34 replies to this topic

[Kermit00]

If I start enigma2 from a telnet console I get

 

[SKIN] Parsing embedded skin <embedded-in-'VirtualKeyBoardExt'>
[Skin] Error: scale

 

Maybe openpli don't support this "scale" attribute in engima2 skins? Why?

[Kermit00]

Yes not funny if someone is decompiling our plugin without permission. 

And I've already fixed that issue.

 

 

Sorry, I just searched a solution. Maybe a mod can remove the attachment (I cannot edit it anymore), if you have a problem with it.

[dhwz]

Yes because the scale parameter is only used on DM7080, but all DMM Images just ignore this parameter, so it doesn't crash.

[littlesat]

Maybe openpli don't support this "scale" attribute in engima2 skins? Why?

Sounds like an attribute that is available in the closed source DMMs stuff (DMM7080)... We can also add it to give a parameter to scale pngs... (scaling is something PLi used for years for e.g. scale the picons). As far I can see we have scale in skin.py.... (row 307) but somehow this possibly crashes...)

 

And why shoot we create solutions for a closed source plugin....?

 

@dhwz,

When you want to have that code removed please Report it to our moderator via the Report button (sorry I do not have moderator rights)

Edited by littlesat, 23 November 2014 - 20:47.

[Kermit00]

Ok, i didn't know it's dmm stuff. I already reported it (asked for removal)

[dhwz]

Ok again, I've already fixed that so it should work with any image with the next update. There is no need to change anything on the image.

[Kermit00]

Sorry, I mean I reported the attached decompiled code (for removal).

Edited by Kermit00, 23 November 2014 - 20:56.

[littlesat]

At least thanks dhwz for solving the "issue"....

 

I also marked the py to the moderator for removal.

 

Just a question.... why not using a build in virtual keyboard?

Edited by littlesat, 23 November 2014 - 21:00.

[dAF2000]

 

Yes not funny if someone is decompiling our plugin without permission. 

And I've already fixed that issue.

 

 

Sorry, I just searched a solution. Maybe a mod can remove the attachment (I cannot edit it anymore), if you have a problem with it.

 

[modbreak=Done]Just removed it[/modbreak]

[dhwz]

Just a question.... why not using a build in virtual keyboard?

 

Because its a modified one, which is also displaying captcha codes and some other things. And importing the virtual keyboard of the image and modifiying it didn't work, thats what we did before but there are too many differences between some images. 

Edited by dhwz, 23 November 2014 - 22:29.

[Taapat]

@littlesat, maybe this is not the right topic, but if we talk about skin errors, then please change haligh to halign in skin_default.xml introduced in this commit: http://sourceforge.n...2de5542cc2ee42/

Edited by Taapat, 24 November 2014 - 08:46.

[littlesat]

Wauw good catch....

[Erik Slagter]

"Thanks" dhwz for supplying a plugin IN SOURCE so we can build it ourselves and there will be less chances for incompatibilities. Oh no, I was dreaming.

[littlesat]

And now I understand the real issue...

 

self.scale is a tupple here.... and later it becomes a function... but where it is called it is still a tupple...

This bug was in there for ever....

I suggest to rename the tupple so it is capable to call the scale function.... 

 

 

def __init__(self, guiObject, desktop, scale = ((1,1),(1,1))):

self.guiObject = guiObject

self.desktop = desktop

self.scale = scale

def applyOne(self, attrib, value):

try:

getattr(self, attrib)(value)

except AttributeError:

print "[Skin] Attribute not implemented:", attrib, "value:", value

except SkinError, ex:

print "[Skin] Error:", ex

except:

print "[Skin] Error:", attrib

Edited by littlesat, 24 November 2014 - 19:45.

[littlesat]

FYI:

http://sourceforge.n...512578722551a7/

Sometimes it is not just ignoring "the bug"...

 

But I understand... this is a typical "python" trap.....

 

In fact it was induced here...

http://sourceforge.n...656db7cedb33b1/

So in there since  2010-07-06.....  And indeed not a 'bug" in DMM's Enigma2, but slightly faster here as we do not have the big if/else...

Edited by littlesat, 24 November 2014 - 19:55.