Yes, please post issues you encounter to this thread.
- Joel
Yes, please post issues you encounter to this thread.
“RokuJoel” wrote:
- Should be all players other than 3.1 legacy devices.
- Not really, but you can get a good idea from the Roku Home Screen and mGo channels.
We just released the beta, for trailblazers to get started blazing trails.
Thanks for the info.
Yeah, umm… i think i’ll wait for the “blaze of gory” :mrgreen: to subside a little.
Esp. i got myself something else really exciting to dig on
I’m not sure if this is a ‘bug’ or a ‘feature’ but I’m seeing a sensitivity to the order in which XML elements are defined. The XML file below defines:
(a) an upper and lower
(b) 3 elements that appear in the middle of the screen, and
(c) 3 elements
The issue is the behavior of the 3rd Label with id=“bottomLabelLeft”. The other two Labels are child nodes of a Rectangle but this one is not. If a element is defined prior to this label, the label will not be visible. To see what I mean, comment out the Posters with ids “avatarMiddle” and “avatarRt”.
Here’s the XML:
<?xml version="1.0" encoding="utf-8" ?>
<component name="testScene01" extends="Scene" >
<script type="text/brightscript" >
<![CDATA[
function init()
m.top.setFocus(true)
end function
]]>
</script>
<children>
<Poster
id="avatarMiddle"
uri="pkg:/images/Oak-Beard.jpg"
width="200"
height="200"
translation="[500,300]"
/>
<Rectangle
id="upperRectangle"
color="0xcc00ccFF"
width="640"
height="60"
translation="[300,120]">
<Label
id="upperLabel"
height="60"
width="0"
text = "Hello World!"
horizAlign = "left"
vertAlign = "center"
translation="[160,10]" />
</Rectangle>
<Poster
id="avatarRt"
uri="pkg:/images/Oak-Beard.jpg"
translation="[900,300]"
/>
<Rectangle
id="bottomRectangle"
color="0x0000FFFF"
width="1280"
height="60"
translation="[0,620]">
<Label
id="bottomLabelRight"
height="60"
width="0"
text = "Can you see me?"
horizAlign = "left"
vertAlign = "center"
translation="[500,0]"
/>
</Rectangle>
<Label
id="bottomLabelLeft"
height="60"
width="0"
text = "Goodbye World!"
horizAlign = "left"
vertAlign = "center"
translation="[160,620]"
/>
<Poster
id="avatarLeft"
uri="pkg:/images/Oak-Beard.jpg"
width="145"
height="145"
translation="[100,300]"
/>
</children>
</component>
I’m having difficulty with the debugger. I’ve created two telnet sessions: one to the 8085 port and one to the 8089 port. While the 8089 session displays the output of any Brightscript defined in the XML components, there is no response to any BrightScript debugger commands entered via this console. The 8085 session responds normally but apparently provides no way to interact with or examine the XML component scripts. Any suggestions?
Update: results are actually inconsistent as clarified in follow-up posting
The video.mute field seems to have no effect. Whether true or false, I’m always getting sound during playback.
“Oak-Beard” wrote:
… the 8089 session displays the output of any Brightscript defined in the XML components, there is no response to any BrightScript debugger commands entered via this console. … Any suggestions?
Sounds pretty… censored. How does it act if you include STOP statement inside the B/S XML?
“EnTerr” wrote:
“Oak-Beard” wrote:
… the 8089 session displays the output of any Brightscript defined in the XML components, there is no response to any BrightScript debugger commands entered via this console. … Any suggestions?
Sounds pretty… censored. How does it act if you include STOP statement inside the B/S XML?
An excellent question.. STOP seems to work OK but a Ctrl-C has inconsistent results. Could be linked to threading? Sometimes the telnet session will respond but only after several seconds. Other times I get nothing until I press a key on the control. It’s not the end of the world but it would be nice to know what’s going on.
I would like to be part of this beta. I want to test rotation of text, which doesn’t work properly in current Roku firmware.
