7 years agoAdd "rt" to libraries
Mike Looijmans [Sun, 3 Feb 2013 12:26:00 +0000 (13:26 +0100)]
Add "rt" to libraries

Newer compilers require -lrt for aio_write and related functions.

7 years agoRevert "ServiceName: getName() cannot be called with zero arguments."
Littlesat [Sun, 3 Feb 2013 10:23:50 +0000 (11:23 +0100)]
Revert "ServiceName: getName() cannot be called with zero arguments."

This reverts commit 027d32076d459084f45dde81dd439c12c26e1e99.

It was a fix for the <n/a> channels but it has a big site effect
that the channelname was not available anymore on the OSD when there
was no explecit #description in the settings

7 years agoServiceName: getName() cannot be called with zero arguments.
Stephen R. van den Berg [Sat, 2 Feb 2013 20:45:04 +0000 (21:45 +0100)]
ServiceName: getName() cannot be called with zero arguments.

7 years agotstools: force m_pts_length to be recalculated when m_pts_begin is set
pieterg [Sat, 2 Feb 2013 13:45:50 +0000 (14:45 +0100)]
tstools: force m_pts_length to be recalculated when m_pts_begin is set

m_pts_length can be correctly calculated (by calcEnd) only when
m_pts_begin is known, so calcBegin should have succeeded before
calcEnd is called.
Under certain circumstances, calcEnd could succeed before calcBegin,
leaving m_pts_length at an illegal value.
In that case, eDVBTSTools::calcLen might return an invalid length.

To avoid this, make sure the end value is invalidated, whenever
the begin value is set.
That ensures that begin and end are always calculated in the right
order, and m_pts_length will always be correct when m_begin_valid
and m_end_valid are both set.

thanks to betacentauri for finding this issue

7 years agopmt: always restart m_PAT/m_PMT when frontend state changes to ok
pieterg [Sat, 2 Feb 2013 10:47:47 +0000 (11:47 +0100)]
pmt: always restart m_PAT/m_PMT when frontend state changes to ok

Even if we have already allocated a demux, it could be that
m_PAT / m_PMT have timed out (or returned invalid data) on
the previous attempt.
When that happened, we would not restart them when the
frontend lost and re-gained it's lock, because we already
allocated a demux.

This situation might occur with a rotor setup for instance,
where the frontend gets a lock before the positioner has
reached the correct orbital position.
(as might happen when there is no current measurement, or
current measurement is not reliable)
As a result, no channel would be started, and no retries
would be attempted.

7 years agoGMEPG: Avoid GSOD when parantal control blocks a service
Littlesat [Sat, 2 Feb 2013 10:24:18 +0000 (11:24 +0100)]
GMEPG: Avoid GSOD when parantal control blocks a service

On OK you never get to a parentel controlled channel. You even cannot
exit the GMEPG. Only with exit this is possible this time - but then you
have to enter a PIN.

7 years agoEnsure that we only have one nofication to enter parantel pin
Littlesat [Sat, 2 Feb 2013 10:21:58 +0000 (11:21 +0100)]
Ensure that we only have one nofication to enter parantel pin

As sometimes with preview we could have mulitple instances

7 years agoGMEPG: possibleAlignmentChoises start with lower case letter
Littlesat [Fri, 1 Feb 2013 21:24:38 +0000 (22:24 +0100)]
GMEPG: possibleAlignmentChoises start with lower case letter

7 years agoGMEPG: Make alignments of Service Name and Events configurable
Littlesat [Fri, 1 Feb 2013 21:08:59 +0000 (22:08 +0100)]
GMEPG: Make alignments of Service Name and Events configurable

7 years agoChange some backslash and continue on next line
Littlesat [Fri, 1 Feb 2013 20:10:04 +0000 (21:10 +0100)]
Change some backslash and continue on next line

As it makes the code less readable

7 years agoCleanup SkinSelector to use os.path and follow symlinks.
Stephen R. van den Berg [Sat, 26 Jan 2013 00:58:34 +0000 (01:58 +0100)]
Cleanup SkinSelector to use os.path and follow symlinks.

7 years agoGetEcmInfo make ecm global to support peeking in the raw lines.
Stephen R. van den Berg [Thu, 31 Jan 2013 08:50:38 +0000 (09:50 +0100)]
GetEcmInfo make ecm global to support peeking in the raw lines.

