News & RSS Ticker

Fetch and parse RSS headlines, clean their text, and display them as sequential marquees. The demo feed keeps this source runnable without a network request.

Complete app

app.star

load("bsoup.star", "bsoup")
load("html.star", "html")
load("http.star", "http")
load("re.star", "re")
load("render.star", "render")
load("schema.star", "schema")
load("xpath.star", "xpath")

DEMO_MODE = True
DEMO_RSS = """
<rss><channel>
  <item><title>Pixel displays brighten the morning commute</title></item>
  <item><title>Developers ship a tiny weather dashboard</title></item>
</channel></rss>
"""

def fetch_feed():
    response = http.get("https://news.example/feed.xml", ttl_seconds = 900)
    if response.status_code != 200:
        fail("Feed request failed: %d" % response.status_code)
    return response.body()

def source_name():
    markup = "<header><span class='source'>PIXEL WIRE</span></header>"
    return html(markup).find(".source").text()

def source_kicker():
    soup = bsoup.parseHtml("<p><strong>LIVE</strong> updates</p>")
    return soup.find("strong").get_text()

def main():
    feed = DEMO_RSS if DEMO_MODE else fetch_feed()
    titles = xpath.loads(feed).query_all("/rss/channel/item/title")
    clean_titles = [re.sub(r"\s+", " ", title) for title in titles]
    label = "%s · %s" % (source_kicker(), source_name())

    slides = []
    for title in clean_titles:
        slides.append(
            render.Column(
                children = [
                    render.Text(label, font = "CG-pixel-3x5-mono", color = "#fb7185"),
                    render.Marquee(
                        width = 64,
                        child = render.Text(title, color = "#f8fafc"),
                        offset_start = 2,
                        offset_end = 8,
                    ),
                ],
            ),
        )

    return render.Root(
        delay = 50,
        child = render.Sequence(children = slides),
    )

def search_sources(pattern):
    options = [
        schema.Option(display = "Pixel Wire", value = "pixel-wire"),
        schema.Option(display = "City Desk", value = "city-desk"),
    ]
    query = pattern.lower()
    return [option for option in options if query in option.display.lower()]

def get_schema():
    return schema.Schema(
        version = "1",
        fields = [
            schema.Typeahead(
                id = "source",
                name = "News source",
                desc = "Search for a news source.",
                icon = "newspaper",
                handler = search_sources,
            ),
        ],
    )

Download this app.star.

How it works

1. xpath selects every RSS title, then re.sub() normalizes whitespace.
2. html demonstrates CSS-selector extraction; bsoup extracts the nested kicker.
3. Each title becomes a Marquee, and Sequence advances through the headlines.
4. The typeahead handler filters schema options as the user types.
5. The live request has a 15-minute TTL to avoid repeatedly downloading the feed.

Connect a live feed

Replace news.example with an RSS URL and set DEMO_MODE to False. If the feed uses a different structure, update the XPath passed to query_all().