{5} Assigned, Active Tickets by Owner (Full Description) (71 matches)

List tickets assigned, group by ticket owner. This report demonstrates the use of full-row display.

(empty) (2 matches)

Ticket Summary Component Milestone Type Created
#9778 Video artifacts when playing recordings made with "Low Quality" profile MythTV - Video Playback unknown Bug Report - General May 15, 2011

[Although this is specific to a recording profile, I'm filing this under "Video Playback" because that is the part that has changed... recordings made with 0.21 behave the same as recordings from 0.23-fixes] [Also, the mixed versioning is because I don't currently have the 0.24-fixes test system up and running..]

MythTV versions from 0.22 onward display graphical artifacts (including the top fourth of the screen being green) when playing recordings made with the "Low Quality" recording profile (with a PVR-150). The same recordings play fine in 0.21 as well as other players (including mplayer, VLC, and the Windows MythTV Player client). An example is currently here: http://organizedinsanity.net/MythTV/green.png

  • Different playback profiles/filters make no difference.
  • I've tried four different computers (via Mythbuntu LiveCD) to rule out hardware incompatibility.
  • "--verbose playback" didn't seem to indicate anything unusual. I can try it again and attach the log if that's of any interest.
  • Behavior is identical whether streaming from the remote backend or playing the files locally via MythVideo?.
  • The commercial flagging process crashes for these recordings (again, 0.21 was fine). That's a more difficult thing to experiment with and investigate, so I only mention it because it may share the same root issue. I'll attach logging for this just in case.

#8066 new keybinding for jump in mythmusic Plugin - MythMusic unknown Patch - Feature Feb 13, 2010

Because seeking in mythmusic is cumbersome if you have longer dj-mixes and can only go forward/back for five seconds, i have made this patch against rev. 23546. With this it's possible to jump forward/back for 60 seconds every keypress. There is a new keybinding like in video playback (JUMPFFWD,JUMPRWND). Don't know it it is a dirty hack, but it would be nice if this would be incorporated in mythtv.

beirdo (5 matches)

Ticket Summary Component Milestone Type Created
#10868 Nuvexport is failing on an uninitialised value error Apps - Nuvexport unknown Bug Report - Crash Jun 27, 2012

Running Nuvexport .25 fixes on a Linux Mint 13 Mate 64 bit server.

Installed Nuvexport from the .25 fixes branch.

While trying to process any files getting the following error.

Use of uninitialized value $info{"video_type"} in string eq at /usr/share/perl5/MythTV/Recording.pm line 327, <STDIN> line 2. Stream type 'lavfpref' is not an mpeg, and will not work with this program.

Cleaning up temp files.

I'm not a perl coder but looking at the code it looked to me like it could be the output produced by mplayer as in the appropriate line it outputs MPG2 not MPEG2

[lavf] stream 2: audio (mp3), -aid 1, -alang eng ID_SUBTITLE_ID=0 ID_SID_0_LANG=eng [lavf] stream 3: subtitle (dvbsub), -sid 0, -slang eng LAVF: Program 1 PROGRAM_ID=1 LAVF: 2 audio and 1 video streams found LAVF: build 3478528 VIDEO: [MPG2] 704x576 0bpp 25.000 fps 15000.0 kbps (1831.1 kbyte/s) [V] filefmt:42 fourcc:0x3247504D size:704x576 fps:25.000 ftime:=0.0400 Load subtitles in /mnt/mythfifofum/

Whereas the perl code at that line is looking for MPEG[12]

# First, we check for the existence of an mpeg info program

my $program = find_program('mplayer');

# Nothing found? Die

die "You need mplayer to use this script on mpeg-based files.\n\n" unless ($program);

# Grab the info we want from mplayer (go uber-verbose to override --really-quiet)

my $idargs = "-v -v -v -v -nolirc -nojoystick -vo null -ao null -frames 0 -identify"; my $data = $program $idargs '$file' 2>/dev/null; study $data; ($info{'video_type'}) = $data =~ m/VIDEO:?\s*(MPEG[12]|H264)/m; ($info{'width'}) = $data =~ m/ID_VIDEO_WIDTH=0*([1-9]\d*)/m; ($info{'height'}) = $data =~ m/ID_VIDEO_HEIGHT=0*([1-9]\d*)/m; ($info{'fps'}) = $data =~ m/ID_VIDEO_FPS=0*([1-9]\d*(?:\.\d+)?)/m; ($info{'audio_sample_rate'}) = $data =~ m/ID_AUDIO_RATE=0*([1-9]\d*)/m; ($info{'audio_bitrate'}) = $data =~ m/ID_AUDIO_BITRATE=0*([1-9]\d*)/m; ($info{'audio_bits_per_sample'}) = $data =~ m/AUDIO:.+?ch,\s*[su](8|16)/mi; ($info{'audio_channels'}) = $data =~ m/ID_AUDIO_NCH=0*([1-9]\d*)/m;

So I modified the script to look for MPG2 rather then MPEG[12] and still get the same error.

#8575 Transcoding to *.nuv truncates end of file MythTV - Mythtranscode unknown Bug Report - General Jun 17, 2010

I record a movie, mark the beginning and the end in the cutlist and then transcode to *.nuv. everything is done in the gui, no command line. the result is too much truncated (the end of the movie is missing). i think the longer the movie, the more is missing. this happens in 0.23 and 0.24. if you need more infos please let me know.

#10509 mythtranscode --showprogress does show out in stdout MythTV - Mythtranscode unknown Bug Report - General Mar 25, 2012

mythtranscode --showprogress no longer show transcode progress in sysout. Eg. mythtranscode --mpeg2 --honorcutlist --showprogress --outfile test.mpg --chanid 1191 --starttime 20120318160000

The above example does not show the status in stdout. Only the current time and version of mythtranscode is displayed in stdout. 2012-03-25 19:37:57.822853 C mythtranscode version: [v0.25pre] www.mythtv.org

Running myth from git v0.25-rc-54-g78ad8be

#10938 Logo detection not working MythTV - Mythcommflag unknown Patch - Bug Fix Jul 25, 2012

I set the commercial detection to only detect Logos. This works best for german TV Station as long as they a broadcasting HDTV (DVB-S2/H.264). For PAL-SD recordings (DVB-S/MPEG-2) mythcommflag returns successfully but always with 0 commercial breaks.

I would be happy to provide traces if you tell me how this is done in mythcommflag ...

#10286 Weather world map for mythweather Plugin - MythWeather unknown Patch - Feature Jan 25, 2012

I've created a patch, which extends MythWeather? with a screen, which displays a world map with weather of different locations (actuall only one can be defined). As a map it uses the NASA blue marble maps. Since both maps are a little bit bigger files, the patch containing both maps is also a little bit bigger (about 40MB). Therefore I've hosted the patch at http://ftp.engsas.de/mythweather-worldweathermap.patch.

The screen is not finished yet, since two questions are not solved.

  1. How can multiple locations can be assigned to one screen?
  2. How can weather locations be mapped to geo coordinates more reliable? (Actually it works, if the weather location is something like CITY, COUNTRY (e.g. London, Great Britain)

Nevertheless I'll hope some others find that idea nice and want to help a bit.

Peter Bennett (8 matches)

Ticket Summary Component Milestone Type Created
#12010 mythcommflag --rebuild does not re-create good seektables with h264 recordings MythTV - Video Decoding unknown Bug Report - General Jan 10, 2014

This looks like a re-run of #11435 and #6243. Initial report here, quoted below.


I have just found that for recordings from DVB-T2 FreeviewHD in the UK, (h264), a listing of the markup data generated after mythcommflag --rebuild contains only a small subset of the lines in a similar listing made immediately after recording.

In the 'original' table, keyframe separation is typically around 24 frames. In the 'rebuilt' table the spacing is variable, sometimes in the hundreds of frames. Points listed there are also in the original table.

hexdump shows that all keyframes in the original list for this recording begin 47 40 65 3x, and to me look genuine. The problem does not affect DVB-T SD recordings, which are mpeg2 format.