7 years agoMinimise code duplication for isPlayable functions.
Stephen R. van den Berg [Thu, 31 Jan 2013 19:34:39 +0000 (20:34 +0100)]
Minimise code duplication for isPlayable functions.

Pull together several similar routines in places sprinkled throughout
the code and centralise them in one place to minimise code duplication.

7 years agoGMEPG Show channels with a different colour if not playable.
Stephen R. van den Berg [Thu, 31 Jan 2013 21:57:48 +0000 (22:57 +0100)]
GMEPG Show channels with a different colour if not playable.

7 years agoGMEPG Actually honour font statements specified in the skin.
Stephen R. van den Berg [Thu, 31 Jan 2013 00:56:57 +0000 (01:56 +0100)]
GMEPG Actually honour font statements specified in the skin.

7 years agoServiceName Reference resolves alternates.
Stephen R. van den Berg [Tue, 22 Jan 2013 00:51:52 +0000 (01:51 +0100)]
ServiceName Reference resolves alternates.

7 years agoOrbitalPosition shows the position of the next channel.
Stephen R. van den Berg [Tue, 22 Jan 2013 02:15:28 +0000 (03:15 +0100)]
OrbitalPosition shows the position of the next channel.

7 years agoAvoid race condition, consolidate check into stat() exception block.
Stephen R. van den Berg [Mon, 21 Jan 2013 01:21:37 +0000 (02:21 +0100)]
Avoid race condition, consolidate check into stat() exception block.

7 years agoGMEPG add ServiceEvent support.
Stephen R. van den Berg [Thu, 24 Jan 2013 21:43:38 +0000 (22:43 +0100)]
GMEPG add ServiceEvent support.

7 years agoGMEPG fix spurious missing descriptions and large timejumps.
Stephen R. van den Berg [Sun, 27 Jan 2013 22:41:32 +0000 (23:41 +0100)]
GMEPG fix spurious missing descriptions and large timejumps.

7 years agoGMEPG Do not repaint backgrounds at this point.
Stephen R. van den Berg [Thu, 13 Dec 2012 02:05:24 +0000 (03:05 +0100)]
GMEPG Do not repaint backgrounds at this point.

7 years agocosmetics in cs.po
ims [Sat, 26 Jan 2013 13:47:22 +0000 (14:47 +0100)]
cosmetics in cs.po

Signed-off-by: pieterg <>

7 years agoPluginBrowser: Add lacking translation string
daddelfalk [Mon, 28 Jan 2013 07:02:13 +0000 (08:02 +0100)]
PluginBrowser: Add lacking translation string

Signed-off-by: pieterg <>

7 years agofont: fix boundingbox calculation for bordered glyphs
ims [Sun, 27 Jan 2013 16:43:16 +0000 (17:43 +0100)]
font: fix boundingbox calculation for bordered glyphs

For left aligned text, the border of the first character
could fall outside the boundingbox, causing it to be cut off.
Compensate the x position of the first character with the
border width.

Signed-off-by: pieterg <>

7 years agoservicedvb: rewrite teletext subtitle page handling
pieterg [Sun, 27 Jan 2013 15:08:20 +0000 (16:08 +0100)]
servicedvb: rewrite teletext subtitle page handling

-only query decoder PTS when we actually need it
-make sure every appended page gets a valid pts value,
so we do not have to make exceptions in checkSubtitleTiming()
-make 'config.subtitles.subtitle_bad_timing_delay' work for
live tv with missing pts as well (as it was probably intended)

7 years agoPluginBrowser: Sort plugin list alphabetically
Mike Looijmans [Sat, 26 Jan 2013 13:25:03 +0000 (14:25 +0100)]
PluginBrowser: Sort plugin list alphabetically delivers them per feed.

7 years agoopkg: Grab more feeds, throw less errors
Mike Looijmans [Fri, 25 Jan 2013 18:05:18 +0000 (19:05 +0100)]
opkg: Grab more feeds, throw less errors

Apparently opkg just grabs all "-feed.conf" files. Do the same so we
don't miss the 3rd party plugins.

Catch errors, no need to crash enigma when a file's missing.

7 years agofix character spacing for centered, right aligned, and right-to-left text
ims [Thu, 24 Jan 2013 19:16:10 +0000 (20:16 +0100)]
fix character spacing for centered, right aligned, and right-to-left text

The kerning parameters, correctly applied in eTextPara::appendGlyph,
were undone because the characters were repositioned
based on just the glyph width, ignoring kerning.