AC
“crawfishmedia” wrote:
I would like to be part of this beta. I want to test rotation of text, which doesn’t work properly in current Roku firmware.The first message in the thread describes the procedure to request it by PM.
I’ve hit a snag when adding a component interface that uses the ‘alias’ attribute. Everything works correctly except when I attempt to alias the ‘state’ or ‘control’ attributes of a Video node. The documentation for a Video node defines the type of the control property as ‘option string’ and type(video.control) will indicate it is ‘roString’. However if I define the alias as
<field
id="control"
type="string"
value="none"
onChange="ctrlVideo"
alias="vidScreen.control"
/>
I get the message:
Interface field alias failed: Type “string” does not match type of the field named “control”
I should point out that aliasing works OK for all other Video properties including ‘errorMsg’ which is defined as type=string. The issue seems to be the ‘option string’ and ‘value string’ types which apparently have no direct corollary in the types available when defining a Component Interface.
Am I missing something I should be doing or is this an actual bug/feature?
I can’t seem to get any of the examples to work on my device. Each example hangs on the splash screen.
I’m using a ROKU 3050X and I get this from the console:
------ Running dev 'Rectangle' main ------
Firmware Version Found: 027.00E08230A
Major Version: 7 Minor Version: 00 Build Number: 08230
in showChannelSGScreen
Any suggestions?
I have a question regarding the rectangle node. It has a couple of fields regarding blending that I’m unclear on. There’s sourceBlendingFactor and destinationBlendingFactor, which have a type of “option string”. What does that mean? I’m a little unclear on what these are even used for. Does anybody know? Thanks
FYI - within brightscript, it is documented that colors are specified as ARGB (search this page for “AARRGGBB”). In scenegraph, it appears that they are specified as RGBA. This made some of the examples confusing to me until I realized this. ![]()
Bug:
Looking at the Z-order example, it works as I would expect, blue is above green, which is above red. But if I add another rectangle at the beginning (under everything), the green rectangle moves to the top. If I add yet another rectangle under everything, the red rectangle moves to the top. This doesn’t make sense.
<Rectangle
id="added2"
color="0xFF00FFFF"
width="400"
height="200"
translation="[900,260]" />
<Rectangle
id="added1"
color="0xFFFFFFFF"
width="400"
height="200"
translation="[800,260]" />
<Rectangle
id="lowestRectangle"
color="0xFF0000FF"
width="180"
height="60"
translation="[190,195]" />
<Rectangle
id="middleRectangle"
color="0x00FF00FF"
width="360"
height="120"
translation="[250,225]" />
<Rectangle
id="highestRectangle"
color="0x0000FFFF"
width="720"
height="240"
translation="[370,285]" />
I’ve got a Roku Streaming Stick running 7.0 build 904, but it’s no longer pairing with Roku remotes (tried several). Remote application on Android phone still works though.
Using that remote application I can install an update to 7.0 build 8236, but after installation the rebooting Roku does not go past the Roku bootup logo anymore. Only power off and on works, with stick booting in the broken 904 build again.
Any ideas on how to solve this?
Hi Everyone,
I have sent PM to RokuShawnS to activate my serial numbers for the SDK, but have not received a reply yet ( it has been 5 days). My dev team is anxious to start using it, can anyone please help with activation?
Thanks
@Yuriy_OM : I didn’t get a response when I added it (which Roku did very quickly). Have your developers see if they can run one of the scenegraph examples.
Oh I see. I thought I will get PM reply confirming it. I will check it now!
“adamkaz” wrote:
@Yuriy_OM: I didn’t get a response when I added it (which Roku did very quickly). Have your developers see if they can run one of the scenegraph examples.
Just wanted to confirm one thing - I should be able to software update to 7.0 to run the SDK? SO far none of my ROKUs want to update and are running 6.2. Just in case, here is the list again:
removed ..
Thanks
Did you send me a PM previously? I got backlogged on a few emails, but they all should have been updated as of Monday evening. I may have missed one or two.
C. Shawn Smith