PR. I created PR on my own PR. I haved added you as collaberator, which should give you the freedom to push changes on top of my PR.

You don't hurt my feelings if you PR my changes directly (I have done that one time, and folks became pissed of, as if I comitted some plagiarism. I never understood that).

phw · 2018-09-12T07:38:03Z

Thanks for including my changes. Looks like we still have some issues with Python 2, I had not tested this locally. I will see whether I can provide fixes for this later this week, and if so will directly push it to this PR.

You don't hurt my feelings if you PR my changes directly (I have done that one time, and folks became pissed of, as if I comitted some plagiarism. I never understood that).

Yes, I also don't mind if people put their corrections on top of my PRs (as long as they leave my authorship intact and don't claim it to be their own work). But I know people sometimes take issue on this hence I rather ask before doing anything :)

phw · 2018-09-12T08:15:08Z

This was easier than expected, just a few instances where for Python 2 a proper unicode string is expected. Builds are running and so far it is looking good :)

Borewit · 2018-09-12T11:39:57Z

Looks like @phw nailed the last issues.

@lazka Can you please review?

lazka · 2018-09-12T11:45:29Z

regarding the code itself I'm not so much into mutagen's internals. What worries me a bit is how the code in _riff.py duplicates a lot of the code for IFF in aiff.py. But I'm also not enough into these formats to say how much this separation is warranted, all I know is that both AIFF and RIFF are based on IFF. @lazka What are your thoughts?

I don't mind the duplication.

@lazka Can you please review?

ok

Borewit · 2018-09-12T12:07:18Z

Thanks @lazka.

@phw, regarding:

I also have a local patch for Picard to use this and tested this with Picard, seems to work just fine.

I do as well, online or offline, have to dig it up. Can we allign the changes for a PR on Picard?
Related Picard issue.

phw · 2018-09-13T06:56:35Z

I do as well, online or offline, have to dig it up. Can we allign the changes for a PR on Picard?

Yes please :) I have my proposed changes linked in the Picard ticket.

lazka · 2018-09-14T22:05:16Z

mutagen/_riff.py

+        raise ValueError("Invalid RIFF-chunk-ID.")
+
+
+class RiffChunkHeader():


inherit from object

lazka · 2018-09-14T22:05:33Z

mutagen/aiff.py

@@ -44,9 +44,6 @@ def is_valid_chunk_id(id):


 def assert_valid_chunk_id(id):
-
-    assert isinstance(id, text_type)


why is this removed?

This was readded

lazka · 2018-09-14T22:17:39Z

mutagen/wave.py

+        try:
+            waveDataChunk = waveFile[u'data']
+            self.number_of_samples = waveDataChunk.data_size / block_align
+        except KeyError:


ZeroDivisionError

lazka · 2018-09-14T22:18:19Z

mutagen/wave.py

+        sample_size (`int`): The audio sample size
+    """
+
+    length = 0


lazka · 2018-09-14T22:20:57Z

mutagen/wave.py

+
+        try:
+            self.info = WaveStreamInfo(fileobj)
+        except ValueError as e:


Where is ValueError raised? Is this needed?

lazka · 2018-09-14T22:24:24Z

mutagen/wave.py

+    """WAVE stream parsing errors."""
+
+
+class WaveFile(RiffFile):


Can we rename this to "_WaveFile"?

lazka · 2018-09-14T22:34:59Z

mutagen/_riff.py

+            raise KeyError(
+                "%r has no %r chunk" % (self._fileobj, id_))
+
+    def __delitem__(self, id_):


I know aiff has the same code, but can you change this to delete_chunk(). It's confusing that del foo['id'] writes to the file and can raise IOError.

lazka · 2018-09-14T22:35:18Z

mutagen/wave.py

+    """Completely removes the ID3 chunk from the RIFF file"""
+
+    try:
+        del RiffFile(filething.fileobj)[u'id3']


change to delete_chunk(), see above

lazka · 2018-09-14T22:36:12Z

mutagen/wave.py

+        RiffFile.__init__(self, fileobj)
+
+        if self.fileType != u'WAVE':
+            raise MutagenError("Expected RIFF/WAVE.")


MutagenError -> error

Ref: - quodlibet#321 (comment) - a5c3785ed8d6a35868bc169f07e40e889087fd2e

…should be a text type

ValueError.

* fix wrong abstract attribute reference * fix reference to root ('RIFF') chuck

…Endian)

Borewit · 2019-10-15T12:21:50Z

Thank you so much for pushing this effort forward @phw!

Borewit · 2019-10-15T18:28:46Z

If you check out your branch please be careful.

Copy that, I will not overwrite it.

For consistency with other formats.

phw · 2019-10-19T20:12:05Z

I looked at the issue reported in #392 in detail, and examined the file structure exactly. Two results:

The issue is likely a foobar2000 bug, I reported it there
The RIFF / WAVE implementation here is very solid. I double checked all the code that adds and removes chunk, especially the file calculation, and checked the actual file structure after every action. All working as intended. Not even a slightly abnormal file as generated by f2k is breaking this. @Borewit great work!

