From 5e30aa3c483629284a737b9b7e913e2012754e25 Mon Sep 17 00:00:00 2001
From: seonghobae <8172694+seonghobae@users.noreply.github.com>
Date: Wed, 15 Jul 2026 17:31:27 +0000
Subject: [PATCH] =?UTF-8?q?=E2=9A=A1=20Bolt:=20[=EC=9A=94=EC=95=BD=20?=
 =?UTF-8?q?=EA=B8=B0=EB=8A=A5=EC=9D=98=20=ED=86=A0=ED=81=B0=20=EC=A0=95?=
 =?UTF-8?q?=EA=B7=9C=EC=8B=9D=20=EC=9A=B0=ED=9A=8C(Fast=20Path)=20?=
 =?UTF-8?q?=EC=B5=9C=EC=A0=81=ED=99=94]?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 .jules/bolt.md | 3 +++
 summarize.py   | 8 ++++++++
 2 files changed, 11 insertions(+)

diff --git a/.jules/bolt.md b/.jules/bolt.md
index 341c7c91..2da6e608 100644
--- a/.jules/bolt.md
+++ b/.jules/bolt.md
@@ -67,3 +67,6 @@
 ## 2025-02-12 - [Fast Path Execution in Directory Traversal and Log Parsing]
 **Learning:** Checking for string existence (`if "silence_" not in stderr`) before invoking regex matchers provides significant speed improvements when parsing large blocks of text. Similarly, moving expensive I/O operations like `os.path.realpath` inside conditional blocks prevents redundant disk access when configuration (like path exclusions) isn't utilized.
 **Action:** When working on large text processing or disk operations, verify if early exit conditions or conditional execution can bypass the expensive system or library calls.
+## 2026-06-25 - [Fast Path for Alphanumeric Regex Replacements]
+**Learning:** Using regex substitution (e.g. `re.sub(r"^\W+|\W+$", "")`) in tight tokenization loops is extremely CPU-bound and slow. However, the majority of tokens in standard text are purely alphanumeric without any punctuation.
+**Action:** When tokenizing or processing text using regular expressions to strip non-word characters, prepend a fast-path check using `str.isalnum()` (or `str.isalpha()`) to bypass the regex overhead for purely alphanumeric words. This safely yields significant performance gains for most valid text inputs.
diff --git a/summarize.py b/summarize.py
index a5b56fe0..81e6b4e9 100644
--- a/summarize.py
+++ b/summarize.py
@@ -94,6 +94,14 @@ def _content_words(sentence):
     """
     words = []
     for raw in sentence.split():
+        # Fast path check: if raw is alphanumeric, skip regex overhead.
+        # This significantly improves performance because most tokens
+        # do not contain punctuation and safely bypass the regex substitution.
+        if raw.isalnum():
+            token = raw.lower()
+            if token not in _STOPWORDS:
+                words.append(token)
+            continue
         token = _TOKEN_STRIP_RE.sub("", raw).lower()
         if token and token not in _STOPWORDS:
             words.append(token)