In my tests I have usually been working with recordings having seektables that have been rebuilt, often several times. This may explain why editing has sometimes been difficult.

mythutil --chanid 1102 --starttime 20140109182600 --getmarkup Haworth_orig.xml mythcommflag --rebuild --chanid 1102 --starttime 20140109182600 mythutil --chanid 1102 --starttime 20140109182600 --getmarkup Haworth_rebuilt.xml

ls -l Haw* -rw-rw-r--. 1 john john 290708 Jan 9 19:13 Haworth_orig.xml -rw-rw-r--. 1 john john 73286 Jan 9 19:16 Haworth_rebuilt.xml

#12907 Shaking while picture panning on Raspberry Pi Ports - rPi unknown Bug Report - General Oct 23, 2016

In sports programs where there is rapid panning of the picture, the video sometimes shakes on a Raspberry Pi 3. The same video plays perfectly smoothly on an amd64 with VDPAU.

#12993 HEVC/H265 Parser missed for Recordings MythTV - Video Playback unknown Bug Report - General Feb 7, 2017

Because there exist no HEVC/H265 parser inside libmythtv (called in DTVRecorder::ProcessVideoTSPacket) there are no information about the video quality in the recording view.

For German DVB-T2 HD this could be fixed by an easy fix (set height to 1080, width to 1920 and frameRate to 50). When DVB-T2 HD the only HEVC source it might be OK.

But at the end a new module is needed similar to mythtv/libs/libmythtv/mpeg/H264Parser.cpp.

#13124 Live TV Using wrong tuner MythTV - Recording needs_triage Bug Report - General Sep 17, 2017

I have 5 tuners, with sequence for Schedule set as 12345 and sequence for LiveTV set as 54321 respectively. Live TV should select the fifth tuner by prefernence, then the 4th etc. If I enter Live TV from the program schedule "Watch this channel" option it selected the first tuner instead of the last. Entering via Live TV on the menu selected the correct tuner.

Work around - after setting the first three tuners to 0 for Live TV sequence, (i.e. 00021) it correctly selected the 5th tuner.

I don't know what it could be doing, if it was using the schedule order or the tuner sequence, after setting the first three to 0 it would have selected the 4th.

#13185 UI Issues with macOS High Sierra Ports - OSX needs_triage Bug Report - General Dec 1, 2017

A MacPorts? user reported various problems with MythTV on macOS 10.13 ("High Sierra"). I've set up a High Sierra VM and installed myth 0.28.2 on it. There are several major UI problems:

1) mythtv-setup shows a blank screen on startup. Clicking anywhere allows the main menu show. It is then easy to trigger the problem where screens stop responding to the keyboard—have to click with a mouse to get focus back For example, adding a new capture card or even just entering the Video Source Setup screen. See #12857 for more description.

2) mythfrontend - shows a menu when it should be in full screen mode. I'll attach a screen grab.

3) mythfrontend - stops responding to the keyboard in certain spots. For example if I enter the Setup > General screen. As above for mythtv-setup, have to click to get focus back.

4) mythfrontend — Dock is properly hidden when starting this app. (The Dock is the macOS application launcher that, by default, floats above other windows.) However, the Dock does not reappear when after exiting mythfrontend. Have to toggle a setting in System Preferences to get the Dock to reappear.

#1 and #3 mean that Myth can no longer be used via a 'ten-foot interface'. It is easy to get into situations where one has to click a mouse to regain control of the application.

#2, #3, and #4 only occur on High Sierra. Need to test #1 under other OS versions to be sure. Myth was built with Qt 5.9.2--version info attached.

#13223 ffmpeg 3.4 and VDPAU problem with some h264 videos MythTV - Video Playback 30.0 Bug Report - General Feb 13, 2018

Certain h264 mkv videos will not play on MythTV with the 3.4.1 ffmpeg version newly incorporated, when using VDPAU decoding. The log displays thousands of errors

2018-02-13 17:41:10.673987 E  AFD: video decode error: Invalid data found when processing input (0)

These videos play correctly with all other setups: Xvideo, OpenGL, Vaapi, OpenMax as well as Standard decoding coupled with VDPAU rendering.

Videos that fail

#13242 Seektable for video can be created but is not recognised MythTV - Mythcommflag needs_triage Bug Report - General Mar 10, 2018
[john@HP_Fed ~]$ mythcommflag --rebuild --video  /mnt/sdb1/Vids/21664_20180309155800_6619.ts
2018-03-10 13:21:06.353216 C  mythcommflag version: master [v30-Pre-v30-Pre-515-ge7173e3] www.mythtv.org
2018-03-10 13:21:06.353263 C  Qt version: compile: 5.9.4, runtime: 5.9.4
MythTV Commercial Flagger, building seek table for:
Rebuild started at Sat Mar 10 13:21:07 2018
2018-03-10 13:21:30.474475 E  decoding error End of file (-541478725)
Rebuild completed at Sat Mar 10 13:21:30 2018
[john@HP_Fed ~]$ 

... but on playing the video and entering the editor it reports "No seektable"

Ticket #10804 refers.

mythutil/commandlineparser.cpp refers to a Video Gallery name, and there is a --parse-video-filename option in mythutil. My filename-fixated approach may conflict with the media-metadata approach, not for the first time, but the log above implies that the intended file was scanned. (It was created by an experimental mythTScut.sh which I have appended to the mythDVBcut entry in the wiki)

#13253 AMD vdpau distorted playback after ffmpeg replace deprecated avcodec_decode_video2 MythTV - Video Playback needs_triage Bug Report - General Mar 26, 2018

After https://github.com/MythTV/mythtv/commit/2fc0b3779e1524d5273165ad9fab48c12f3bd7f3 users of AMD platform using vdpau have distorted picture after playback jump forward/back.

Pls look at video: http://warped.inet2.org/sample6.zip

In sample6, at 00:10 there was was first 30s jump forward. At 00:14 was second jump and after that playback becomes distorted.

Issue is seen only on AMD with mesa provided VDPAU. Nvidia/VDPAU and AMD/ffmpeg plays OK. Version of MESA doesn't matter (tested 17.1.10 and 17.3.6)

cpinkham (4 matches)

Ticket Summary Component Milestone Type Created
#10700 RFE: recording failed system event MythTV - General unknown Bug Report - General May 7, 2012

I had a number of recordings fail due to a faulty tuner as such:

May  6 19:03:01 pvr mythbackend[2119]: I Scheduler scheduler.cpp:2505 (HandleTuning) Canceling recording since tuning timeout, 180 seconds, has been exceeded.
May  6 19:03:01 pvr mythbackend[2119]: I Scheduler scheduler.cpp:2459 (HandleRecordingStatusChange) Canceled recording (Recorder Failed): "Once Upon a Time": channel 3129 on cardid 6, sourceid 2

My "recording finished" system event didn't get fired for these though, I'd guess due to the fact that they failed.

I'd like to be able to hook on a failed recording event though.

#11387 Mythtranscode Deleting Aspect Ratio Data MythTV - Mythtranscode unknown Bug Report - General Jan 30, 2013

I've been running mythtranscode to losslessly remove commercials from my mpeg2 recordings. It appears that all data for the recording is being deleted from recordedmarkup when mythtranscode is run.

I noticed this because the aspect ratio of recordings played in mythweb in flash after transcoding are defaulting to 4:3 aspect ratio. By looking through the source code for mythweb, I noticed that this was caused by the recordedmarkup being deleted from the database. I believe this is occurring when it is transcoded.

#10595 Video Setup Wizard can hang if sample videos can not be downloaded. MythTV - General unknown Bug Report - Hang/Deadlock Apr 13, 2012

I tried the setup wizard. Downloading the Sd or HD sample file hangs the frontend. Found in 0.25, same in master from now. One thing is the failing download, the other thing is the hanging frontend.

#9435 mythcommflag requires full path of mythvideo files to be enterred to work correctly. MythTV - Mythcommflag unknown Patch - Feature Jan 4, 2011


If mythvideo called from /, and storage group is /MythVideo/Videos?, and filename is Video.mpg. Filename used for saving filemarkup entries is based on Video.mpg instead of /MythVideo/Videos/Video?.mpg. Since the storage group / doesn't exist, Video.mpg is saved.

I have attached a patch to fix this.

dblain (1 match)

Ticket Summary Component Milestone Type Created
#11075 services API returns errors as XML no matter what you ask for MythTV - UPnP unknown Bug Report - General Sep 9, 2012
curl --header accept:application/json 'http://localhost:6544/Dvr/GetRecordeengList'


<?xml version="1.0" encoding="utf-8"?>
  <errorDescription>Invalid Action</errorDescription>

that ain't no json I ever heard of!