Instead of recalculating the character positions, just move each
glyph (and its bounding box) with the same offset, keeping
the original spacing.

Signed-off-by: pieterg <>

7 years agoRevert "GMEPG add ServiceEvent support."
Littlesat [Thu, 24 Jan 2013 17:44:41 +0000 (18:44 +0100)]
Revert "GMEPG add ServiceEvent support."

This reverts commit 525d1d9cdd6332355f7a970f13635d6c9614b0e8.
As it seems to break some plugins

7 years agoRevert "Bring EPGSelection in line with other screens and support ServiceEvent."
Littlesat [Thu, 24 Jan 2013 17:43:22 +0000 (18:43 +0100)]
Revert "Bring EPGSelection in line with other screens and support ServiceEvent."

This reverts commit f9007432d542b544f93a5ce117bd1c8b6316663d.
As it seems to break some plugins

7 years agoInputDeviceSetup: fix config entry label translation
pieterg [Thu, 24 Jan 2013 10:12:42 +0000 (11:12 +0100)]
InputDeviceSetup: fix config entry label translation

as in 37e27e3b6814f6cece350986c283ed6a3f89388c

(xgettext does not find doubly quoted messages)

7 years agoRevert "changes for gettext in input devices"
pieterg [Thu, 24 Jan 2013 10:01:02 +0000 (11:01 +0100)]
Revert "changes for gettext in input devices"

This reverts commit 37e27e3b6814f6cece350986c283ed6a3f89388c.

We should not try to translate input device names, these
names are defined by device drivers, we have no control
over them.

7 years agoNimManager: skip empty 'Mode' lines in /proc/bus/nim_sockets
pieterg [Wed, 23 Jan 2013 23:15:09 +0000 (00:15 +0100)]
NimManager: skip empty 'Mode' lines in /proc/bus/nim_sockets

dm800 shows a trailing empty mode line for multi-mode
vtuner devices, which we should ignore, e.g.:

Mode 0: DVB-S2
Mode 1: DVB-S
Mode 2:

(the same might be the case for other dreambox models)

7 years agoRevert "unpause timeshift with play/pause button"
Littlesat [Wed, 23 Jan 2013 17:07:51 +0000 (18:07 +0100)]
Revert "unpause timeshift with play/pause button"

This reverts commit 2dcd5bbe1b3597ff1afd9039564b04d5f045fede.

As it breaks the pauze button in some cases when you e.g. go to the
movieplayer list.

7 years agounpause timeshift with play/pause button
ims [Tue, 22 Jan 2013 21:39:50 +0000 (22:39 +0100)]
unpause timeshift with play/pause button

7 years agoRevert "Disable overcommit for non-workstations."
Littlesat [Mon, 21 Jan 2013 14:39:53 +0000 (15:39 +0100)]
Revert "Disable overcommit for non-workstations."

This reverts commit 9a9f5fdb93794afef8692726cf84f8c38a5ceb03.
As it seems playing some MKVs and movies on some boxes

7 years agoGetEcmInfo: Also clear cached info when no ecm info
Littlesat [Mon, 21 Jan 2013 06:58:45 +0000 (07:58 +0100)]
GetEcmInfo: Also clear cached info when no ecm info

And just return as the True and False is not used

7 years agoGetEcmInfo: Reply empty data when there is no
Littlesat [Mon, 21 Jan 2013 00:11:32 +0000 (01:11 +0100)]
GetEcmInfo: Reply empty data when there is no

And remove some spaces as intend and whitespaces and a ;

7 years agoWirelessLan plugin: do not rely on pythonwifi.iwlibs.getWNICnames
pieterg [Sun, 20 Jan 2013 19:05:58 +0000 (20:05 +0100)]
WirelessLan plugin: do not rely on pythonwifi.iwlibs.getWNICnames

getWNICnames reads /proc/net/wireless, which does not necessarily
contain all wlan interfaces.
For instance, the rt2800usb driver on a 3.2 kernel does not cause its
interface to be listed in /proc/net/wireless.
The same might be the case for other drivers and kernel versions.

