VDR developer version 2.1.7 is now available at
ftp://ftp.tvdr.de/vdr/Developer/vdr-2.1.7.tar.bz2
A 'diff' against the previous version is available at
ftp://ftp.tvdr.de/vdr/Developer/vdr-2.1.6-2.1.7.diff
MD5 checksums:
1c954bad31ce74cd1cbd7987e62d2a98 vdr-2.1.7.tar.bz2 88a90327a75833b7723942d3bd25f954 vdr-2.1.6-2.1.7.diff
WARNING: ========
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 2.1.6:
- No longer logging an error message in DirSizeMB() if the given directory doesn't exist. This avoids lots of log entries in case several VDRs use the same video directory and one of them has already physically removed a recording directory, while the others still have it in their list of deleted recordings. - Updated the Italian OSD texts (thanks to Diego Pierotto). - A cCamSlot that has WantsTsData set to true in its constructor now also gets the CAT and EMM PIDs data. - Fixed a possible division by zero in frame rate detection. - VDR now reads command line options from *.conf files in /etc/vdr/conf.d (thanks to Lars Hanisch). See vdr.1 and vdr.5 for details. - Fixed a possible crash in the LCARS skin (thanks to Thomas Reufer). - Updated the dvbhddevice plugin source. - Fixed a bug in the Makefile when installing plugins with LCLBLD=1 (thanks to Stefan Huelswitt). - The pid of the PMT in which the CA descriptors of a given channel are broadcast is now stored together with the CA descriptors and can be retrieved by calling GetPmtPid() (this information is only required to receive encrypted channels with the OctopusNet receiver via the 'satip' plugin). - Channels that are not listed in the SDT are now only marked as OBSOLETE if "Setup/DVB/Update channels" is set to a value other than "no" or "PIDs only". - Fixed multiple OBSOLETE marks in channels that are not listed in the SDT in case "Setup/Miscellaneous/Show channel names with source" is set to "yes". - The new function cOsd::DrawScaledBitmap() is now used for drawing subtitles. This function can be reimplemented by high level OSDs which may be able to do the scaling in hardware or otherwise more efficiently (thanks to Thomas Reufer). - Fixed detaching receivers from devices in case a CAM needs to receive the TS (reported by Dietmar Spingler). - Fixed resetting the receiver for EMM pids for CAMs that need to receive the TS (reported by Dietmar Spingler). - Fixed (well, actually worked around) a problem with subtitles not being displayed because the broadcaster doesn't set the data's version numbers as required by the DVB standard (thanks to Rolf Ahrenberg). - Fixed support for systemd (thanks to Christopher Reimer). - Added a missing backslash to the help text of the SVDRP command MOVR (thanks to Lars Hanisch). - Added subsystem id support for DVB devices connected via USB (thanks to Jose Alberto Reguero). - Added the functions IndexOf(), InsertUnique(), AppendUnique() and RemoveElement() to the cVector class (thanks to Stefan Schallenberg). - Fixed a possible out-of-bounds access in cVector::Remove(). - Added functions to set and retrieve the priority of a cReceiver (suggested by Frank Schmirler). - Added the new parameters "Setup/Miscellaneous/Volume steps" and ".../Volume linearize" (thanks to Claus Muus). See the MANUAL for details. - Fixed jumping to an absolute position via the Red key in case replay was paused (reported by Dieter Ferdinand). - Changed the German weekday names from "MonDieMitDonFreSamSon" to "Mo.Di.Mi.Do.Fr.Sa.So." (thanks to Stefan Blochberger). - Now handling CAT sections that consist of more than one TS packet. - Added handling for DTS audio tracks to cPatPmtParser::ParsePmt() (thanks to Thomas Reufer). - Added support for PGS subtitles (thanks to Thomas Reufer). - Use of the function cOsd::GetBitmap() outside of derived classes is now deprecated, and it may be made 'protected' in a future version, since it doesn't work with TrueColor OSDs. Plugin authors may want to modify their code so that it works without this function. - Modified the descriptions of several threads, so that the important information (like device or frontend numbers) is within the first 15 characters of the string, because only these are displayed