Karl Egly (3 matches)

Ticket Summary Component Milestone Type Created
#12330 channel scan ignores subsequent channels with same generated callsign MythTV - DVB 29.2 Bug Report - General Dec 4, 2014

In the UK on Freesat DVB-S there are channels which have different service IDs but the same name (what the code calls serviceShortName). The logic in in libs/libmythtv/channelscan/channelscan_sm.cpp sets the callsign, which mythtv requires to be unique, to the serviceShortName, therefore further channels with the same name are ignored and not picked up in the scan.

The attached patch demonstrates this by changing the callsign to be the catenation of the serviceShortName with service id.

Eg on transport 2047 there are two channels called "BBC ALBA"

MariaDB [mythconverg]> select transportid,frequency,polarity,symbolrate,mplexid from dtv_multiplex where transportid=2047; +-------------+-----------+----------+------------+---------+ | transportid | frequency | polarity | symbolrate | mplexid | +-------------+-----------+----------+------------+---------+ | 2047 | 10802750 | h | 22000000 | 973 | +-------------+-----------+----------+------------+---------+


MariaDB [mythconverg]> select callsign,serviceid,mplexid from channel where mplexid=973; +----------------+-----------+---------+ | callsign | serviceid | mplexid | +----------------+-----------+---------+ | ETV2 | 6407 | 973 | | BBC One Scot | 6421 | 973 | | BBC Two Scot | 6422 | 973 | | BBC ALBA | 6423 | 973 | | BBC One N West | 6441 | 973 | | BBC One Yorks | 6451 | 973 | | BBC One S East | 6461 | 973 | | BBC One NE&C | 6471 | 973 | +----------------+-----------+---------+ 8 rows in set (0.00 sec)


MariaDB [mythconverg]> select callsign,serviceid,mplexid from channel where mplexid=973; +---------------------+-----------+---------+ | callsign | serviceid | mplexid | +---------------------+-----------+---------+ | ETV2-6407 | 6407 | 973 | | BBC One Scot-6421 | 6421 | 973 | | BBC Two Scot-6422 | 6422 | 973 | | BBC ALBA-6423 | 6423 | 973 | | BBC ALBA-6424 | 6424 | 973 | | BBC One N West-6441 | 6441 | 973 | | BBC One Yorks-6451 | 6451 | 973 | | BBC One S East-6461 | 6461 | 973 | | BBC One NE&C-6471 | 6471 | 973 | +---------------------+-----------+---------+ 9 rows in set (0.00 sec)

I'm not suggesting this patch be merged, just providing it so as to explain what the problem is.

This is an issue for me becasue the tables which map bouquets and regions to logical channel numbers often seem to reference serviceids which mythtv doesn't have in the channels table because they are the 2nd or subsequent occurances of serviceShortName the scanner has encountered.



#12662 EITFixup Glashart-media DVB-C NL MythTV - EIT 29.2 Bug Report - General Feb 22, 2016

EIT on DVB-C from Glashart-media in NL is not correctly showing special characters, like umlaut etc.

See attached dvbsnoop output.

Probably a EITfixup required.

#12012 Abbreviated show names don't work for Title MythTV - Video Library unknown Patch - Bug Fix Jan 12, 2014

When parsing the filename to create the title for the video, the system removes all dots (periods), however, when there are multiple single characters these are likely an abbreviation. Examples

House M.D G.I Joe

The issue is that when a lookup is done against tmdb, thetvdb, there is no match as what is searched is "House M D" or "G I Joe".

There are 2 ways to solve this. In the metadata Lookup, or in the logic to parse the filename. Metadata lookup would cover more areas, however, at this point, the user may have altered the name, so we should lookup exactly what they have asked.

Pull will be attached shortly.

gigem (2 matches)

Ticket Summary Component Milestone Type Created
#13358 Revert Some More Suspect OpenGL Commits MythTV - General needs_triage Bug Report - General Dec 10, 2018

Following up on commit [23b59d73], some other commits were identified by Mark Kendall as suspect and good candidates for reversion. This ticket is intended to track those changes. The main commits to be considered are these:

[59acf5c1] - OpenGL: Add GLSL deinterlace filters that accept YV12 input textures

[364e09ba] - UI: cache OpenGL shader parameters to reduce render time

[2b01ed38] - Video output: Fix OpenGL 1 output after b9971146

and maybe

[b9971146] - Video output: Add default fragment shader for OpenGL 2

#10682 Add excluded_cards to GET_FREE_RECORDER_LIST protocol command MythTV - General 29.2 Developer Task May 2, 2012

Currently we have to make many individual backend queries to work around this situation.

Also, change vector<> to QVector<> in this area of code.

See http://code.mythtv.org/cgit/mythtv/commit/?id=dc6a18af542c9de80f06cc22485d6ac8b932dfeb

jpoet (2 matches)

Ticket Summary Component Milestone Type Created
#13321 Seg Fault when upgrading Steppes Theme MythTV - User Interface Library needs_triage Bug Report - Crash Sep 19, 2018

When upgrading to Steppes theme 2.40, mythfrontend fails with a seg fault. After restart it is ok.

#13187 HLSPlaylist doesn't follow 302 redirects correctly MythTV - HTTP Streaming needs_triage Bug Report - General Dec 4, 2017

When using HLS to IPTV stream NZ channels the base URL for the playlist re-directs to the correct location via a 302.

This re-direct can't be permanent as the streaming URL changes over time.

The redirected playlist provides the correct TS streams but HLSStream tries to use the original base playlist domain rather than the domain provided via re-direction.

JYA (3 matches)

Ticket Summary Component Milestone Type Created
#12295 LiveTV stalling and exiting at programme transitions after [4ff7c1fca] MythTV - Video Playback unknown Bug Report - General Oct 12, 2014

Running Gentoo, installing using ebuilds. When watching liveTV all versions newer than 317d5b7 keep stalling briefly every now and then (Waited XXXms for video buffers in log) and when one show ends and another one starts liveTV exits to main menu. This seems to be frontend problem only, my backend is updated to latest and works no problem.

#12805 Plugin MythMusic crashes on a specific WMA file Plugin - MythMusic unknown Bug Report - Crash Jun 6, 2016

Playing one of my WMA files causes a segfault in mythfrontend after 5 seconds of playing. Nothing obviously wrong I can spot with the file. Pressing I on the file in mythMusic brings up all the metadata correctly.

I suspect the embedded album art.

Last time I had a problem with one file https://code.mythtv.org/trac/ticket/12403 the fantastic Paul Harrison fixed it very quickly once he had the file in hand. So rather than attaching a trace, I'm linking to the file.


This file has caused crashes for a long time. So it's not due to a recent change. It's taken me a while to track down as the place in the playlist is forgotten so I had to catch it happening.

#10370 DVD playback fail with default settings on MacOS X Ports - OSX unknown Bug Report - General Feb 22, 2012

Using MacOS 10.7 (or 10.6)

MythFrontend will fail to play any DVD

When selecting the play DVD option in the Optical Menu ; in the log you get:

libdvdread: Using libdvdcss version 1.2.9 for DVD access
libdvdread: Can't stat /dev/r/dev/disk1
No such file or directory
libdvdread: Could not open /dev/r/dev/disk1
libdvdnav: vm: failed to open/read the DVD
2012-02-22 21:33:51.365260 I  MediaMonitor::selectDrivePopup(Select a DVD drive) - No suitable devices
2012-02-22 21:33:51.366886 I  DVDInfo: Trying dvd://dev/r/dev/disk1
2012-02-22 21:33:51.369193 E  DVDInfo: Failed to open device at /dev/r/dev/disk1
2012-02-22 21:33:51.369200 I  DVDInfo: Finishing.

programs/mythfrontend/main.cpp automatically prefix the device name with /dev/r

Yet, the default device name it isn't defined by the user is /dev/disk1 (defined in libs/libmyth/mediamonitor_darwin.h

The documentation in the settings statea: "This device must exist, and the user playing the DVD needs to have read permission on the device. 'default' will let the MediaMonitor? choose a device"

using "default" does nothing, it uses /dev/disk1 unless defined in the settings, and default is /dev/disk1 which myth try to play as /dev/r/dev/disk1 which will *always* be an invalid device

The log also shows: MediaMonitor::selectDrivePopup(Select a DVD drive) - No suitable devices

No such popup ever appears...

MythFrontend should either automatically determine what the right DVD device is, not prepend /dev/r to any path name or properly display the selectDrivePopup

Nicolas Riendeau (2 matches)

Ticket Summary Component Milestone Type Created
#11763 MythWeb doesn't use UTF-8 aware regex and string functions Plugin - MythWeb unknown Bug Report - General Aug 19, 2013

MythWeb doesn't currently use UTF-8 aware functions to process the program category file, they would have to be replaced by their multibyte equivalent.