7 years agoInfoBarGenerics: Initialize TimeshiftActivateActions Inactive
Littlesat [Sun, 20 Jan 2013 18:16:16 +0000 (19:16 +0100)]
InfoBarGenerics: Initialize TimeshiftActivateActions Inactive

At startup of Enigma2 timeshift will work right a way without any
zap before.

7 years agoDo not break timeshift when going to MovieList
Littlesat [Sun, 20 Jan 2013 18:10:59 +0000 (19:10 +0100)]
Do not break timeshift when going to MovieList

Initializing InfoBarBase in MoviePlayer does corrupt timeshift
functionality. When timeshift is Enabled and we do not Initialize
the InfoBarBase the timeshift will not behave strange anymore.
For now this solves at least a known timeshift issue.

7 years agoMovieSelection: Add purging of trashcan subdirectories
Mike Looijmans [Sun, 20 Jan 2013 17:59:03 +0000 (18:59 +0100)]
MovieSelection: Add purging of trashcan subdirectories

When "deleting" a folder in the trashcan, it will be purged instead
of attempting to move it again. Remaining issue is that the folders
will remain (because files are being removed in background). But it
is a major improvement over just erroring out.

7 years agoRevert "GMEPG Fix forgotten repaint background colour."
Littlesat [Sat, 19 Jan 2013 19:45:30 +0000 (20:45 +0100)]
Revert "GMEPG Fix forgotten repaint background colour."

This reverts commit 4e4ea6fc94f82c28c2be33ee23ffd2e13f4bcd12.

Fix was incomplete as it overwrites the png background

7 years agoBring EPGSelection in line with other screens and support ServiceEvent.
Stephen R. van den Berg [Fri, 28 Dec 2012 21:25:19 +0000 (22:25 +0100)]
Bring EPGSelection in line with other screens and support ServiceEvent.

7 years agoExtend ServiceName to support References for non-current channels.
Stephen R. van den Berg [Sat, 19 Jan 2013 00:48:19 +0000 (01:48 +0100)]
Extend ServiceName to support References for non-current channels.

7 years agoExtend ClocktoText with AsLengthHours.
Stephen R. van den Berg [Mon, 24 Dec 2012 00:40:54 +0000 (01:40 +0100)]
Extend ClocktoText with AsLengthHours.

7 years agoCleanup the calculations in EventTime.
Stephen R. van den Berg [Mon, 24 Dec 2012 09:47:31 +0000 (10:47 +0100)]
Cleanup the calculations in EventTime.

7 years agoStreamline RemainingtoText and add support for Progress.
Stephen R. van den Berg [Sun, 23 Dec 2012 19:11:24 +0000 (20:11 +0100)]
Streamline RemainingtoText and add support for Progress.

7 years agoExtend and partial rewrite ECM info.
Stephen R. van den Berg [Sat, 22 Dec 2012 15:33:21 +0000 (16:33 +0100)]
Extend and partial rewrite ECM info.

7 years agoGMEPG add ServiceEvent support.
Stephen R. van den Berg [Tue, 25 Dec 2012 02:24:49 +0000 (03:24 +0100)]
GMEPG add ServiceEvent support.

7 years agoGMEPG Center time-labels above columns.
Stephen R. van den Berg [Thu, 13 Dec 2012 02:23:23 +0000 (03:23 +0100)]
GMEPG Center time-labels above columns.

This means that the date cannot be shown at the upperlefthand corner anymore
(there is no room left).  The date should be visible in the detail info
already (or can be made visible at some other available spot on the screen).

7 years agoGMEPG Fix forgotten repaint background colour.
Stephen R. van den Berg [Thu, 13 Dec 2012 02:05:24 +0000 (03:05 +0100)]
GMEPG Fix forgotten repaint background colour.

7 years agoGMEPG More consistent up/down arrows when moving between channels.
Stephen R. van den Berg [Thu, 13 Dec 2012 02:18:40 +0000 (03:18 +0100)]
GMEPG More consistent up/down arrows when moving between channels.

The previous behaviour was erratic or chaotic at best.  Moving down and up
again now allows you to stay in the same column vertically.

7 years agoGMEPG Align service entries to the left to ease finding where it starts.
Stephen R. van den Berg [Fri, 21 Dec 2012 15:13:56 +0000 (16:13 +0100)]
GMEPG Align service entries to the left to ease finding where it starts.

Also turn off line wrap in those entries, because the display can get really
cluttered otherwise.

