This guide covers how to parse an XML content feed using a task node in SceneGraph. This is a continuation from Setting up a project of the SDK Development Guide.
Parsing an XML feed is one of the most common ways to retrieve content for Roku channels. This guide will be using this sample MRSS feed.
Steps:
- Create Task node
- Retrieve XML content feed
- Parse XML content feed
- Setup Task thread
- Addendum: Customizing grid size
First create a new XML file in the components folder called FeedParser.xml. All XML files need to be prefixed with:
<?xml version="1.0" encoding="UTF-8"?>Create a new component within this file called FeedParser that extends from the Task node class.
<component name="FeedParser" extends="Task">Next, create an interface field called content of type node that will be used to store each content item from the parsed XML feed.
<interface>
<field id="content" type="node" />
</interface>We also want to keep BrightScript code separate of SceneGraph. We can do this by referencing a separate .brs file using the <script> tag.
In the components folder, create a new file named FeedParser.brs. Now in FeedParser.xml, we can reference FeedParser.brs with the following <script> tag:
<script type="text/brightscript" uri = "pkg:/components/FeedParser.brs"/>FeedParser.xml should look like below:
<?xml version="1.0" encoding="UTF-8"?>
<component name="FeedParser" extends="Task">
<interface>
<field id="content" type="node" />
</interface>
<script type="text/brightscript" uri = "pkg:/components/FeedParser.brs"/>
</component>Next, we need to retrieve the feed and convert it to a string so that it can be parsed. In FeedParser.brs, create a new function called GetContentFeed() using the code below:
Function GetContentFeed() 'This function retrieves and parses the feed and stores each content item in a ContentNode
url = CreateObject("roUrlTransfer") 'component used to transfer data to/from remote servers
url.SetUrl("http://api.delvenetworks.com/rest/organizations/59021fabe3b645968e382ac726cd6c7b/channels/1cfd09ab38e54f48be8498e0249f5c83/media.rss")
rsp = url.GetToString() 'convert response into a string
responseXML = ParseXML(rsp) 'Roku includes it's own XML parsing method
End FunctionIn the last line, we'll pass the response string to a ParseXML() function that we'll cover next.
We now have to parse the response string using the ParseXML method. ParseXML is a function that will try to parse the response string on the XMLElement or return invalid if it fails.
Function ParseXML(str As String) As dynamic 'Takes in the content feed as a string
if str = invalid return invalid 'if the response is invalid, return invalid
xml = CreateObject("roXMLElement") '
if not xml.Parse(str) return invalid 'If the string cannot be parsed, return invalid
return xml 'returns parsed XML if not invalid
End FunctionIn GetContentFeed(), it will need to be modified to handle what ParseXML() returns:
Function GetContentFeed() 'This function retrieves and parses the feed and stores each content item in a ContentNode
url = CreateObject("roUrlTransfer") 'component used to transfer data to/from remote servers
url.SetUrl("http://api.delvenetworks.com/rest/organizations/59021fabe3b645968e382ac726cd6c7b/channels/1cfd09ab38e54f48be8498e0249f5c83/media.rss")
rsp = url.GetToString() 'convert response into a string
responseXML = ParseXML(rsp) 'Roku includes it's own XML parsing method
if responseXML<>invalid then 'Fall back in case Roku's built in XML Parse method fails
responseXML = responseXML.GetChildElements() 'Access content inside Feed
responseArray = responseXML.GetChildElements()
End if
End FunctionIn a fallback case when invalid is returned, the response string will have to be parsed manually. We'll modify GetContentFeed() one more time using the code below:
Function GetContentFeed() 'This function retrieves and parses the feed and stores each content item in a ContentNode
url = CreateObject("roUrlTransfer") 'component used to transfer data to/from remote servers
url.SetUrl("http://api.delvenetworks.com/rest/organizations/59021fabe3b645968e382ac726cd6c7b/channels/1cfd09ab38e54f48be8498e0249f5c83/media.rss")
rsp = url.GetToString() 'convert response into a string
responseXML = ParseXML(rsp) 'Roku includes it's own XML parsing method
if responseXML<>invalid then 'Fall back in case Roku's built in XML Parse method fails
responseXML = responseXML.GetChildElements() 'Access content inside Feed
responseArray = responseXML.GetChildElements()
End if
'manually parse feed if ParseXML() is invalid
result = [] 'Store all results inside an array. Each element represents a row inside our RowList stored as an Associative Array (line 63)
for each xmlItem in responseArray 'For loop to grab content inside each item in the XML feed
if xmlItem.getName() = "item" 'Each individual channel content is stored inside the XML header named <item>
itemAA = xmlItem.GetChildElements() 'Get the child elements of item
if itemAA <> invalid 'Fall back in case invalid is returned
item = {} 'Creates an Associative Array for each row
for each xmlItem in itemAA 'Goes through all content of itemAA
item[xmlItem.getName()] = xmlItem.getText()
if xmlItem.getName() = "media:content" 'Checks to find <media:content> header
item.stream = {url : xmlItem.url} 'Assigns all content inside <media:content> to the item AA
item.url = xmlItem.getAttributes().url
item.streamFormat = "mp4"
mediaContent = xmlItem.GetChildElements()
for each mediaContentItem in mediaContent 'Looks through MediaContent to find poster images for each piece of content
if mediaContentItem.getName() = "media:thumbnail"
item.HDPosterUrl = mediaContentItem.getattributes().url 'Assigns images to item AA
item.hdBackgroundImageUrl = mediaContentItem.getattributes().url
end if
end for
end if
end for
result.push(item) 'Pushes each AA into the Array
end if
end if
end for
return result ' Returns the array
End Functionℹ️ The following code is specific to the schema of the sample feed used in this example. Your own content feed may vary.
Next, we setup the task node to call these functions when initialized. Because the init() method spawns on the render thread and we want the task node to run on a separate task thread, we setup the init() function to spawn a separate function on the task thread so that content can be loaded asynchronously.
Sub Init()
m.top.functionName = "loadContent"
End Sub
Sub loadContent()
list = GetContentFeed()
m.top.content = ParseXMLContent(list)
End Sublist is passed into a separate function named ParseXMLContent that takes all of the content and assigns them to content nodes that can be passed into a RowList component. The formatted content node is then passed into the content field of the task node so it can accessed from outside the scope of the task thread.
Function ParseXMLContent(list As Object) 'Formats content into content nodes so they can be passed into the RowList
RowItems = createObject("RoSGNode","ContentNode")
'Content node format for RowList: ContentNode(RowList content) --<Children>-> ContentNodes for each row --<Children>-> ContentNodes for each item in the row)
for each rowAA in list
row = createObject("RoSGNode","ContentNode")
row.Title = rowAA.Title
for each itemAA in rowAA.ContentList
item = createObject("RoSGNode","ContentNode")
item.SetFields(itemAA)
row.appendChild(item)
end for
RowItems.appendChild(row)
end for
return RowItems
End FunctionContinue to the next guide on Building a User Interface in SceneGraph.