In at least some distributions PHP support for multibyte strings is in a separate package which would introduce a new dependency.

#12997 [PATCH] Country name correction MythTV - General unknown Patch - Bug Fix Feb 10, 2017

Country name correction for future localization tasks (zh_TW).


Reference (if anyone is interested): https://en.wikipedia.org/wiki/Taiwan

David Hampton (4 matches)

Ticket Summary Component Milestone Type Created
#13344 C++11 code modernizations MythTV - General 30.0 Developer Task Oct 31, 2018

Update MythTV to take advangate of features included in the C++11 standard.

  1. Use the boolean literals true/false instead casting the numbers 1/0 from integer to boolean.
  2. Use the nullptr keyword instead of NULL/0.
  3. Use the 'override' keyword on all subclassed functions that override a parent/ancestor function.
  4. Convert from std::random_shuffle to std::shuffle. The former has been removed from the C++17 standard.
  5. Use the =default declaration where possible, which allows the compiler to make additional optimizations.
  6. Use the =delete declaration for unimplemented private functions.
  7. Use the new list emplace_back function to eliminate creating temporary objects.
  8. Remove the OVERRIDE/FINAL macros that are no longer needed.
  9. Replace deprecated C header files with their C++ equivalents.
  10. Remove unused instances of the 'using' keyword.
  11. Update some functions with default member initialization.

#12298 Sorting improperly assuming an Article MythTV - General 30.0 Bug Report - General Oct 14, 2014

A new show started this season called "A to Z" (http://thetvdb.com/?tab=series&id=281588&lid=7)

In the watch recordings screen, it assumes "A" is an article, and not part of the actual title, and therefore this show is sorted as "to Z, A" - which can bury it in a large library of recordings.

#13164 Changes to fix cppcheck warnings MythTV - General 30.0 Developer Task Oct 31, 2017

Track changes committed to cleanup warnings from cppcheck.

#13026 Clean up technical debt from Qt4 changes MythTV - General 30.0 Patch - Bug Fix Apr 15, 2017

Changes from Qt3 to Qt4 removed the need to explicitly call the detach or copy functions on qt classes. These will be called internally as necessary. See http://doc.qt.io/qt-4.8/porting4.html

Mark Spieth (1 match)

Ticket Summary Component Milestone Type Created
#12638 Switch to using DVB APIv5 MythTV - DVB 30.0 Developer Task Feb 4, 2016

DVB API v3 is depreciated, and v5 should be used with any kernel from 3.3 onwards

See https://linuxtv.org/docs/libdvbv5/index.html

sphery (8 matches)

Ticket Summary Component Milestone Type Created
#9954 Bad channel crash loop - Problem since v 0.18 MythTV - General unknown Bug Report - General Jul 29, 2011

Minor Problem: Channel scan finds 100s of invalid, non-existent channels. Tuning to an invalid channel in LiveTV crashes to main menu.

Major problem: No fast way to recover. Fastest way is to close down, run backend setup and manually change the starting channel. No big deal if it's one channel, one time. 5 minutes or so to fix. Major problem is that if you are trying to label 500 channels, 200 crashes are going to add an extra 16 hours to the time it takes.

I regret to report that this has been a problem since at least version 0.18 (see link). I have seen lots of people asking for this to be fixed.


Forget about fixing the "crash when the data is bad" problem.

The real problem (can't quickly recover from crash) is an incredibly simple problem to make a good workaround. Please accept my suggestion for one or more of the following seven ways to workaround it.

1) On crash, return to last working channel 2) On crash, return to a pre-set channel (e.g., channel 2_1) 3) On crash, tune to analog tuner (for those of us with analog and digital tuners) 4) Option for a "what channel do you want" to pop up when selecting live TV, before trying to turn to TCC (the Crash Channel). 5) An option to have a fixed starting channel rather than a variable one. 6) A simple command line program to change to another startup channel while frontend is running. 7) Let us tune to a channel from the channel guide in "schedule recordings" instead of only live TV (which crashes before we can get there). 8) When it crashes, offer the option to delete the channel then and there (then tune to default channel when starting).

In short, if you can't stop the thing from crashing on particular channels, please provide some very fast way to get away from a bad channel. I would be happy to tab to a terminal, then re-run a command line every time it crashes.

If it matters, I have HDhomerun. Oh, and this isn't the SD/HD bug. It happens regardless of what channel you start from, only the destination channel matters.

So could somebody please, please fix this so that those of us whose cable company shuffles the channels from time to time don't have to spend days weeding out the bad ones?

#11004 Master backend tries to record to non-existent directory MythTV - Recording unknown Bug Report - General Aug 15, 2012

I've had a persistent issue for about a year where the master backend will try to record to a directory listed in the storage group that isn't present on that system. The only way I've been able to get the system to work is to mount the directory using NFS, however I would prefer to have it work without doing that.

I have the storage group defined on the master backend with the following:


The first three directories are on the master backend and the last two are on the slave backend. Nothing is defined on the slave backend. My understanding is that the master backend will realize that the last two directories do not exist on that system and not use them, however, on my system, it repeatedly tries to anyway.

I have tried removing all storage group information and re-entering it with no difference.

I posted my issue on the mythtv-users mailing list and Michael Dean asked me to provide some details which I have included below.

master backend:

bob@mythbox:~$ ls -ld /var/{,mythtv/}tv*/recordings
ls: cannot access /var/tv*/recordings: No such file or directory
drwxrwxrwx 2 bob bob  24576 2012-07-19 22:33 /var/mythtv/tv2/recordings
drwxrwxrwx 2 bob bob 126976 2012-07-24 11:43 /var/mythtv/tv3/recordings
drwxrwxrwx 2 bob bob 135168 2012-07-24 12:39 /var/mythtv/tv/recordings

bob@mythbox:~$ df -k /var/{,mythtv/}tv*/recordings
df: `/var/tv*/recordings': No such file or directory
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/sda2            709363588 692880936  16482652  98% /var/mythtv
/dev/sdb1            1953446464 1896418576  57027888  98% /var/mythtv/tv3
/dev/sdc1            1952558140 1895880720  56677420  98% /var/mythtv/tv

slave backend:

bob@mail:~$ ls -ld /var/{,mythtv/}tv*/recordings
ls: cannot access /var/mythtv/tv*/recordings: No such file or directory
drwxrwxrwx 2 bob bob 49152 2012-07-24 12:39 /var/tv2/recordings
drwxrwxrwx 2 bob bob 20480 2012-07-16 18:14 /var/tv/recordings

bob@mail:~$ df -k /var/{,mythtv/}tv*/recordings
df: `/var/mythtv/tv*/recordings': No such file or directory
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/sdb1            976694464 919833692  56860772  95% /var/tv2
/dev/sda2            200184560 183212220  16972340  92% /var/tv

Here is the storage group table:

| id | groupname   | hostname | dirname                          |
|  9 | Default     | mythbox  | /var/mythtv/tv/recordings/       |
| 12 | Default     | mythbox  | /var/tv/recordings/              |
| 10 | Default     | mythbox  | /var/mythtv/tv2/recordings/      |
| 11 | Default     | mythbox  | /var/mythtv/tv3/recordings/      |
| 14 | Default     | mythbox  | /var/tv2/recordings/             |
| 15 | Videos      | mythbox  | /var/mythtv/videos/              |
| 16 | Coverart    | mythbox  | /var/mythtv/artwork/coverart/    |
| 17 | Fanart      | mythbox  | /var/mythtv/artwork/fanart/      |
| 18 | Screenshots | mythbox  | /var/mythtv/artwork/screenshots/ |
| 19 | Banners     | mythbox  | /var/mythtv/artwork/banners/     |

He requested that I run the backend with the parameters -v file,schedule --loglevel debug. The error appeared pretty quickly. I have attached the log file.

I cut off the log file right after the error appeared. Please let me know if you need any additional information.

#11509 mythfilldatabase pulls incorrect channel information from Schedules Direct MythTV - General unknown Bug Report - General Apr 26, 2013

I was having this issue on version 0.25 and recently upgraded to 0.26.0_p2013022 (Gentoo) in hopes of fixing it, but no luck. I am using the HDHR Prime with Verizon FIOS and Schedules Direct. My issue is I have missing channels, duplicate channels and channels with the wrong channel numbers.

