Spaces:
Sleeping
Sleeping
Create parser/section_extractor.py
Browse files- parser/section_extractor.py +38 -0
parser/section_extractor.py
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# parser/section_extractor.py
|
| 2 |
+
import re
|
| 3 |
+
from typing import List, Dict, Any, Tuple
|
| 4 |
+
|
| 5 |
+
SECTION_KEYWORDS = ["الكتاب", "الباب", "الفصل"]
|
| 6 |
+
|
| 7 |
+
def is_section_line(line: str) -> bool:
|
| 8 |
+
return bool(re.match(rf"^(?:{'|'.join(SECTION_KEYWORDS)})\b", line))
|
| 9 |
+
|
| 10 |
+
def extract_sections_from_text_blocks(text_blocks: List[str]) -> Tuple[List[Dict[str, Any]], List[str]]:
|
| 11 |
+
sections = []
|
| 12 |
+
preamble = []
|
| 13 |
+
current = None
|
| 14 |
+
found = False
|
| 15 |
+
|
| 16 |
+
for block in text_blocks:
|
| 17 |
+
parts = re.split(r"(?<=\.)|(?=\b(?:الكتاب|الباب|الفصل)\b)", block)
|
| 18 |
+
|
| 19 |
+
for part in parts:
|
| 20 |
+
part = part.strip()
|
| 21 |
+
if not part:
|
| 22 |
+
continue
|
| 23 |
+
|
| 24 |
+
if is_section_line(part):
|
| 25 |
+
found = True
|
| 26 |
+
current = {"title": part, "lines": []}
|
| 27 |
+
sections.append(current)
|
| 28 |
+
continue
|
| 29 |
+
|
| 30 |
+
if not found:
|
| 31 |
+
preamble.append(part)
|
| 32 |
+
else:
|
| 33 |
+
if current is None:
|
| 34 |
+
current = {"title": "", "lines": []}
|
| 35 |
+
sections.append(current)
|
| 36 |
+
current["lines"].append(part)
|
| 37 |
+
|
| 38 |
+
return sections, preamble
|