in thread listings. Plugin authors may want to do the same. - Added the channel name to log messages that reference a channel (suggested by Dietmar Spingler). - Modified the CAM API so that it is possible to implement CAMs that can be freely assigned to any devices (thanks to Jasmin Jessich). - Plugins can now implement the function SetMenuSortMode() in their skin objects derived from cSkinDisplayMenu, to get informed about the currently used sort mode, if applicable (suggested by Martin Schirrmacher). - Added cOsdProvider::OsdSizeChanged(), which plugins that implement an output device can call to signal a change in the OSD that requires a redraw of the currently displayed object (thanks to Thomas Reufer). - Added a comment to cRecorder::Activate() about the need to call Detach() in the destructor (suggested by Eike Sauer). - Now returning from removing deleted recordings after at most 10 seconds, or if the user presses a remote control key, to keep the system from getting unresponsive when removing a huge number of files (reported by Dieter Ferdinand). - Fixed generating the index file of an existing recording in case at the of a TS file there is less data in the buffer than needed by the frame detector. In such a case it was possible that frames were missed, and there was most likely a distortion when replaying that part of a recording. This is mostly a problem for recordings that consist of more than one *.ts file. Single file recordings could only lose some frames at their very end, which probably doesn't matter. At any rate, if you have generated an index file with VDR version 2.0.6, 2.1.5 or 2.1.6, you may want to do so again with this version to make sure the index is OK. - Added the new command line option --updindex, which can be used to update an incomplete index of a recording (based on a patch from Helmut Auer).
Have fun!
Klaus
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Am 18.01.2015 11:38, schrieb Klaus Schmidinger:
VDR developer version 2.1.7 is now available at
ftp://ftp.tvdr.de/vdr/Developer/vdr-2.1.7.tar.bz2
Attached patch will fix some issues with gcc-4.8 and CXXFLAGS -std=C++11 warning: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wliteral-suffix]
in dvbsubtitle.c recording.c remote.c
Tested also with gcc-4.8 without extended -std-=C++11
Thanks for the new version :)
/dev/joerg
On 18.01.2015 13:58, Joerg Bornkessel wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Am 18.01.2015 11:38, schrieb Klaus Schmidinger:
VDR developer version 2.1.7 is now available at
ftp://ftp.tvdr.de/vdr/Developer/vdr-2.1.7.tar.bz2
Attached patch will fix some issues with gcc-4.8 and CXXFLAGS -std=C++11 warning: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wliteral-suffix]
in dvbsubtitle.c recording.c remote.c
I hated that "PRId64" rubbish when it was introduced, now I hate it even more...
(This is not against you, Joerg - just a general rant ;-).
Klaus
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Am 18.01.2015 13:58, schrieb Joerg Bornkessel:
Am 18.01.2015 11:38, schrieb Klaus Schmidinger:
VDR developer version 2.1.7 is now available at
ftp://ftp.tvdr.de/vdr/Developer/vdr-2.1.7.tar.bz2
Attached patch will fix some issues with gcc-4.8 and CXXFLAGS -std=C++11 warning: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wliteral-suffix]
in dvbsubtitle.c recording.c remote.c
Tested also with gcc-4.8 without extended -std-=C++11
passed also the test with gcc-4.9.2 on x86_64 without any problems
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Am 18.01.2015 11:38, schrieb Klaus Schmidinger:
VDR developer version 2.1.7 is now available at
- VDR now reads command line options from *.conf files in
/etc/vdr/conf.d (thanks to Lars Hanisch). See vdr.1 and vdr.5 for details.
Is there an example what exactly can we do with this? The Manpage's give there very small information :( Lars?
On 01/18/2015 02:43 PM, Joerg Bornkessel wrote:
- VDR now reads command line options from *.conf files in
/etc/vdr/conf.d (thanks to Lars Hanisch). See vdr.1 and vdr.5 for details.
Is there an example what exactly can we do with this? The Manpage's give there very small information :( Lars?
