From e9798bf021847689084f951de3d0111d97167ea5 Mon Sep 17 00:00:00 2001
From: bozarnr <97791974+bozarnr@users.noreply.github.com>
Date: Fri, 14 Aug 2026 21:05:43 +0800
Subject: [PATCH 1/3] Fix shared-prefix ticker hint extraction

---
 quantmind/preprocess/news.py  | 37 +++++++++++++++++++++++++++++------
 tests/preprocess/test_news.py | 33 +++++++++++++++++++++++++++++++
 2 files changed, 64 insertions(+), 6 deletions(-)

diff --git a/quantmind/preprocess/news.py b/quantmind/preprocess/news.py
index 12e7706..31d3695 100644
--- a/quantmind/preprocess/news.py
+++ b/quantmind/preprocess/news.py
@@ -56,6 +56,10 @@
     r"(?<!\*)\*{1,2}([A-Z][A-Z0-9.-]{0,9})\*{1,2}(?!\*)",
     re.IGNORECASE,
 )
+_EXCHANGE_TICKER_LIST_MEMBER_RE = re.compile(
+    r"\s*,\s*([A-Z][A-Z0-9.-]{0,9})\b",
+    re.IGNORECASE,
+)
 _EMAIL_PROTECTION_LINK_RE = re.compile(
     r"\[\[email protected]\]\(/cdn-cgi/l/email-protection#[^)]+\)"
 )
@@ -410,21 +414,42 @@ def extract_exchange_ticker_hints(text: str) -> tuple[NewsTickerHint, ...]:
     scan_text = _MARKDOWN_EMPHASIS_RE.sub(r"\1", scan_text)
     hints: list[NewsTickerHint] = []
     seen: set[tuple[str, str | None]] = set()
-    for match in _EXCHANGE_TICKER_RE.finditer(scan_text):
-        raw_exchange = " ".join(match.group(1).upper().split())
-        exchange = _EXCHANGE_NAMES.get(raw_exchange, raw_exchange)
-        symbol = match.group(2).upper()
+
+    def append_hint(
+        *,
+        symbol: str,
+        exchange: str,
+        raw: str,
+    ) -> None:
         key = (symbol, exchange)
         if key in seen:
-            continue
+            return
         seen.add(key)
         hints.append(
             NewsTickerHint(
                 symbol=symbol,
                 exchange=exchange,
-                raw=match.group(0).strip(),
+                raw=raw,
             )
         )
+
+    for match in _EXCHANGE_TICKER_RE.finditer(scan_text):
+        raw_exchange = " ".join(match.group(1).upper().split())
+        exchange = _EXCHANGE_NAMES.get(raw_exchange, raw_exchange)
+        symbol = match.group(2).upper()
+        append_hint(
+            symbol=symbol,
+            exchange=exchange,
+            raw=match.group(0).strip(),
+        )
+
+        tail = re.split(r"[;)]", scan_text[match.end() :], maxsplit=1)[0]
+        for list_match in _EXCHANGE_TICKER_LIST_MEMBER_RE.finditer(tail):
+            append_hint(
+                symbol=list_match.group(1).upper(),
+                exchange=exchange,
+                raw=f"{raw_exchange}: {list_match.group(1).upper()}",
+            )
     return tuple(hints)
 
 
diff --git a/tests/preprocess/test_news.py b/tests/preprocess/test_news.py
index 7a0b634..c4db38c 100644
--- a/tests/preprocess/test_news.py
+++ b/tests/preprocess/test_news.py
@@ -117,6 +117,39 @@ def test_exchange_ticker_hints_ignore_markdown_decoration(self):
                     expected,
                 )
 
+    def test_exchange_ticker_hints_capture_shared_prefix_lists(self):
+        cases = (
+            (
+                "(OTCID: QVCAQ, QVCGQ, QVCPQ)",
+                [],
+            ),
+            (
+                "(NYSE: EVEX, EVEXW; B3: EVEB31)",
+                [("EVEX", "NYSE"), ("EVEXW", "NYSE")],
+            ),
+            (
+                "(NYSE: TME and HKEX: 1698)",
+                [("TME", "NYSE")],
+            ),
+            (
+                "(NASDAQ: VMAR; TSXV: VMAR)",
+                [("VMAR", "NASDAQ")],
+            ),
+            (
+                "(NYSE: ASR; BMV: ASUR)",
+                [("ASR", "NYSE")],
+            ),
+        )
+
+        for text, expected in cases:
+            with self.subTest(text=text):
+                hints = extract_exchange_ticker_hints(text)
+
+                self.assertEqual(
+                    [(hint.symbol, hint.exchange) for hint in hints],
+                    expected,
+                )
+
     def test_build_sec_news_identity(self):
         self.assertEqual(
             build_sec_news_identity(

From 76ece5b9d4c15a09b8bc79783edd3c3ea1fda6c2 Mon Sep 17 00:00:00 2001
From: Newman Gao <newmangao@outlook.com>
Date: Sat, 15 Aug 2026 15:09:02 +0800
Subject: [PATCH 2/3] Keep shared-prefix ticker members uppercase

---
 quantmind/preprocess/news.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/quantmind/preprocess/news.py b/quantmind/preprocess/news.py
index 31d3695..33427ce 100644
--- a/quantmind/preprocess/news.py
+++ b/quantmind/preprocess/news.py
@@ -57,8 +57,7 @@
     re.IGNORECASE,
 )
 _EXCHANGE_TICKER_LIST_MEMBER_RE = re.compile(
-    r"\s*,\s*([A-Z][A-Z0-9.-]{0,9})\b",
-    re.IGNORECASE,
+    r"\s*,\s*([A-Z][A-Z0-9.-]{0,9})\b"
 )
 _EMAIL_PROTECTION_LINK_RE = re.compile(
     r"\[\[email protected]\]\(/cdn-cgi/l/email-protection#[^)]+\)"

From c89b53259a5884afa1aae9a723d69caa7e93e26b Mon Sep 17 00:00:00 2001
From: Newman Gao <newmangao@outlook.com>
Date: Sat, 15 Aug 2026 15:09:06 +0800
Subject: [PATCH 3/3] Test lowercase prose after ticker lists

---
 tests/preprocess/test_news.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/tests/preprocess/test_news.py b/tests/preprocess/test_news.py
index c4db38c..64d9cbd 100644
--- a/tests/preprocess/test_news.py
+++ b/tests/preprocess/test_news.py
@@ -127,6 +127,10 @@ def test_exchange_ticker_hints_capture_shared_prefix_lists(self):
                 "(NYSE: EVEX, EVEXW; B3: EVEB31)",
                 [("EVEX", "NYSE"), ("EVEXW", "NYSE")],
             ),
+            (
+                "(NASDAQ: ABC, a leading provider announced results)",
+                [("ABC", "NASDAQ")],
+            ),
             (
                 "(NYSE: TME and HKEX: 1698)",
                 [("TME", "NYSE")],