From e229b567ad7e8b7ea8ef00ee1cb2dace93fd513a Mon Sep 17 00:00:00 2001
From: Chaoying <32626585+Chaoyingz@users.noreply.github.com>
Date: Thu, 24 Mar 2022 19:49:25 +0800
Subject: [PATCH] Support feature names contain Chinese punctuation (#1003)

---
 qlib/utils/__init__.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/qlib/utils/__init__.py b/qlib/utils/__init__.py
index 8852155bc..de60b18b7 100644
--- a/qlib/utils/__init__.py
+++ b/qlib/utils/__init__.py
@@ -274,9 +274,11 @@ def parse_field(field):
 
     if not isinstance(field, str):
         field = str(field)
+    # Chinese Punctuation Regex: \nff08 = (, \nff09 = ), \nff1a = :, \n3001 = 。
+    chinese_punctuation_regex = r"\u3001\uff1a\uff08\uff09"
     for pattern, new in [
-        (r"\$\$(\w+)", r'PFeature("\1")'),  # $$ must be before $
-        (r"\$(\w+)", rf'Feature("\1")'),
+        (rf"\$\$([\w{chinese_punctuation_regex}]+)", r'PFeature("\1")'),  # $$ must be before $
+        (rf"\$([\w{chinese_punctuation_regex}]+)", r'Feature("\1")'),
         (r"(\w+\s*)\(", r"Operators.\1("),
     ]:  # Features  # Operators
         field = re.sub(pattern, new, field)