VDR developer version 2.1.6 is now available at


A 'diff' against the previous version is available at


MD5 checksums:

79519dac59166fabc2029b916bd61d00  vdr-2.1.6.tar.bz2
99f2f7094a242462696c0da5e52bb4c3  vdr-2.1.5-2.1.6.diff


This is a *developer* version. Even though *I* use it in my productive
environment, I strongly recommend that you only use it under controlled
conditions and for testing and debugging.

The changes since version 2.1.5:

- Revoked "Fixed some compiler warnings with Clang 3.4.1" from ci.c, because 
  did not compile with older versions of gcc (thanks to Sören Moch).
- Fixed keeping the current position in the Recordings menu if a recording was
  deleted in a sub folder.
- Fixed handling transfer mode on full featured DVB cards for encrypted channels
  that have no audio pid (reported by Christian Winkler).
- Fixed a possible endless loop in cH264Parser::GetGolombUe(), which caused 
  on some HD channels to get stuck and resulted in buffer overflows.
- Fixed handling PAT packets when detecting frames, so that they can be properly
  taken into account when regenerating the index of a recording.
- Fixed adding new source types in case they are already registered (reported 
by Rolf
- Removed an unnecessary assignment from cMenuRecordings::~cMenuRecordings().
- The Recordings menu now remembers the last recording the cursor was 
positioned on,
  independent of the last replayed recording. When a replay ends, however, the 
  will initially be positioned to the last replayed recording again when the 
  is opened.
- Updated the Finnish OSD texts (thanks to Antti Hartikainen).
- Fixed drawing the live indicator in the LCARS skin in case there are no 
- When checking for obsolete channels, those with an RID that is not 0 are now
  ignored (suggested by Oliver Endriss).
- The SDT is now only parsed *after* the NIT has been read, and it explicitly 
  the source value derived from the NIT. This should prevent new channels from 
  created with the wrong source.
- Added a log message in case a receiver is detached from its device because the
  assigned CAM can't decrypt the channel.
- Refactored setup parameter handling for output devices:
  + The function cDevice::GetVideoSystem() has been deprecated and will be 
    in a future version. In order to check whether a particular plugin needs to 
    modified if this function is removed, you can comment out the line
    in device.h.
  + Handling the "video (display) format" (things like 16:9, 4:3, pan&scan, 
    etc) shall now be done by the individual output devices, because the types 
    numbers of parameters are too device specific. The Setup/DVB parameters
    "Video format" and "Video display format" are still there for now and can 
be used
    by SD devices. HD devices, however, shall not use these parameters (any 
    but rather implement their own setup menu with the necessary parameters for
    controlling output.
  + The dvbhdffdevice plugin has been modified accordingly.
  + Made it clear that cDevice::SetDigitalAudioDevice() merely tells the output 
    that the current audio track is Dolby Digital. This function was only used 
by the
    original "full featured" DVB cards - do not use it for new developments!
    If an output device has several ways of replaying audio (like HDMI or 
analog jack)
    it shall implement the proper options in its plugin's SetupMenu() function.
- Added support for "Pilot", "T2-System-Id" and "SISO/MISO" parameters (thanks 
  Rolf Ahrenberg).
- Now initializing the isOnVideoDirectoryFileSystem member of cRecording when
  scanning the video directory, so that it won't cause a delay when opening the 
  on a system with a large number of recordings.
- Now resetting the isOnVideoDirectoryFileSystem member of a cRecording to -1 
  renaming it, so that it will be re-checked upon the next call to
- Added support for systemd (thanks to Christopher Reimer). To activate this you
  need to add "SDNOTIFY=1" to the 'make' call.

Have fun!


vdr mailing list

Reply via email to