Skip to content

Latest commit

 

History

History
117 lines (87 loc) · 4.17 KB

CHANGES.md

File metadata and controls

117 lines (87 loc) · 4.17 KB
Raw

Version 0.10.1

  • revert to not throwing an exception on playback buffer underrun; instead, return -EPIPE like PCM.read() does on overrun; #131
  • type hints

Version 0.10.0

  • assorted improvements (#123 from @ossilator)
    • support for periods in the PCM constructor.
    • new functions PCM.state(), PCM.drop() and PCM.drain()
    • improved underrun/overrun handling
    • documentation improvements/consolidation (docstrings were removed in favour of online documentation)
    • more sampling rates
    • bug fixes

Version 0.9.2

  • Fix alsamixer_getvolume (#112 from @stephensp)

Version 0.9.1:

  • Support decibel, percentage, and raw volumes in getvolume, setvolume, and getrange (#109 from @chrisdiamand)

Version 0.9.0:

  • Added keyword arguments for channels, format, rate and periodsize
  • Deprecated setchannel, setformat, setrate and setperiodsize

Version 0.8.6:

  • Added four methods to the PCM class to allow users to get detailed information about the device:

    • getformats() returns a dictionary of name / value pairs, one for each of the card's supported formats - e.g. {"U8": 1, "S16_LE": 2},
    • getchannels() returns a list of the supported channel numbers, e.g. [1, 2],
    • getrates() returns supported sample rates for the device, e.g. [48000],
    • getratebounds() returns the device's official minimum and maximum supported sample rates as a tuple, e.g. (4000, 48000).

    (#82 contributed by @jdstmporter)

  • Prevent hang on close after capturing audio (#80 contributed by @daym)

Version 0.8.5:

  • Return an empty string/bytestring when read() detects an overrun. Previously the returned data was undefined (contributed by @jcea)

  • Unlimited setperiod buffer size when reading frames (contributed by @jcea)

Version 0.8.4:

  • Fix Python3 API usage broken in 0.8.3

Version 0.8.3:

  • Add DSD sample formats (contributed by @lintweaker)
  • Add Mixer.handleevents() to acknowledge events identified by select.poll (contributed by @PaulSD)
  • Add functions for listing cards and their names (contributed by @chrisdiamand)
  • Add a method for setting enums (contributed by @chrisdiamand)

Version 0.8.2:

  • fix #3 (we cannot get the revision from git for pip installs)

Version 0.8.1:

  • document changes (this file)

Version 0.8:

  • PCM() has new device and cardindex keyword arguments.

    The keyword device allows to select virtual devices, cardindex can be used to select hardware cards by index (as with mixers() and Mixer()).

    The card keyword argument is still supported, but deprecated.

    The reason for this change is that the card keyword argument guessed a device name from the card name, but this only works sometimes, and breaks opening virtual devices.

  • new function pcms() to list available PCM devices.

  • mixers() and Mixer() take an additional device keyword argument. This allows to list or open virtual devices.

  • The default behaviour of Mixer() without any arguments has changed. Now Mixer() will try to open the default Mixer instead of the Mixer that is associated with card 0.

  • fix a memory leak under Python 3.x

  • some more memory leaks in error conditions fixed.

Version 0.7:

  • fixed several memory leaks (patch 3372909), contributed by Erik Kulyk)

Version 0.6:

  • mostly reverted patch 2594366: alsapcm_setup did not do complete error checking for good reasons; some ALSA functions in alsapcm_setup may fail without rendering the device unusable

Version 0.5:

  • applied patch 2777035: Fixed setrec method in alsaaudio.c This included a mixertest with more features
  • fixed/applied patch 2594366: alsapcm_setup does not do any error checking

Version 0.4:

  • API changes: mixers() and Mixer() now take a card index instead of a card name as optional parameter.
  • Support for Python 3.0
  • Documentation converted to reStructuredText; use Sphinx instead of LaTeX.
  • added cards()
  • added PCM.close()
  • added Mixer.close()
  • added mixer.getenum()

Version 0.3:

  • wrapped blocking calls with Py_BEGIN_ALLOW_THREADS/Py_END_ALLOW_THREADS
  • added pause

Version 0.2:

  • Many bugfixes related to playback in particular
  • Module documentation in the doc subdirectory

Version 0.1:

  • Initial version