I tried re-adding the video sources after I upgraded. Did a mythfilldatabase and grabbed all the channels from Schedules Direct.

The issue is 100% reproducible.

#8701 OSD timeouts need revisiting MythTV - User Interface Library unknown Developer Task Jul 28, 2010

Most OSD displays would probably be OK with a standard timeout or even require user action to clear them, but IMO, there is a small subset which I still feel the timeout should have some reasonable configurability.

#9946 Frontend cannot show recent logs MythTV - General unknown Developer Task Jul 26, 2011

My database has two table for log entries. New log entries are going into the "logging" table, but mythfrontend (and mythweb) shows the log entries which are in the "mythlog" table.

This appears to have been introduced with the merge of the "new-logging" branch.

#12312 mythfilldatabase (SD/DD) does not honour $TMPDIR MythTV - Mythfilldatabase unknown Patch - Bug Fix Nov 2, 2014

The Schedules Direct grabber does not honour (i.e. use) $TMPDIR if it's set.

I will attach a patch that I *think* should do the trick. I only very very occasionally find myself C++ programming so it might be quite right. Constructive criticism welcomed.

#9267 Add "Jump to Random" feature in libmythui MythTV - User Interface Library unknown Patch - Feature Nov 20, 2010

Similar to ticket #5626 from long ago, this patch adds a "Jump to Random" entry option to libmythui. This allows components such as MythGame? and MythVideo? to utilize a hotkey to pick a random choice in the list.

Feel free to adjust the default hotkey.

#10056 Move SQL statement for jumppoint and keybinding MythTV - General unknown Patch - Feature Sep 23, 2011

It make sense to group all the SQL statements relative to the key bindings and jump points together

mspieth (1 match)

Ticket Summary Component Milestone Type Created
#13230 Android green line when watching certain videos Ports - Android 30.0 Bug Report - General Feb 28, 2018

When watching videos that are resized there is a green line along the botton and right of the picture.

The green line only occurs if it is using texture format YV12. There are two texture formats used for most systems - YV12 and UYVY. The logic for choosing which to use is rather obscure. It is choosing UYVY on regular Linux and choosing YV12 on Android. Testing shows that both formats work for Linux as well as for Android, but YV12 is causing the green line on Android.

Changing to use UYVY on Android solves the problem in most cases. However, watching interlaced SD content shows the problem again at the bottom of the screen, even using UYVY.

There seems to be a problem with resizing and deinterlacing.

Nigel (1 match)

Ticket Summary Component Milestone Type Created
#10252 Mac OS X menu bar, dock, and mouse not hidden if GUI size nonzero Ports - OSX unknown Bug Report - General Jan 11, 2012

If you change the GUI size and offset to compensate for overscan, MythTV incorrectly uses a normal window instead of a full-screen window, resulting in Mac OS X showing the menu bar and doc. MythTV should use these values for scaling the *content* of the window, not for setting the actual size and position of the window itself.

The net result is that unless you can find a reduced-size theme that will fit on your screen, you're pretty much hosed.

paulh (1 match)

Ticket Summary Component Milestone Type Created
#12640 mythplugins/configure fails to find cdio/cdda.h Plugin - MythMusic 29.2 Bug Report - General Feb 5, 2016

I haven't attempted to build mythplugins for a couple of years. All the pre-reqs are installed. However, mythplugins reports

        libcdio        support will not be included in MythMusic

config.ep includes:

check_lib cdio/cdio.h cdio_open -lcdio
check_header cdio/cdio.h
BEGIN /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.vMrvu20m.c
    1	#include <cdio/cdio.h>
    2	int x;
END /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.vMrvu20m.c
/usr/bin/clang -D_ISOC99_SOURCE -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_DARWIN_C_SOURCE -DPIC -pipe -std=c99 -fomit-frame-pointer -fPIC -g -Wall -Wno-parentheses -Wno-format-zero-length -Wdisabled-optimization -Wpointer-arith -Wno-pointer-sign -Wwrite-strings -Wtype-limits -Wundef -Wno-pointer-to-int-cast -Wstrict-prototypes -O3 -fno-math-errno -fno-signed-zeros -O3 -fno-math-errno -fno-signed-zeros -Qunused-arguments -Werror=implicit-function-declaration -Werror=return-type -E -o /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.WyESZ2FS.o /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.vMrvu20m.c
check_func cdio_open -lcdio
check_ld -lcdio
BEGIN /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.vMrvu20m.c
    1	extern int cdio_open();
    2	int main(void){ cdio_open(); }
END /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.vMrvu20m.c
/usr/bin/clang -D_ISOC99_SOURCE -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_DARWIN_C_SOURCE -DPIC -pipe -std=c99 -fomit-frame-pointer -fPIC -g -Wall -Wno-parentheses -Wno-format-zero-length -Wdisabled-optimization -Wpointer-arith -Wno-pointer-sign -Wwrite-strings -Wtype-limits -Wundef -Wno-pointer-to-int-cast -Wstrict-prototypes -O3 -fno-math-errno -fno-signed-zeros -O3 -fno-math-errno -fno-signed-zeros -Qunused-arguments -Werror=implicit-function-declaration -Werror=return-type -c -o /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.WyESZ2FS.o /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.vMrvu20m.c
/usr/bin/clang -Wl,-dynamic,-search_paths_first -o /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.RJcXchI4 /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.WyESZ2FS.o -lcdio -framework CoreFoundation -framework VideoDecodeAcceleration -framework QuartzCore -lxml2 -lcrypto -lass -lfftw3_threads -lfftw3f -lfftw3 -lmp3lame -lm -liconv -L/opt/local/lib -lfreetype -lx264 -lmp3lame -lfaac -lm -lbz2 -lz
check_lib cdio/cdda.h cdio_cddap_open -lcdio_cdda
check_header cdio/cdda.h
BEGIN /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.vMrvu20m.c
    1	#include <cdio/cdda.h>
    2	int x;
END /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.vMrvu20m.c
/usr/bin/clang -D_ISOC99_SOURCE -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_DARWIN_C_SOURCE -DPIC -pipe -std=c99 -fomit-frame-pointer -fPIC -g -Wall -Wno-parentheses -Wno-format-zero-length -Wdisabled-optimization -Wpointer-arith -Wno-pointer-sign -Wwrite-strings -Wtype-limits -Wundef -Wno-pointer-to-int-cast -Wstrict-prototypes -O3 -fno-math-errno -fno-signed-zeros -O3 -fno-math-errno -fno-signed-zeros -Qunused-arguments -Werror=implicit-function-declaration -Werror=return-type -E -o /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.WyESZ2FS.o /opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.vMrvu20m.c
/opt/local/var/macports/build/_Users_craigtreleaven_MacPortsTemp_mythtv.27/mythtv-plugins.27/work/.tmp/mythplugins_conf.vMrvu20m.c:1:10: fatal error: 'cdio/cdda.h' file not found
#include <cdio/cdda.h>
1 error generated.

I think configure should be using pkg-config. On this system pkg-config reports:

$ pkg-config --libs libcdio libcdio++ libcdio_paranoia
-L/opt/local/lib -lcdio++ -lcdio -lm -Wl,-framework,CoreFoundation -Wl,-framework,IOKit -Wl,-framework,DiskArbitration -lcdio_paranoia -lcdio_cdda -lcdio -lm -liconv -Wl,-rpath,/opt/local/lib -Wl,-framework,CoreFoundation -Wl,-framework,IOKit -Wl,-framework,DiskArbitration 

Attaching config.ep

Roger Siddons (6 matches)

Ticket Summary Component Milestone Type Created
#12290 Commercial Flagging isn't being queued. MythTV - General 29.2 Bug Report - General Oct 3, 2014

In: TVRecEvent programinfo.cpp (GetFilesize?) the existing returned filesize is sometimes 0 and TVRec::FinishedRecording? won't allow the job to be queued if the file size is less than 1000 bytes.

This has deactivated all commercial flagging. A workaround is attached (not a fix.)

#12287 Extend Watch Recording 'jump' to all groups MythTV - General unknown Patch - Feature Sep 29, 2014

Ticket https://code.mythtv.org/trac/ticket/12157 introduced a very useful feature for navigating between the groups containing a specific recording. However it is currently limited to title groups.

If the group filter does not contain titles then the menu for recordings in "All Programmes" still shows a "List Recorded Episodes" button that performs no action.

For recordings in other groups (watchlist, category, search rules etc) it will navigate to "All Programmes" but then there is no convenient way of returning to the original group.

