fix(preprocess): parse shared exchange ticker lists

contexts/design/flow/news.md

@@ -230,10 +230,17 @@ three limited checks:
 3. fetch up to 25 unique articles and find the first supported exchange-coded
    symbol in every group that has one (100% recall).
 
+Ticker-hint extraction supports one exchange-qualified symbol and parenthesized
+comma-separated symbol lists that share a supported exchange prefix. A shared
+group ends at a semicolon, conjunction, or closing parenthesis. Additional list
+members record a reconstructed `EXCHANGE: SYMBOL` raw value; unsupported
+exchanges remain outside the whitelist policy.
+
 The ticker check uses a separate comparison regex rather than the production
 extractor. It accepts plain, Markdown-link, and emphasis-wrapped symbols. It
-ignores later symbols in a multi-symbol list and exchanges that the extractor
-does not support. The script prints PASS/FAIL plus short listing, article,
+recognizes comma-separated symbols that share a supported exchange prefix and
+ignores exchanges that the extractor does not support. The script prints
+PASS/FAIL plus short listing, article,
 failure, and recall summaries. It fails when RSS or listing data is invalid,
 no sampled article parses, or supported symbols fall below 100% recall. A
 parsed sample with no supported symbols reports `SKIP` and still passes.

examples/preprocess/02_news_ticker_lists.py

@@ -0,0 +1,8 @@
+"""Extract ticker hints from a shared exchange-prefix list."""
+
+from quantmind.preprocess import extract_exchange_ticker_hints
+
+text = "Example issuer (NYSE: EVEX, EVEXW; B3: EVEB31) announced results."
+
+for hint in extract_exchange_ticker_hints(text):
+    print(hint)

quantmind/preprocess/news.py

@@ -56,6 +56,10 @@
     r"(?<!\*)\*{1,2}([A-Z][A-Z0-9.-]{0,9})\*{1,2}(?!\*)",
     re.IGNORECASE,
 )
+_SHARED_EXCHANGE_SYMBOL_RE = re.compile(
+    r"\s*,\s*([A-Z][A-Z0-9.-]{0,9})",
+    re.IGNORECASE,
+)
 _EMAIL_PROTECTION_LINK_RE = re.compile(
     r"\[\[email protected]\]\(/cdn-cgi/l/email-protection#[^)]+\)"
 )
@@ -401,8 +405,12 @@ def canonicalize_source_url(url: str) -> str:
 def extract_exchange_ticker_hints(text: str) -> tuple[NewsTickerHint, ...]:
     """Extract exchange-qualified ticker mentions from PR-style text.
 
-    Examples matched include ``(NASDAQ: NVDA)`` and ``NYSE: IBM``. The result is
-    only a hint; downstream instrument resolution should still validate it.
+    Examples matched include ``(NASDAQ: NVDA)``, ``NYSE: IBM``, and a
+    parenthesized comma list such as ``(NYSE: EVEX, EVEXW)``. A shared exchange
+    applies only to comma-separated symbols before the closing parenthesis;
+    semicolons and conjunctions end the group. Additional members use a
+    reconstructed ``EXCHANGE: SYMBOL`` raw value. The result is only a hint;
+    downstream instrument resolution should still validate it.
     Markdown link and emphasis decoration is removed from a scan-only copy so
     stored news text and its content hash retain their original representation.
     """
@@ -425,6 +433,29 @@ def extract_exchange_ticker_hints(text: str) -> tuple[NewsTickerHint, ...]:
                 raw=match.group(0).strip(),
             )
         )
+        if not match.group(0).lstrip().startswith("("):
+            continue
+
+        closing_parenthesis = scan_text.find(")", match.end())
+        if closing_parenthesis == -1:
+            continue
+        suffix = scan_text[match.end() : closing_parenthesis]
+        position = 0
+        while continuation := _SHARED_EXCHANGE_SYMBOL_RE.match(
+            suffix, position
+        ):
+            symbol = continuation.group(1).upper()
+            key = (symbol, exchange)
+            if key not in seen:
+                seen.add(key)
+                hints.append(
+                    NewsTickerHint(
+                        symbol=symbol,
+                        exchange=exchange,
+                        raw=f"{exchange}: {symbol}",
+                    )
+                )
+            position = continuation.end()
     return tuple(hints)
 
 

tests/preprocess/test_news.py

@@ -117,6 +117,45 @@ def test_exchange_ticker_hints_ignore_markdown_decoration(self):
                     expected,
                 )
 
+    def test_exchange_ticker_hints_capture_shared_prefix_comma_lists(self):
+        cases = (
+            (
+                "supported shared prefix",
+                "(NYSE: EVEX, EVEXW; B3: EVEB31)",
+                (
+                    ("EVEX", "NYSE", "(NYSE: EVEX"),
+                    ("EVEXW", "NYSE", "NYSE: EVEXW"),
+                ),
+            ),
+            ("unsupported exchange", "(OTCID: QVCAQ, QVCGQ, QVCPQ)", ()),
+            (
+                "conjunction boundary",
+                "(NYSE: TME and HKEX: 1698)",
+                (("TME", "NYSE", "(NYSE: TME"),),
+            ),
+            (
+                "semicolon boundary TSXV",
+                "(NASDAQ: VMAR; TSXV: VMAR)",
+                (("VMAR", "NASDAQ", "(NASDAQ: VMAR"),),
+            ),
+            (
+                "semicolon boundary BMV",
+                "(NYSE: ASR; BMV: ASUR)",
+                (("ASR", "NYSE", "(NYSE: ASR"),),
+            ),
+        )
+
+        for name, text, expected in cases:
+            with self.subTest(name=name):
+                hints = extract_exchange_ticker_hints(text)
+
+                self.assertEqual(
+                    tuple(
+                        (hint.symbol, hint.exchange, hint.raw) for hint in hints
+                    ),
+                    expected,
+                )
+
     def test_build_sec_news_identity(self):
         self.assertEqual(
             build_sec_news_identity(