7 years agoGMEPG Increase enforced limits to support bigger screens.
Stephen R. van den Berg [Thu, 13 Dec 2012 02:15:49 +0000 (03:15 +0100)]
GMEPG Increase enforced limits to support bigger screens.

7 years agoDisable overcommit for non-workstations.
Stephen R. van den Berg [Fri, 7 Dec 2012 22:16:46 +0000 (23:16 +0100)]
Disable overcommit for non-workstations.

This allows the STB to reliably report how much memory (including swap)
is available at all times, and consequently means that the OOM-killer
is never invoked (and thus no unexpected crashes due to that).

7 years agoFix Button and Label classes to mimick Source to support generic templates in skins.
Stephen R. van den Berg [Thu, 29 Nov 2012 23:44:43 +0000 (00:44 +0100)]
Fix Button and Label classes to mimick Source to support generic templates in skins.

7 years agoMovieSelection: Keep settings and network items in list
Mike Looijmans [Sat, 19 Jan 2013 18:51:52 +0000 (19:51 +0100)]
MovieSelection: Keep settings and network items in list

A skin change is needed to "see" the colored buttons, so people with
other skins will wonder where they went.

7 years agoFix to work with older opkg versions
Mike Looijmans [Sat, 19 Jan 2013 18:05:55 +0000 (19:05 +0100)]
Fix to work with older opkg versions

Prevent "duplicate" packages.

7 years agoopkg: Handle new packaging output
Mike Looijmans [Mon, 31 Dec 2012 10:54:28 +0000 (11:54 +0100)]
opkg: Handle new packaging output

Just parse the OPKG files directly, that's much easier than parsing
opkg's rendering. The v4 packaging has some nasty habits:
- multiline descriptions
- repeating package name and version in the description
- wrapping lines in odd places (adds extra spaces to versions)
This works around those.

7 years agotypo in NationalOptionSubsets for czech
ims [Sat, 19 Jan 2013 13:29:36 +0000 (14:29 +0100)]
typo in NationalOptionSubsets for czech

7 years agosettings and net moved from movielist menu under color buttons
ims [Sat, 19 Jan 2013 15:33:17 +0000 (16:33 +0100)]
settings and net moved from movielist menu under color buttons

7 years agoInfoBarGenerics: set seekstate to play when not seekable in timeshift
Littlesat [Thu, 17 Jan 2013 23:35:19 +0000 (00:35 +0100)]
InfoBarGenerics: set seekstate to play when not seekable in timeshift

Thanks ims and also ensure to stop when zap was not a good idea...

7 years agoInfoBarGenerics: Ensure timeshift is stopped when zap
Littlesat [Thu, 17 Jan 2013 23:03:18 +0000 (00:03 +0100)]
InfoBarGenerics: Ensure timeshift is stopped when zap

7 years agoInfoBarGenerics: Some cosmetic cleanups regards Timeshift
Littlesat [Thu, 17 Jan 2013 22:20:16 +0000 (23:20 +0100)]
InfoBarGenerics: Some cosmetic cleanups regards Timeshift

Use True and False instead of 0, 1 and False mixed up together
for timeshift_enabled and remove not used declared timeshift_state

7 years agoInfoBarGenerics: Some tiny timeshift related fixes
Littlesat [Thu, 17 Jan 2013 21:23:58 +0000 (22:23 +0100)]
InfoBarGenerics: Some tiny timeshift related fixes

Fixed when going to MovieSelection and directly exit that timeshift
could be started, but not played anymore.
Remove PVRState when zapping/selecting different service that also
stops the timeshift.

I still discover some strange things in case the timeshift buffer
does not have much time yet (< 1 min) and we step skip back. This
is not fixed (yet)

7 years agoMovieSelection: Remove extension code/data duplication and speed it up
Mike Looijmans [Wed, 16 Jan 2013 19:35:56 +0000 (20:35 +0100)]
MovieSelection: Remove extension code/data duplication and speed it up

Put the "known extension types" lists in one spot for easy maintenance.

Doing "a in (very,big,tuple)" will sequentially scan all elements of
the tuple. This is much slower than a set or frozenset which uses a
hash lookup.

