fix: Keep single-element RSS feed elements as lists

src/edgar_tool/rss.py

@@ -17,6 +17,10 @@
 RSS_COMPANY_TICKERS_URL = "https://www.sec.gov/files/company_tickers.json"
 UNKNOWN_TICKER_PLACEHOLDER = "UNKNOWN"
 
+# Elements that repeat in the feed and are always iterated over. xmltodict collapses a
+# single-element list into a plain dict, so these are pinned to lists explicitly.
+FORCE_LIST_ELEMENTS = ("item", "edgar:xbrlFile")
+
 
 def _fetch_company_tickers(
     request_headers: Dict[str, Any], refresh_tickers_mapping: bool
@@ -147,8 +151,12 @@ def parse_rss_feed_data(
     :return: Iterator of parsed dicts for each item in the RSS feed
     """
 
-    # Parse RSS feed data and get all items
-    items = xmltodict.parse(response.content)["rss"]["channel"]["item"]
+    # Parse RSS feed data and get all items. force_list keeps the repeated elements as
+    # lists even when the feed carries exactly one of them; without it xmltodict
+    # collapses a single element to a dict and the loops below iterate over its keys.
+    items = xmltodict.parse(response.content, force_list=FORCE_LIST_ELEMENTS)["rss"][
+        "channel"
+    ]["item"]
     for i in items:
 
         try:

tests/test_rss_parsing.py

@@ -0,0 +1,102 @@
+import pytest
+
+from edgar_tool.rss import parse_rss_feed_data
+
+TICKERS_MAPPING = {"320193": ["AAPL"]}
+
+XBRL_FILE_HTM = '<edgar:xbrlFile edgar:url="https://example.invalid/a.htm"/>'
+XBRL_FILE_XSD = '<edgar:xbrlFile edgar:url="https://example.invalid/a.xsd"/>'
+
+
+def _item(xbrl_files: str) -> str:
+    return f"""
+    <item>
+      <title>8-K - EXAMPLE CORP</title>
+      <link>https://example.invalid/a.htm</link>
+      <description>8-K</description>
+      <pubDate>Mon, 03 Aug 2026 16:30:00 EDT</pubDate>
+      <edgar:xbrlFiling>
+        <edgar:companyName>EXAMPLE CORP</edgar:companyName>
+        <edgar:formType>8-K</edgar:formType>
+        <edgar:filingDate>08/03/2026</edgar:filingDate>
+        <edgar:cikNumber>0000320193</edgar:cikNumber>
+        <edgar:accessionNumber>0000320193-26-000001</edgar:accessionNumber>
+        <edgar:fileNumber>001-36743</edgar:fileNumber>
+        <edgar:acceptanceDatetime>20260803163000</edgar:acceptanceDatetime>
+        <edgar:period>20260803</edgar:period>
+        <edgar:assistantDirector>Office of Technology</edgar:assistantDirector>
+        <edgar:assignedSic>3571</edgar:assignedSic>
+        <edgar:fiscalYearEnd>0927</edgar:fiscalYearEnd>
+        <edgar:xbrlFiles>
+          {xbrl_files}
+        </edgar:xbrlFiles>
+      </edgar:xbrlFiling>
+    </item>
+"""
+
+
+def _feed(items: str) -> bytes:
+    return (
+        '<?xml version="1.0" encoding="ISO-8859-1"?>\n'
+        '<rss version="2.0" xmlns:edgar="https://www.sec.gov/Archives/edgar">\n'
+        f"  <channel>\n    <title>Latest Filings</title>\n{items}  </channel>\n</rss>\n"
+    ).encode()
+
+
+class _FakeResponse:
+    def __init__(self, content: bytes):
+        self.content = content
+
+
+@pytest.mark.parametrize(
+    "item_count,xbrl_files,expected_xbrl_files",
+    [
+        pytest.param(
+            2,
+            f"{XBRL_FILE_HTM}\n{XBRL_FILE_XSD}",
+            ["https://example.invalid/a.htm", "https://example.invalid/a.xsd"],
+            id="many-items-many-files",
+        ),
+        pytest.param(
+            1,
+            f"{XBRL_FILE_HTM}\n{XBRL_FILE_XSD}",
+            ["https://example.invalid/a.htm", "https://example.invalid/a.xsd"],
+            id="single-item",
+        ),
+        pytest.param(
+            2,
+            XBRL_FILE_HTM,
+            "https://example.invalid/a.htm",
+            id="single-xbrl-file",
+        ),
+        pytest.param(
+            1,
+            XBRL_FILE_HTM,
+            "https://example.invalid/a.htm",
+            id="single-item-and-single-xbrl-file",
+        ),
+    ],
+)
+def test_parses_feeds_with_single_repeated_elements(
+    item_count, xbrl_files, expected_xbrl_files
+):
+    """
+    xmltodict collapses a single-element list into a plain dict, so a feed with exactly
+    one <item>, or an item with exactly one <edgar:xbrlFile>, used to make the parser
+    iterate over dict keys and drop every item.
+
+    Note `xbrl_files` is a bare string rather than a one-element list when the filing has
+    a single file, because `unpack_singleton_list` unwraps it on the way out.
+    """
+    # GIVEN
+    response = _FakeResponse(_feed(_item(xbrl_files) * item_count))
+
+    # WHEN
+    rows = list(parse_rss_feed_data(response, ["AAPL"], TICKERS_MAPPING))
+
+    # THEN
+    assert len(rows) == item_count
+    assert all(r["ticker"] == "AAPL" for r in rows)
+    assert all(r["company_name"] == "EXAMPLE CORP" for r in rows)
+    assert all(r["cik"] == "0000320193" for r in rows)
+    assert all(r["xbrl_files"] == expected_xbrl_files for r in rows)