From 54ab172c6af0e97c2771dfb1c577962955220820 Mon Sep 17 00:00:00 2001
From: phoneee <phoneee@gmail.com>
Date: Sun, 29 Mar 2026 16:43:22 +0700
Subject: [PATCH 1/2] fix: guard han_solo list_cut[-1] and narrow longest.py
 exception catch

han_solo: if CRF predicts "0" for first char, list_cut is empty and
list_cut[-1] raises IndexError. Add fallback to append when empty.

longest: except BaseException catches KeyboardInterrupt and SystemExit.
Narrow to except IndexError which is the only expected exception.
---
 pythainlp/tokenize/han_solo.py | 2 +-
 pythainlp/tokenize/longest.py  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/pythainlp/tokenize/han_solo.py b/pythainlp/tokenize/han_solo.py
index dc1b18ecf..efcbf43b0 100644
--- a/pythainlp/tokenize/han_solo.py
+++ b/pythainlp/tokenize/han_solo.py
@@ -162,7 +162,7 @@ def segment(text: str) -> list[str]:
     y_pred = tagger.tag(x)
     list_cut = []
     for j, k in zip(list(text), y_pred):
-        if k == "1":
+        if k == "1" or not list_cut:
             list_cut.append(j)
         else:
             list_cut[-1] += j
diff --git a/pythainlp/tokenize/longest.py b/pythainlp/tokenize/longest.py
index e2453c7de..3ea12d599 100644
--- a/pythainlp/tokenize/longest.py
+++ b/pythainlp/tokenize/longest.py
@@ -105,7 +105,7 @@ def __longest_matching(self, text: str, begin_pos: int) -> str:
                     return text[0 : len_word_valid + 1]
                 else:
                     return word_valid
-            except BaseException:
+            except IndexError:
                 return word_valid
         else:
             return ""

From 06e016ff02da758411da351d511f11a681d81cbf Mon Sep 17 00:00:00 2001
From: Arthit Suriyawongkul <arthit@gmail.com>
Date: Mon, 30 Mar 2026 19:07:56 +0100
Subject: [PATCH 2/2] Update pythainlp/tokenize/han_solo.py

---
 pythainlp/tokenize/han_solo.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pythainlp/tokenize/han_solo.py b/pythainlp/tokenize/han_solo.py
index efcbf43b0..4e455f394 100644
--- a/pythainlp/tokenize/han_solo.py
+++ b/pythainlp/tokenize/han_solo.py
@@ -160,7 +160,7 @@ def segment(text: str) -> list[str]:
     tagger = _get_tagger()
     x = _to_feature.featurize(text)["X"]
     y_pred = tagger.tag(x)
-    list_cut = []
+    list_cut: list[str] = []
     for j, k in zip(list(text), y_pred):
         if k == "1" or not list_cut:
             list_cut.append(j)