There seem to be an error with my dm920 that caused my problems regarding freeze!
Beiträge von FoxyRabbit
-
-
It seems to me that it uses component ScrollLabel like this
and that seem right if you see picture with Default skin.I can see from code that this component reads other values from skin, so cant see that it should be that much work to read inn values from windowsstylescrollbar id=4.
Python: ScrollLabel.py
Alles anzeigendef applySkin(self, desktop, parent): if self.skinAttributes is None: return False skin.applyAllAttributes(self.long_text, desktop, self.skinAttributes, parent.scale) widget_attribs = [ ] scrollbar_attribs = [ ] for (attrib, value) in self.skinAttributes: if attrib.find("borderColor") != -1 or attrib.find("borderWidth") != -1: scrollbar_attribs.append((attrib,value)) if attrib.find("transparent") != -1 or attrib.find("backgroundColor") != -1: widget_attribs.append((attrib,value)) skin.applyAllAttributes(self.instance, desktop, widget_attribs, parent.scale) skin.applyAllAttributes(self.scrollbar, desktop, scrollbar_attribs+widget_attribs, parent.scale) s = self.long_text.size() self.instance.move(self.long_text.position()) lineheight=fontRenderClass.getInstance().getLineHeight( self.long_text.getFont() ) if not lineheight: lineheight = 30 # assume a random lineheight if nothing is visible lines = (int)(s.height() / lineheight) self.pageHeight = (int)(lines * lineheight) self.instance.resize(eSize(s.width(), self.pageHeight+(int)(lineheight/6))) scrollbarwidth, scrollbarborderwidth = self.scrollbar.updateScrollLabelProperties(20, 1) self.scrollbar.move(ePoint(s.width()-scrollbarwidth,0)) self.scrollbar.resize(eSize(scrollbarwidth,self.pageHeight+(int)(lineheight/6))) self.scrollbar.setOrientation(eSlider.orVertical); self.scrollbar.setRange(0,100) self.scrollbar.setBorderWidth(scrollbarborderwidth) self.long_text.move(ePoint(0,0)) self.long_text.resize(eSize(s.width()-scrollbarwidth-10, self.pageHeight*16)) self.setText(self.message) return True
Another thing that could be nice when we talk about Console and the new feature with horisontal scrollbar possibility is that it often would be a case when info in Console is wider than the listbox size. Could it be an idea to implent for both vertical and horisontal scrollbar for Console? But that would be some more work I guess. -
Thanks folks! You are doing a grate job! Keep on doing that!
By the way, can you also confirm that you will make ScrollLabel use windowstylescrollbar id = 4 values in the future?
-
Does anybody else have this problem?
After latest update vertical scrollbar in Console screen is messed up with Default skin (not HD).
I have studied it and find that windowstylescrollbar id=5 (horisontal scrollbar) interfere with this vertical scrollbar. If I switch to Default HD skin Console uses component ScrollLabel scrollbarwidth 20px and scrollbarborderwidt 1px (defined within python/component). Because this skin does not have windowsstylescrollbar id=5 defined.
Could this be fixed, and done so that Console screen uses windowsstylescrollbar id=4 (vertical scrollbar) values. Scrollbar for Console screen will then be as skinmakers want it to be.
Edit: If I change values in windowsstylescrollbar id=5, the scrollbar in Default skin i changing width and borderwidth.
-
I am using 0.8 Thor 5/6/7 & Intelsat 10-02
-
Does anybody else get freeze on Animal Planet HD after latest updates? I am using dm920.
Subtitles do work by the way!
-
-
Ahh, of course! I knew that offset right adjusted the config value in config lists, but got tricked by seeing TimerLog used listboxcontent and asumed the right offset then would do the same job for description there. Bad thinking!
Thanks for educating me!
You could see from time to time that config_descriptions is colliding with config values. It might be due to translation differences in text length, font choices and font size choice, or that text is just to long for some reasons. Would it be possible to make that not happen? In the same time it could be calculated for a right offset cut of config_description if config_value is None and config_description is longer than width of list. This should fix all possible look issues regarding config lists.
