From 3857bd8ded0246c57d81612be7c5c28d86a07747 Mon Sep 17 00:00:00 2001
From: arsam <me@asabbagh.com>
Date: Fri, 18 Sep 2026 22:41:11 +0330
Subject: [PATCH] Add text chunking and Persian language helpers

---
 utils/chunker.py | 26 ++++++++++++++++++++++++++
 utils/persian.py | 45 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 71 insertions(+)
 create mode 100644 utils/chunker.py
 create mode 100644 utils/persian.py

diff --git a/utils/chunker.py b/utils/chunker.py
new file mode 100644
index 0000000..12199df
--- /dev/null
+++ b/utils/chunker.py
@@ -0,0 +1,26 @@
+TG_LIMIT = 4096
+
+
+def chunk_text(text: str, max_length: int = TG_LIMIT) -> list[str]:
+    """Split text into ≤max_length chunks, preferring newline then space boundaries."""
+    if len(text) <= max_length:
+        return [text]
+
+    chunks: list[str] = []
+    while text:
+        if len(text) <= max_length:
+            chunks.append(text)
+            break
+
+        split_at = text.rfind('\n', 0, max_length)
+        if split_at == -1:
+            split_at = text.rfind(' ', 0, max_length)
+        if split_at == -1:
+            split_at = max_length
+
+        chunk = text[:split_at].strip()
+        if chunk:
+            chunks.append(chunk)
+        text = text[split_at:].strip()
+
+    return chunks
diff --git a/utils/persian.py b/utils/persian.py
new file mode 100644
index 0000000..7501a03
--- /dev/null
+++ b/utils/persian.py
@@ -0,0 +1,45 @@
+import re
+import unicodedata
+
+# Normalize Arabic lookalikes to Persian
+_ARABIC_TO_PERSIAN = str.maketrans({
+    'ي': 'ی',  # Arabic ya  → Persian ya
+    'ك': 'ک',  # Arabic kaf → Persian kaf
+    'ى': 'ی',  # Alef maqsura → Persian ya
+})
+
+_PUPPY_RE = re.compile(r'puppy|پاپی|پاپي', re.IGNORECASE | re.UNICODE)
+
+# Rough threshold: if >15% of chars are Arabic/Persian script → Persian
+_PERSIAN_BLOCK = re.compile(r'[؀-ۿ]')
+
+
+def normalize(text: str) -> str:
+    text = unicodedata.normalize('NFC', text)
+    text = text.translate(_ARABIC_TO_PERSIAN)
+    text = text.replace('‌', '').replace('‍', '')  # ZWNJ / ZWJ
+    return text
+
+
+def is_puppy_mentioned(text: str) -> bool:
+    if not text:
+        return False
+    return bool(_PUPPY_RE.search(normalize(text)))
+
+
+def detect_language(text: str) -> str:
+    if not text:
+        return 'en'
+    persian_count = len(_PERSIAN_BLOCK.findall(text))
+    return 'fa' if persian_count > len(text) * 0.15 else 'en'
+
+
+def strip_puppy_prefix(text: str) -> str:
+    """Remove 'Puppy/پاپی' name prefix from a command-style message."""
+    cleaned = re.sub(
+        r'^[\s]*(puppy|پاپی|پاپي)[،,:\s]+',
+        '',
+        normalize(text),
+        flags=re.IGNORECASE | re.UNICODE,
+    )
+    return cleaned.strip() or text.strip()