This patch reimplements the feature as a cyclic search that navigates between all displayed groups (in order of display).

If the group filter contains only titles (typical configuration?) then the behaviour is unchanged, ie. it will toggle between "All Programmes" and the Title group (although the menu options "List All Recordings"/"List Recorded Episodes" are replaced by "Show Next Group").

However, when the group filter also contains other groups (showing the recording), it will cycle through them all of them. For example. for a group filter of Titles, Watchlist & Category (where the title group is shown before the Category group) the navigation will be Watchlist -> the Title Group -> the Category -> All Programmes -> Watchlist etc.

This is much more useful to Watchlist users in particular.

#12296 Watchlist Improvements MythTV - General 30.0 Patch - Feature Oct 12, 2014

The watchlist has 2 functions:

a) to select the 'first' episode of each series of recordings, b) to present the resulting titles in a convenient order for watching


(a) 1. It currently selects the earliest/oldest episode so the wrong recording is shown when episodes are recorded out of order.

(a) 2. Series are grouped by recording rule. For rules that record different titles, the later ones will be hidden. Multiple episodes of the same title are shown when multiple rules record the same title.

(b) 3. Current ordering uses arbitrary weightings with a highly subjective set of rules summarised as:

  • promote series with many episodes
  • promote series that are about to record again
  • promote new recordings for 42 hrs
  • promote recordings from the same time of day, demote others
  • promote daily/weekly recordings
  • hide series that have been recently deleted
  • promote quickly watched series
  • demote old recordings

In practice this produces a seemingly random and unintuitive order which is then frequently reshuffled.

A series of patches is attached for ease of review, but should be applied in sequence.

  1. Group watchlist recordings by title rather than recording rule. Thus different titles recorded by a one rule are all shown and a single title recorded by multiple rules are grouped together. Also a pre-requisite for selection by episode number.
  1. 'First' episodes are selected by season/episode as well as age. Initial selection is age. However if the oldest episode has season/episode numbers assigned then it is superseded by any earlier season/episode that exists. Specials or untagged recordings that co-exist with tagged ones thus appear in proper sequence.
  1. The watchlist currently caches its scores in ProgramInfo::recpriority2 in order to display the program score/state on the Info/ProgDetails screen. This is confusing and an abuse of ProgramInfo.
    The score/state is of no interest to the user, who cannot alter it. It is debug information that belongs in the logs and has been removed along with the ProgramInfo caching.
    Themes should remove WATCH_LIST_SCORE & WATCH_LIST_STATUS widgets from progdetails*.html. However they are set as empty to prevent breakage for now.
  1. A new UI widget "watchtotal" enables the watchlist to show the number of episodes of each title, for example a "recordings" buttonlist widget of:
    <textarea name="title">
      <template>%title%% (|watchtotal| episodes)%% - "|subtitle|"%</template>
    will produce:
    The Simpsons (15 episodes) - "Homer does something stupid"
    This widget only exists when the Watchlist group is selected in the groups list - for other groups it will show the usual:
    The Simpsons - "Homer does something stupid"
  1. Restructures the code to support multiple sort strategies. Adds a setting "WatchListOrder" so the user can select their preferred sorting strategy in Setup/TV Settings/TV Playback/Page 6. It defaults to "Classic" - the existing strategy.
  1. Adds a new simplified sort strategy "LimitedOldest" that is IMHO more stable, intuitive & useful. Titles are ordered by record date (oldest first). New titles & quickly-watched titles are placed at the top; titles that remain unwatched are relegated to the bottom. Two new settings "PlaybackWLRecentLimit" & "PlaybackWLOldLimit" are introduced so the user can adjust this behaviour.

#12297 Upcoming Show Active filter MythTV - General unknown Patch - Feature Oct 12, 2014

The Upcoming Recordings screen has filters for "All" or "Important".

However "Important" still shows inactive rules and 'episode-limited' rules that aren't going to record. If there are lots (for example, to top up kids programmes) they clutter the view.

This patch adds a third filter "Active" that also filters these out to show only programmes that are going to record.

The menu is converted from a DialogBox to a MythMenu in order to use a sub-menu.

#12683 Add Zoomable guide/EPG MythTV - User Interface Library 30.0 Patch - Feature Mar 10, 2016

Prompted by: https://forum.mythtv.org/viewtopic.php?t=1358&p=6554

Allow themes to define zoomable guide widgets. When detected Zoom In/Out menu options appear to allow user to switch between them.

Also adds TV Frontend/Zoom In/Out key bindings but the number of different zoom bindings is growing - maybe they should be rationalised.

The widgets that are zoomable is prescribed by the code and should be added to the Theme Development Guide.

Zoomable theme widgets use the same principles as the Gallery - an integer suffix on the widget name indicates zoom level (0 = 'most zoomed out'). eg. guidegrid0, guidegrid1, guidegrid2...

Any number of zoom levels may be defined but they must be contiguous (discontiguous ones will be ignored)

Each widget may define different zoom levels. For zoom levels above those defined, the last widget will be used. eg. If a theme defines guidegrid0, guidegrid1, guidegrid2, timelist0, timelist1, then timelist1 will also be used for zoom level 2.

Backwards-compatible with existing themes, for which no zoom options appear and behaviour is unchanged. (An absent zoom suffix is equivalent to suffix "0").

Code defaults to zoom level 0 until user creates the setting by zooming. It could easily default to most zoomed in, but there's no simple way for theme to specify the default zoom level.

The theme file attached merges the modifications from the above thread with Mythbuntu 28.20. It's not pretty (different base theme versions ?) but serves as a demo and an example of how to easily define zoomable widgets. It would be even easier if relative sizing (<area>0,50%,100%,20%</area>) were used.

#12809 Use lastPlayPos rather than bookmark MythTV - Video Playback 30.0 Patch - Feature Jun 11, 2016

This patch offers a solution to https://forum.mythtv.org/viewtopic.php?f=9&t=163 '"Recently watched" recordings playback filter' requesting a way of locating part-played recordings more easily.

It involves the last-played position introduced by #11713 and Watchlist changes from #12296. Most importantly, it alters bookmarking behaviour to separate last-play position from 'user marks'.

As the changes are connected, they are presented together - 18 commits have been aggregated into 2 mega-patches for convenience, but I can easily push a branch if it helps.


  1. Last-played position is re-enabled and becomes exclusive indicator of playback position. Video Library and the preview generator also use it.
  2. Bookmarks become a user feature only and are never set/cleared automatically. 2 related settings are removed.
  3. Buttonlist items gain a progressbar widget and progress percentage value. 'Watch Recordings' uses this to show the playback position; 'Upcoming Recordings' uses it to show a (proxy) recording progress. See screenshots.
  4. The Watchlist promotes part-watched recordings to the top.
  5. Introduces a Myth-compatible QtConcurrent feature for conveniently running background threads.

These patches are fully revertable. However you will lose the last-played position of all part-played recordings/videos that were played whilst the patches were installed (because bookmark won't be set). Manually bookmark them first before reverting.

1. Extend lastPlayPos

Last-played position was disabled by 38443b8e (the menu option to resume playback never appears). It is re-enabled for recordings and videos.

As it duplicates (betters) the bookmark's 'save position on exit' function, this mark assumes that responsibility and is also updated on playback exit (provided playback has been going for at least 30 sec). At end of recording it is reset.

Note: 'Automatically Mark watched' is now also actioned by the end-of-playback dialog (unnoticed bug: it isn't currently).

By default, playback now starts from:

  • last-played position, if present
  • bookmark, if present (so user can override program start mark)
  • program start mark, if present (recordings only)
  • beginning of file

Menu options allow user to start from last-played position, bookmark or 'beginning' (program start mark or file start) where applicable. Menu options allow user to clear bookmark and reset last-played position.

This mark always indicates playback position (whereas bookmark may not) and so can be used to identify part-played recordings/videos.

Previews are now generated from:

  • last-played position, if present
  • bookmark, if present,
  • program start mark, if present
  • existing offset calculation, which is unchanged.

Possibly the latter two should be combined to 'x minutes from program start'

2. Simplify bookmarks

Bookmarks have always been confusing because they try to do (at least) 2 functions: mark last-played point or mark a user-designated point, i.e. program start or a point of interest.

This depends on 2 settings "Clear bookmark on playback" & "Action on playback exit/Save position and exit".

Either the user must remember to manually bookmark before exiting an unfinished show or forego the ability to mark anything (because Myth will overwrite it on exit).