7 years agoMovieSelection: Remove delayTimer use callLater instead
Littlesat [Wed, 16 Jan 2013 18:52:26 +0000 (19:52 +0100)]
MovieSelection: Remove delayTimer use callLater instead

As callLater does in fact the same

7 years agoMovieList: Add pictures to the known extensions
Littlesat [Wed, 16 Jan 2013 11:41:06 +0000 (12:41 +0100)]
MovieList: Add pictures to the known extensions

And do it slightly different - thanks DeBeek

7 years simplify vu+ boxtype detection
pieterg [Wed, 16 Jan 2013 09:52:44 +0000 (10:52 +0100)] simplify vu+ boxtype detection

7 years agoUsageConfig: Latvian should be lav instead of lat
Littlesat [Tue, 15 Jan 2013 13:30:27 +0000 (14:30 +0100)]
UsageConfig: Latvian should be lav instead of lat

Thanks Artis


7 years agoMovieSelection: Better description an simpler coding
Littlesat [Mon, 14 Jan 2013 22:50:59 +0000 (23:50 +0100)]
MovieSelection: Better description an simpler coding

7 years agoMovieSelection: Add option to hide known extensions
Littlesat [Mon, 14 Jan 2013 22:46:56 +0000 (23:46 +0100)]
MovieSelection: Add option to hide known extensions

And remove some more white spaces

7 years agoMovieSelection: Fix some whitespaxes
Littlesat [Mon, 14 Jan 2013 19:02:07 +0000 (20:02 +0100)]
MovieSelection: Fix some whitespaxes

7 years agoRemove DemoPlugins
Mike Looijmans [Mon, 14 Jan 2013 17:40:26 +0000 (18:40 +0100)]
Remove DemoPlugins

Not built, not needed, so remove them. Even as examples they fail.

7 years agoGraphMultiEPG: More fixes regards alternatives
Littlesat [Sun, 13 Jan 2013 16:48:26 +0000 (17:48 +0100)]
GraphMultiEPG: More fixes regards alternatives

7 years agoGraphMultiEPG: Fix for previous commit regards alternatives
Littlesat [Sun, 13 Jan 2013 16:07:45 +0000 (17:07 +0100)]
GraphMultiEPG: Fix for previous commit regards alternatives

The reference could be an alternative, so compare for the

7 years agoGraphMultiEpg: Change zap behaviour in GMEPG
Littlesat [Sun, 13 Jan 2013 15:47:16 +0000 (16:47 +0100)]
GraphMultiEpg: Change zap behaviour in GMEPG

Only quit GMEPG when you zap to the current played channel. This will
lead to the effect that during normal zapping the GMEPG is not quited. It
will quit only when you select 2 x the same channel.

7 years agobg.po: translation updates by marto
pieterg [Sun, 13 Jan 2013 00:43:04 +0000 (01:43 +0100)]
bg.po: translation updates by marto

7 years agofix typo in decoder.cpp
Mirakels [Thu, 10 Jan 2013 22:56:44 +0000 (23:56 +0100)]
fix typo in decoder.cpp

Thanks IMS

7 years agoUpdates for show single frame.
Mirakels [Sat, 5 Jan 2013 22:28:13 +0000 (23:28 +0100)]
Updates for show single frame.

- allow h264 still images to be shown (based on showiframe patch from
  Erik slagter)
- vusolo2 (and probably dus2 too) did not properly show single
  iframes in enigma2 with drivers from before 2013. This was very
  noticable when in radio mode.

7 years agocosmetics in cs.po
ims [Tue, 8 Jan 2013 20:58:45 +0000 (21:58 +0100)]
cosmetics in cs.po

Signed-off-by: pieterg <>

7 years agoFastScan: follow nim_list --> nimList rename
pieterg [Wed, 2 Jan 2013 22:36:22 +0000 (23:36 +0100)]
FastScan: follow nim_list --> nimList rename

in 19f4f7bf2f39c9662fc9bd9ab6aead26fada310e

7 years agono.po: translation updates by andy1
pieterg [Tue, 1 Jan 2013 15:54:43 +0000 (16:54 +0100)]
no.po: translation updates by andy1

7 years agofrontend: Add SNR approximation in dB for MaxDigital XP1000 NIMs.
Erik Slagter [Tue, 1 Jan 2013 12:52:49 +0000 (13:52 +0100)]
frontend: Add SNR approximation in dB for MaxDigital XP1000 NIMs.

Thanks to theparasol for the percentage data. The approximation
might improve with more data.