Have a look at the vdr4arch repository: https://github.com/VDR4Arch/vdr4arch
We already used this feature using the patch created by Lars.
This new feature is especially useful when using systemd or other event driven init systems. It now no longer is required to "somehow" construct a command line. Just start VDR
https://github.com/VDR4Arch/vdr4arch/blob/master/vdr/vdr.service
The VDR settings can be configured in a file like this one: https://github.com/VDR4Arch/vdr4arch/blob/master/vdr/10-vdr.conf
And to load/configure a plugin the config file has to look like this:
https://github.com/VDR4Arch/vdr4arch/blob/master/plugins/vdr-softhddevice/50...
vdr4arch places the plugin config files to /etc/vdr/conf.available first. It's the user's job to create a symlink to /etc/vdr/conf.d to enable the plugin or delete the symlink to disable a plugin.
Manuel
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Am 18.01.2015 15:11, schrieb Manuel Reimer:
On 01/18/2015 02:43 PM, Joerg Bornkessel wrote:
- VDR now reads command line options from *.conf files in
/etc/vdr/conf.d (thanks to Lars Hanisch). See vdr.1 and vdr.5 for details.
Is there an example what exactly can we do with this? The Manpage's give there very small information :( Lars?
Have a look at the vdr4arch repository: https://github.com/VDR4Arch/vdr4arch
We already used this feature using the patch created by Lars.
This new feature is especially useful when using systemd or other event driven init systems. It now no longer is required to "somehow" construct a command line. Just start VDR
Oh Oh, systemd crap...
vdr4arch places the plugin config files to /etc/vdr/conf.available first. It's the user's job to create a symlink to /etc/vdr/conf.d to enable the plugin or delete the symlink to disable a plugin.
this looks like, as has the user a lot of activity to do in my opinion, prevent the end user from a lot of editing some files or symlinking etc. anyway, systemd is not my working place and i will see what ideas comes from Lucian M. He is the manager of the systemd crap part in the gentoo-vdr-scripts
Thx for your reply
Hi,
Am 18.01.2015 um 15:40 schrieb Joerg Bornkessel:
Am 18.01.2015 15:11, schrieb Manuel Reimer:
On 01/18/2015 02:43 PM, Joerg Bornkessel wrote:
- VDR now reads command line options from *.conf files in
/etc/vdr/conf.d (thanks to Lars Hanisch). See vdr.1 and vdr.5 for details.
Is there an example what exactly can we do with this? The Manpage's give there very small information :( Lars?
Yes, I should have documented that feature a bit... My fault, will try to do in the next days.
Have a look at the vdr4arch repository: https://github.com/VDR4Arch/vdr4arch
We already used this feature using the patch created by Lars.
This new feature is especially useful when using systemd or other event driven init systems. It now no longer is required to "somehow" construct a command line. Just start VDR
Oh Oh, systemd crap...
It's not only systemd. In fact, it hasn't anything to do with systemd. But it helps to strip down the various init scripts, regardless if systemd, Upstart or SysV.
And of course everything works with the old scripts. :)
vdr4arch places the plugin config files to /etc/vdr/conf.available first. It's the user's job to create a symlink to /etc/vdr/conf.d to enable the plugin or delete the symlink to disable a plugin.
this looks like, as has the user a lot of activity to do in my opinion, prevent the end user from a lot of editing some files or symlinking etc. anyway, systemd is not my working place and i will see what ideas comes from Lucian M. He is the manager of the systemd crap part in the gentoo-vdr-scripts
It helps distributors for preconfiguring the vdr and its plugins, but also you can created multiple conf.d directories and symlink the one or the other to /etc/vdr/conf.d to test different configuration without messing with the old and working config.
Just play a bit with it, you may find it helpful. "vdr --showargs[=DIR]" will output all options from /etc/vdr/conf.d or the given directory. You might even use this to generate lines for "old style" init scripts with all options on the commandline.
Lars.
Thx for your reply
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
-
Joerg Bornkessel -
Klaus Schmidinger -
Lars Hanisch -
Manuel Reimer