Just for discussion...
-
Hi,
We are building some skins and find that offset right do not seem to work for listboxcontent. Offset left works.
We are working with screen TimeLog from screen TimerEditList. Attached screenshot. Offset set to 10,10 left/right, and list shortened to demonstrate error.
Anybody else have this issue?
-
I've found a great plugin to mask logo to prevent ruin my oled tv.
Does oled tv's get ruined by logos!?
-
While it's discussable if providers are meant to use the upper bits,
Back to my questions again, and to learn from this, why do they want to suddenly use upper bits? Does it have any advantages over lower bits? And why do we see this coming now? If everybody understands this it might be less speculations around it.
-
Does someone know why this DVB change is coming now? Is the new option any way "better" than before, and will we see lots of other services that will use this in the near future? If so, it would be natural to update OE2.0 also? Else there will be a lot of dreambox tuners out there that no longer can be used with original images.
-
In such cases like this, who would normally report this to the devs? Since this issue is touching more than just dreambox. Eventually everybody gets the fix...
-
Someone contacted us (Peter Pan team) regarding missing DVB subtitle on this channel. After some time people in the same thread also reportet the same issue for other image makers eg. OpenPli, OpenATV, and also for other tuner producers like Mutant.
Could this be somewhat the same issue discussed i OpenPli forum in 2010. Then there was discusses three other channels with the same problem. It was mentioned in this thread some framebuffer acelleration not supported by the driver.
https://forums.openpli.org/topic/11468-dm800-subtitles/
Have tested a bit and the subtitle box comes around after a while but it is transparent with no text in it.
-
Hi,
- GSOD when switching filterlist Countries/Genre
- GREEN text in different lists
When pushing GREEN twice it will make an GSOD.
CodeTraceback (most recent call last): File "/usr/lib/enigma2/python/enigma.py", line 7824, in __call__ return ret(*args, **kwargs) File "/usr/lib/enigma2/python/Components/ActionMap.py", line 63, in action res = self.actions[action]() File "/usr/lib/enigma2/python/Plugins/Extensions/InternetRadio/InternetRadioScreen.py", line 560, in green_pressed config.plugins.internetradio.filter.value = self.filterSwitch[config.plugins.internetradio.filter.value] KeyError: 'Countries'
Have studied InternetRadioScreen.py and it seem that "removing" (make workaround) some _() regarding config.plugins.internetradio.filter.value fixes the problem.I also find that the GREEN button "status" text ("Countries" / "Genres") is misleading from what you see on screen when GREEN and OK are used in different lists.
This is my code that seems to solve this problem. Be nice with me, it is only a suggestion for what I think it should work like. It may be done more pretty for experts...
-
Well, we just want to make a note of an error that should be corrected. This is also a feedback from users of the PP image that says that it does not work to install ipk packages on dm900 & dm920. Not everyone understands that the arm platform should have deb, and mipsel platform ipk packages.
In my mind, if not grandmother understands, it's not good enough.
-
Hi again,
In Software Management there is wrong text for arm platform tuners. The text on screen is "ipkg-install". Should be "dpkg-install" for these tuners.
-
It happens when cookies list is empty. This code fixes the problem, but not sure if its done right.
Old code
BrowserMenu.pyPythondef __ckBuildList(self): print "[BrowserMenu].__ckBuildList" cookies = [] for c in self.__ckList: cookies.append(self.__ckGetEntryComponent(c)) self.detailList.setList(cookies)
code that works (no gsod when emtpty cookie list)
-
Ha, you choked it with to much power man! LOL!
-
See that one!
Enough said, the most important for me is that pixmapWidth and pixmapHeight are kept correct, and when that is true you still should be able to set the selectionPixmap margin as the value you use for other lists with selectionPixmap, and the correct y position thats needed for that specific pixmap size.
Your arguments of manipulating pixmapWidth and pixmapHeight is just not the way it should be done.