7 years agotstools: Correctly and efficiently calculate recording length
Mike Looijmans [Mon, 31 Dec 2012 14:10:44 +0000 (15:10 +0100)]
tstools: Correctly and efficiently calculate recording length

A recording with dicontinuities, such as when you've cut out commercials
or when there were big dropouts during recording, did not account for those
when the length was requested.

To fix this, take the discontinuities into account, which are calculated from
the .ap file. Use the "fixupPTS" function to correctly estimate the length.

Improve efficiency: Keep the calculated length as a member, because it is being
requested many times per second while playing.

7 years agoeDVBFrontend::readFrontendData: only get SNR/AGC/BER when we have a lock
pieterg [Sun, 30 Dec 2012 20:15:24 +0000 (21:15 +0100)]
eDVBFrontend::readFrontendData: only get SNR/AGC/BER when we have a lock

avoids a few ioctl's while we're tuning

7 years agoFastScan: Another small optimalization
Littlesat [Sun, 30 Dec 2012 12:38:16 +0000 (13:38 +0100)]
FastScan: Another small optimalization

When a recording is busy we do not need to search though NIMs

7 years agoFastScan: Only search for suitable NIMs once
Littlesat [Sun, 30 Dec 2012 12:25:10 +0000 (13:25 +0100)]
FastScan: Only search for suitable NIMs once

7 years agoscan.cpp: Add missing fcntl.h include
Mike Looijmans [Sat, 29 Dec 2012 18:28:27 +0000 (19:28 +0100)]
scan.cpp: Add missing fcntl.h include

Some compilers might complain like this:
error: 'O_RDONLY' was not declared in this scope

7 years agoFastScan: Give warning screen when no suitable NIMs are available
Littlesat [Sat, 29 Dec 2012 13:14:32 +0000 (14:14 +0100)]
FastScan: Give warning screen when no suitable NIMs are available

There was already a warning message screen available when there
was no sat tuner available now extend that to suitable tuners
and remove the disappear from menu as it was some kind of double
wise implemented during my previous commit

7 years agoFastscan: Only appear when there are suitable NIMs
Littlesat [Sat, 29 Dec 2012 12:43:46 +0000 (13:43 +0100)]
Fastscan: Only appear when there are suitable NIMs

7 years agoFastscan: Remember last used fastscan configuration
Littlesat [Sat, 29 Dec 2012 11:05:28 +0000 (12:05 +0100)]
Fastscan: Remember last used fastscan configuration

7 years agoServiceList: remove too much indents
Littlesat [Thu, 27 Dec 2012 21:18:44 +0000 (22:18 +0100)]
ServiceList: remove too much indents

7 years agoRelocate refreshServiceList to
Littlesat [Thu, 27 Dec 2012 21:13:48 +0000 (22:13 +0100)]
Relocate refreshServiceList to

7 years agoFastscan: Use tables to configure fastscan lists
Littlesat [Thu, 27 Dec 2012 20:48:31 +0000 (21:48 +0100)]
Fastscan: Use tables to configure fastscan lists

In addition relead service list after a fastscan was succesfull

7 years agoAudioSelection: No QuickSubtitleConfigMenu while playing DVDs
Littlesat [Thu, 27 Dec 2012 15:35:41 +0000 (16:35 +0100)]
AudioSelection: No QuickSubtitleConfigMenu while playing DVDs

7 years agoFastscan: made transponder parameters configurable in python
Littlesat [Thu, 27 Dec 2012 15:05:38 +0000 (16:05 +0100)]
Fastscan: made transponder parameters configurable in python

7 years agogdi: Implement drawing non-black lines on surfaces without clut
Mike Looijmans [Thu, 27 Dec 2012 11:06:20 +0000 (12:06 +0100)]
gdi: Implement drawing non-black lines on surfaces without clut

Line was only implemented for color lookup tables, not for regular
RGB surfaces. Refactor a bit and make it work. drwaLine is now

7 years agodvb.cpp: Refactor getNextSourceSpan
Mike Looijmans [Wed, 26 Dec 2012 11:02:58 +0000 (12:02 +0100)]
dvb.cpp: Refactor getNextSourceSpan

Get rid of code duplication by introducing diff_upto().
tstools always returns aligned offsets, so there's no need to align
them again. Otherwise, the more sensible thing to do would be to
fix tstools, instead of sprinkling the code with align() calls.