Spaces:
Running
Running
Create utils.py
Browse files- helpers/utils.py +51 -0
helpers/utils.py
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import re
|
| 2 |
+
from helpers.cleaner import clean_text
|
| 3 |
+
|
| 4 |
+
ARABIC_INDIC_DIGITS = str.maketrans("٠١٢٣٤٥٦٧٨٩", "0123456789")
|
| 5 |
+
|
| 6 |
+
def normalize_digits(s: str) -> str:
|
| 7 |
+
return s.translate(ARABIC_INDIC_DIGITS)
|
| 8 |
+
|
| 9 |
+
# ---------------------------------------
|
| 10 |
+
# تنظيف النصوص + تحويل الأرقام
|
| 11 |
+
# ---------------------------------------
|
| 12 |
+
def clean_text_block(text: str) -> str:
|
| 13 |
+
if not isinstance(text, str):
|
| 14 |
+
return ""
|
| 15 |
+
return clean_text(normalize_digits(text.strip()))
|
| 16 |
+
|
| 17 |
+
# ---------------------------------------
|
| 18 |
+
# التعرف على المواد القانونية (يشمل "مكرر" وحروف)
|
| 19 |
+
# ---------------------------------------
|
| 20 |
+
ARTICLE_PATTERN = re.compile(
|
| 21 |
+
r"^\s*(?:المادة|مادة)\s*[\(\s]*([0-9٠-٩]+(?:\s*مكرر)?)\s*[\)\s]*([أ-ي\-]*)",
|
| 22 |
+
re.IGNORECASE
|
| 23 |
+
)
|
| 24 |
+
|
| 25 |
+
def is_article(text: str) -> bool:
|
| 26 |
+
if not isinstance(text, str):
|
| 27 |
+
return False
|
| 28 |
+
text_norm = normalize_digits(text)
|
| 29 |
+
return bool(ARTICLE_PATTERN.match(text_norm))
|
| 30 |
+
|
| 31 |
+
def extract_article_number(text: str):
|
| 32 |
+
if not isinstance(text, str):
|
| 33 |
+
return None
|
| 34 |
+
text_norm = normalize_digits(text)
|
| 35 |
+
m = ARTICLE_PATTERN.match(text_norm)
|
| 36 |
+
if m:
|
| 37 |
+
return m.group(1).replace(" ", "") # إزالة المسافات في "مكرر"
|
| 38 |
+
return None
|
| 39 |
+
|
| 40 |
+
# ---------------------------------------
|
| 41 |
+
# التعرف على الأقسام (الباب / الفصل / القسم)
|
| 42 |
+
# ---------------------------------------
|
| 43 |
+
SECTION_PATTERN = re.compile(
|
| 44 |
+
r"^\s*(الباب|الفصل|القسم)\s+([^\n]+)",
|
| 45 |
+
re.IGNORECASE
|
| 46 |
+
)
|
| 47 |
+
|
| 48 |
+
def is_section(text: str) -> bool:
|
| 49 |
+
if not isinstance(text, str):
|
| 50 |
+
return False
|
| 51 |
+
return bool(SECTION_PATTERN.match(text))
|