By losing the last-play-point responsibility the bookmark becomes purely a user aid - it will never be automatically updated/cleared.

The 2 settings are now redundant and have been removed from the UI.

An example of the benefit: If my kids have part-watched a show I can resume playback, bookmark it, then rewind to watch from the beginning, exiting/resuming as I often as I wish. When I've finished it (their) next playback will resume from the bookmark automatically. Or I could jump-to-bookmark and leave the last-played-pos in its original location to retain the part-played indication.

Some menu options "Set last-play-pos to bookmark" and "Swap bookmark/last-play-pos" could prove useful to allow 2 people to part-watch shows simultaneously.

Although not ideal, it's an improvement (until user-assignable bookmarks get implemented)

3. Add UI

Buttonlist buttons can now contain a progressbar, as an intuitive way to show contextual 'progress'.

'Watch Recordings' uses it to show playback posiiton. Previews aren't regenerated whilst a file is in use, so playback by another frontend only updates the progress bar/counter, until playback ceases.

'Upcoming Recordings' uses the bar to show the progress of time for recordings - it will progress to 100% for failing (0-byte) recordings. Filesize is also available for 'Upcoming Recordings' now (although Scheduler Updates will reset it to 0.00 for 10 secs)

A percentage progress value is also provided that can be used for sorting, or as a "depends" for a 'part-watched' icon.

The modified MythCenter-Wide theme demonstrates the UI changes (in a cluttered and ugly way) by using the button background as a progress bar. Themers can do a better job.

Video Library has no UI indication of progress.

4. Show part-played shows

Whilst a 'Recently Watched' filter would be useful, it should probably include more than part-played recordings.

Sorting is of dubious benefit. If there are so many part-played recordings then the user is unlikely to select one based on playback positiion!

The important point is to list part-played recordings separately. As dekarl noted, this feature is a better fit for the Watchlist. An updated version of the Watchlist #12296 promotes part-watched to the top of the list.

Note that #12287 is useful for switching between the watchlist and 'all episodes'.

5. Convenience thread-runner

QtConcurrent::Run is a useful way of running verbiage-free background threads. I use it often in Myth without problem.

However, strictly it is incompatible with Myth because MThreadPool is not a subclass of QThreadPool. I can't find any evidence to justify this, but include a tailored version of QtConcurrent::Run that uses MThreadPool.

Stuart Auchterlonie (10 matches)

Ticket Summary Component Milestone Type Created
#7434 max episodes code doesn't respect autoexpire flag on a recording MythTV - General 29.2 Bug Report - General Oct 26, 2009

I've a recording rule setup as follows.

  • Max episodes = 20
  • Record new / delete old
  • Allow autoexpire

I decided i wanted to keep 1 of the 20 recordings and set autoexpire=off on that recording.

However when it recorded a new episode, that episode was removed because it was the oldest, in direct violation of autoexpire=off.


#12932 Patch to EIT table version handling in DVBStreamData MythTV - DVB 29.2 Patch - Bug Fix Nov 21, 2016

DVBStreamData current uses code from MPEGStreamData to handle redundancy (duplicate checking) and version checking. This is fine where the tables concerned only use the mpeg PSIPTable TableIDExtension field to uniquely identify instances of sub tables. This is not the case for all DVB tables. In particular the Service Desctiption Table (SDT) and the Event Information Table (see ETSI EN 300 468 V1.15.1, section 3.1 Definitions, definition of sub_table). This patch currently deals with the EIT only. In the EIT case a unique instance of a subtable is identified by the TableID, the TableIDExtension(service_id), the transport stream id, and the original network id.

This patch implements a 64 bit key constructed from the above 4 values as an index instead of the 32 bit key used in the current code.

I have also implemented the following two items.

A function to check EIT table completeness in line with some of the other tables. I think it would be a good idea to pass this flag on to the EIT handling code where it can be used to delay handling of tables sections until the table is complete thus saving unnecessary scheduler calls. It would be even betwwen if the caching necessary to acheive this was also implemented in DVBStreamData in line with the other tables that do this.

Code to handle the "severe" (their words) derogation from the ETSI standard regarding the original network id field here in the UK. As an aside I suspect that the EIT handling only worked here in the UK because the EIT table version handling is broken.


#8731 DVB-T scan: muxes not tuned during full scan but tuned using NIT info MythTV - Channel Scanner unknown Bug Report - General Aug 5, 2010

Using the same configuration as reported in #8726.

Scan details: full UK scan using Sutton Coldfield transmitter with 6 available muxes. Signal/tuning timeouts raised to 2000/6000ms. ALL services (DVB+MPEG) scanned, with decryption testing disabled.

During the scan, a signal and successful lock is achieved on only 1 of the possible muxes (746000000), even when using doubled timeouts. The remaining 5 muxes are discovered in the NIT of this mux, and subsequent tuning of these muxes is successful.

Checking the log output reveals the following for sample mux 634166670:

i) During the initial full scan using default scan parameters, the following was seen:

                        DVBChan(1:/dev/dvb/adapter102/frontend0): New Params: 634166670 auto 0 auto auto 8 a 1/32 n v fec: auto msys: UNDEFINED rolloff: 0.35
2010-08-05 06:42:04.455 DVBChan(1:/dev/dvb/adapter102/frontend0): Tune(): Frequency tuning successful.
2010-08-05 06:42:04.455 DTVSM(/dev/dvb/adapter102/frontend0)::SetChannel(-1, -1):
2010-08-05 06:42:04.455 SM(/dev/dvb/adapter102/frontend0)::Start: begin
2010-08-05 06:42:04.455 SM(/dev/dvb/adapter102/frontend0)::Start: end
2010-08-05 06:42:04.504 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:04.504 SM(/dev/dvb/adapter102/frontend0)::RemoveFlags: Seen() Match() Wait(Pos,)
2010-08-05 06:42:04.566 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:04.628 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:04.690 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:04.752 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:04.814 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:04.876 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:04.937 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:05.893 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:05.955 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:06.017 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:06.079 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:06.141 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:06.203 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:06.264 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:06.326 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:06.388 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:42:06.449 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)

No signal is seen within the 2000ms timeout value

ii) Having extracted this sample mux's details from the NIT of the 1 mux locked to (746000000), the later "fill-in" scan was succesful:

                        DVBChan(1:/dev/dvb/adapter102/frontend0): New Params: 634166670 qam_16 a 3/4 3/4 8 2 1/32 n v fec: auto msys: UNDEFINED rolloff: 0.35
2010-08-05 06:45:51.455 DVBChan(1:/dev/dvb/adapter102/frontend0): Tune(): Tuning to 634166670Hz
2010-08-05 06:45:52.356 dvbchannel.cpp:wait_for_backend: Status:
2010-08-05 06:45:52.356 DVBChan(1:/dev/dvb/adapter102/frontend0): Tune(): Frequency tuning successful.
2010-08-05 06:45:52.356 DTVSM(/dev/dvb/adapter102/frontend0)::SetChannel(-1, -1):
2010-08-05 06:45:52.356 SM(/dev/dvb/adapter102/frontend0)::Start: begin
2010-08-05 06:45:52.356 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:45:52.356 SM(/dev/dvb/adapter102/frontend0)::Start: end
2010-08-05 06:45:52.356 SM(/dev/dvb/adapter102/frontend0)::RemoveFlags: Seen() Match() Wait(Pos,)
2010-08-05 06:45:52.420 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:45:52.481 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen() Match() Wait(Sig,)
2010-08-05 06:45:52.510 DVBSM(/dev/dvb/adapter102/frontend0): UpdateValues -- Signal Locked
2010-08-05 06:45:52.540 SM(/dev/dvb/adapter102/frontend0)::AddFlags: Seen(PAT,) Match() Wait()
2010-08-05 06:45:52.540 ChannelScanSM(/dev/dvb/adapter102/frontend0): Got a Program Association Table for TransportID 4101

Is it possible to modify the default scan parameters in order to increase the likelihood of finding more/all muxes during the initial scan?

Full log output of the scan is attached.

#8732 DVB-T scan: duplicate muxes created for MPEG channels MythTV - Channel Scanner 29.2 Bug Report - General Aug 5, 2010

Using the same configuration as reported in #8726.

Scan details: full UK scan using Sutton Coldfield transmitter with 6 available muxes. Signal/tuning timeouts raised to 2000/6000ms. ALL services (DVB+MPEG) scanned, with decryption testing disabled.