@lazka I think this is good for a final review. If you want I can cleanup the git history a bit and combine some commits (but make sure to maintain authorship).

phw · 2019-10-21T06:58:20Z

@lazka Looks like this needs some more work, also for AIFF actually. See my comments on #392

phw · 2019-10-23T06:58:08Z

New PR with updated code at #408

lazka · 2020-05-04T18:25:18Z

See #408

Borewit mentioned this pull request Aug 18, 2017

Support RIFF/LIST-INFO (EXIF) tags Borewit/music-metadata#28

Closed

lazka reviewed Aug 20, 2017

View reviewed changes

Borewit added a commit to Borewit/mutagen that referenced this pull request Aug 26, 2017

KeyError => ValueError

8f67deb

Ref: - quodlibet#321 (comment) - a5c3785ed8d6a35868bc169f07e40e889087fd2e

Borewit added a commit to Borewit/mutagen that referenced this pull request Aug 26, 2017

Minor changes based on lessons learned from quodlibet#321

521f195

Borewit mentioned this pull request Aug 26, 2017

Minor changes in AIFF functionality #322

Merged

wsngamerz mentioned this pull request Apr 16, 2018

Support RIFF INFO chunk metadata (AVI, WAV, XMA, xWMA, RMI, DLS) #207

Open

spl0k mentioned this pull request Aug 26, 2018

wav support? spl0k/supysonic#110

Closed

phw pushed a commit to phw/mutagen that referenced this pull request Sep 11, 2018

KeyError => ValueError

f8bc6c7

Ref: - quodlibet#321 (comment) - a5c3785ed8d6a35868bc169f07e40e889087fd2e

lazka reviewed Sep 14, 2018

View reviewed changes

lazka mentioned this pull request Sep 29, 2018

WAV format - ModuleNotFoundError: No module named 'mutagen.wave' #353

Closed

Borewit added a commit to Borewit/mutagen that referenced this pull request Sep 2, 2019

KeyError => ValueError

6c78c6a

Ref: - quodlibet#321 (comment) - a5c3785ed8d6a35868bc169f07e40e889087fd2e

Borewit force-pushed the feature/riff-wave branch 4 times, most recently from 5b5f636 to 6ff97bf Compare September 2, 2019 19:29

Borewit and others added 14 commits October 15, 2019 09:58

Try to fix: isinstance(id, text_type) return false on chunk-id which …

8b097ac

…should be a text type

Remove unused import

21552ef

Fix lint errors

26d7670

Restore test file to include all tests

22a9a05

Fix unit tests loading RIFF/WAVE by throwing MutagenError instead of

62f7eb0

ValueError.

Fix function RiffFile.insert_chunk:

1be502a

* fix wrong abstract attribute reference * fix reference to root ('RIFF') chuck

Fix endianness off RIFF size field (should be Little-Endian, was Big-…

825da85

…Endian)

Normalize RIFF ID (FourCC) to exclude spaces on 'bean' level

395911b

Remove ID3 header remainder

beec64f

Add a unit test mocking Picard saving an ID3 header to a WAVE file.

a09c826

Resolve review

c4374b8

Wave: Fixed style errors

9c2fd9b

wave: Fixed tests

e507764

Wave: Python 2 fixes

b3abeeb

phw force-pushed the feature/riff-wave branch from 65cb514 to c4b0b38 Compare October 15, 2019 07:58

phw mentioned this pull request Oct 15, 2019

WAV corrupted when tagging with foobar2k #392

Closed

phw added 3 commits October 15, 2019 21:12

wave: Code review fixes

51cf54b

wave: Added API docs

9d29fbd

wave: Renamed sample_size to bits_per_sample

52274d7

For consistency with other formats.

phw force-pushed the feature/riff-wave branch from c4b0b38 to 52274d7 Compare October 15, 2019 19:13

riff: Minor documentation fixes

e4f54e5

phw mentioned this pull request Oct 22, 2019

Full RIFF/WAVE support with ID3 tags #408

Merged

lazka closed this May 4, 2020

phw deleted the feature/riff-wave branch February 14, 2023 09:41

		raise ValueError("Invalid RIFF-chunk-ID.")


		class RiffChunkHeader():

		@@ -44,9 +44,6 @@ def is_valid_chunk_id(id):


		def assert_valid_chunk_id(id):

		assert isinstance(id, text_type)

RIFF/WAVE support, using ID3v2 #321

RIFF/WAVE support, using ID3v2 #321

Conversation

Borewit commented Aug 17, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Borewit Aug 21, 2017 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Borewit Aug 21, 2017 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Borewit Aug 21, 2017 • edited Loading

Choose a reason for hiding this comment

Borewit commented Sep 6, 2017

lazka commented Sep 7, 2017 • edited Loading

Borewit commented Mar 1, 2018 • edited Loading

Borewit commented Apr 18, 2018

phw commented Sep 11, 2018

Borewit commented Sep 11, 2018

phw commented Sep 12, 2018

phw commented Sep 12, 2018

Borewit commented Sep 12, 2018

lazka commented Sep 12, 2018

Borewit commented Sep 12, 2018 • edited Loading

phw commented Sep 13, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Borewit commented Oct 15, 2019

Borewit commented Oct 15, 2019

phw commented Oct 19, 2019

phw commented Oct 21, 2019

phw commented Oct 23, 2019

lazka commented May 4, 2020

Borewit Aug 21, 2017 •

edited

Loading

Borewit Aug 21, 2017 •

edited

Loading

Borewit Aug 21, 2017 •

edited

Loading

lazka commented Sep 7, 2017 •

edited

Loading

Borewit commented Mar 1, 2018 •

edited

Loading

Borewit commented Sep 12, 2018 •

edited

Loading