During a successful DVB-T scan, all DVB channels are eventually discovered (although only 1 mux was successfully tuned during the initial scan phase and the remaining 5 muxes were found in the Network Information Table of the successfully scanned mux.)

However, although there are only 6 available muxes, 9 muxes are created in the database at the end of the scan.

Further checking reveals that the creation/configuration of the 4 MPEG channels found during the scan creates 3 new mux entries in the database, rather than configuring them with one of the existing mux entries used for the DVB channels on those muxes. Additionally, networkid is set to 0 on the additional muxes.

Full log output is attached to #8731

#12105 EPG: EITCache: Ignoring channel since it is locked MythTV - EIT unknown Bug Report - General Mar 29, 2014

After a mythbackend restart EPG is not loaded anymore and see the following messages:

2014-03-29 11:34:47.681760 I [410/435] TVRecEvent eitscanner.cpp:265 (StartActiveScan) - EITScanner (31): StartActiveScan called with 37 multiplexes
2014-03-29 11:34:47.905509 I [410/436] EIT eitscanner.cpp:143 (run) - EITScanner (31): Now looking for EIT data on multiplex of channel 81
2014-03-29 11:34:47.905563 I [410/436] EIT eitcache.cpp:386 (PruneOldEntries) - EITCache: Pruning all entries that ended before UTC 2014-03-28T10:40:12Z
2014-03-29 11:34:47.905573 I [410/436] EIT eitcache.cpp:112 (delete_in_db) - EITCache: Deleting old cache entries from the database
2014-03-29 11:34:49.046523 N [410/515] DVBRead recorders/dtvsignalmonitor.cpp:367 (HandlePMT) - DTVSigMon[31](/dev/dvb/adapter6/frontend0): PMT says program 81 is encrypted
2014-03-29 11:34:49.628863 I [410/435] TVRecEvent eitscanner.cpp:205 (StartPassiveScan) - EITScanner (31): Started passive scan.
2014-03-29 11:34:49.783281 I [410/515] DVBRead eitcache.cpp:161 (lock_channel) - EITCache: Ignoring channel 10099 since it is locked.
2014-03-29 11:34:49.844445 I [410/515] DVBRead eitcache.cpp:161 (lock_channel) - EITCache: Ignoring channel 1081 since it is locked.
2014-03-29 11:34:49.892038 I [410/515] DVBRead eitcache.cpp:161 (lock_channel) - EITCache: Ignoring channel 4930 since it is locked.

Before the restart mythbackend was in the middle of an active scan.

A little bit ugly is also the message when doing an active scan, that it shows it is doing a passive scan:

(StartPassiveScan) - EITScanner (31): Started passive scan

As an intermediate fix:

delete from eit_cache where status=1; 

#12976 Channel Icons not being downloaded from www.lyngsat-logo.com Services - Icon Downloads 29.2 Bug Report - General Jan 15, 2017

Install of 0.28, mythtv-setup seems to have tried downloading a package of Australian channel icons, but each one contains HTML errors like :

<title>403 Forbidden</title>
<p>You don't have permission to access /hires/aa/abc1_au.png
on this server.<br />
<address>Apache Server at www.lyngsat-logo.com Port 80</address>

#13014 switching to required delivery system MythTV - DVB 30.0 Bug Report - General Mar 12, 2017

The Hauppaugge Win soloHD and dualHD USB sticks (Silicon Labs Si2168), there are multiple delivery systems available: DVB-T, DVB-T2 and DVB-C/ANNEX_A.

The default at boot seems to be DVB-T. Using the tool dvb-fe-tool it is easy to switch to the required delivery system (dvb-fe-tool -a 0 -d DVBC/ANNEX_A for example). But when running mythtv-setup, the delivery system is recognized from the card. And I have not found a way to switch it.

My work-around is to run a quick systemd unit to use dvb-fe-tool to switch the card, but I feel I shoud be able to select the delivery system from the mythtv-setup utility.

#8774 Store channel type in database MythTV - Channel Scanner 29.2 Developer Task Aug 13, 2010

Store channel type (TV/Radio/Data etc) into database when we find that information during channel scanning.

This information can then be leveraged in other areas, such as when setting up a recorder and the type of streams it expects to find.

#13178 Fix reading run level for smolt hardware profile on system with systemd MythTV - General 30.0 Patch - Bug Fix Nov 19, 2017

On systems with systemd determining the run level via /sbin/runlevel fails which triggers an exception. The code catches the exception and writes an error message to stderr. The error message is translated using the i18n function _(). However the file software.py is missing the appropriate import which causes it to fail with "NameError?: global name '_' is not defined". Adding the correct import statement "from i18n import _" fixes the error.

#12264 Add --bootstrap option to mythbackend MythTV - General 29.2 Patch - Feature Sep 7, 2014

See pull request


stuartm (2 matches)

Ticket Summary Component Milestone Type Created
#11482 Implement support for LNAs through DVB API MythTV - DVB 29.2 Developer Task Apr 10, 2013


Since a lot of users have problems with drivers that default to disabling the LNA built into their tuners we should implement support for enabling the LNA when using the card with MythTV.

My experience of handling user support requests in IRC and mailing lists suggests that most users need the LNA enabled, so the simplest solution would just enable by default with a GUI option to disable it.

A better implementation would automatically the LNA if the signal strength/quality falls below a reasonable threshold. The difficulty here is that signal strength can change over time and with changes in weather, so the very best implementation would be tied to the signal monitor and work per-tune rather than be set once when scanning.

#11491 Patch to service api (content endpoint) allowing querying of recorded markup data MythTV - Services API - Backend unknown Developer Task Apr 17, 2013

This is a patch to allow the querying of RecordedMarkup? data via services api for third party player commskip.

Raymond Wagner (4 matches)

Ticket Summary Component Milestone Type Created
#7990 Rework of status handling in mythjobqueue MythTV - Mythjobqueue 29.2 Developer Task Jan 29, 2010

The goal of this task is to remove (most of) the custom handling of mythcommflag and mythtranscode within the jobqueue code. The jobqueue will only be responsible for startup and success/failure logs and status updates. The jobs themselves will be responsible for any task specific information displayed in the terminal logs, database logs, or jobqueue table.

#10801 Video screenshot images don't load under certain conditions MythTV - Video Library unknown Bug Report - General Jun 5, 2012

When using storage groups for all video info (Videos themselves as well as meta images), individual episode screenshots do not load under certain conditions and the season cover art is shown instead.

Required conditions:

  • Remote frontend (ie screenshot directory is not locally accessible)
  • Video file that has screenshot and where file is not located in a top level directory (See below for further info)
  • View mode: Browse Library, Show Directory Structure

With the above conditions, if the file is located 1 directory level down from the Storage Group root, the screenshot loads OK. If it is more than 1 level down, the cover image is loaded, not the screenshot. Eg Storage group: /video/ File 1: /video/MyShow/episode1.avi <--- Works File 2: /video/MyShow/season 1/episode1.avi <--- Doesn't work

Note that if you do the same test on the machine where the screenshot files are actually located (Eg combined backend/frontend) everything works OK. It is only when on a remote frontend that this occurs.

#11838 Screen Setup Wizard doesn't work with Frontend Control Socket MythTV - General unknown Bug Report - General Sep 10, 2013

I've had some overscan issues that I used to control via nvidia, but thought I'd try the screen setup wizard after upgrading to 0.27-fixes (RC). I tried to leverage the MythMote? app, which uses the frontend control socket to send keys to the frontend, but none of the relevant keys (up, down, left, right, enter, esc) worked when in the Screen Setup Wizard.

LIRC and the Keyboard did work, however - so this bug seems limited to just the frontend control socket. Also the frontend control socket has worked in all other screens, and during playback, without issue.

#8962 playbackbox.cpp:extract_job_state() inefficient MythTV - General 29.2 Developer Task Sep 18, 2010

Right now it queries the info independently for every recording on screen. But it is much more efficient to get this info in MainServer::HandleQueryRecordings?() where we already do a partial bulk load of this info to set FL_COMMPROCESSING.

xris (1 match)

Ticket Summary Component Milestone Type Created
#10456 Change packaging scripts to build noarch packages as noarch Packaging unknown Patch - Bug Fix Mar 14, 2012

Add BuildArch?: noarch to mythtv.spec and mythweb.spec in packaging/rpm. This will build packages that are not platform specific as noarch.

Note: See TracReports for help on using and creating reports.