update reader pdf by page

main.py

@@ -225,7 +225,7 @@ async def server_status():
 
 
 @app.post("/upload")
-async def upload_file(file: UploadFile = File(...), page: Optional[str] = Form(None)):
+async def upload_file(file: UploadFile = File(...), page: Optional[str] = Form("")):
     fname = file.filename
     ext = os.path.splitext(fname)[1].lower()
     contents = await file.read()
@@ -246,7 +246,7 @@ async def upload_file(file: UploadFile = File(...), page: Optional[str] = Form(N
         elif ext == ".xlsx":
             df = read_csv(str(tmp_path))
         elif ext == ".pdf":
-            tbl = read_pdf(tmp_path)
+            tbl = read_pdf(tmp_path, page)
             if len(tbl) == 0:
                 response = {
                     "message": "Tidak ditemukan tabel valid",

services/reader_pdf.py

@@ -1,6 +1,7 @@
 import pdfplumber
 import re
 import pandas as pd
+from services.filter_column import filter_geo_admin_column
 
 def is_number(s):
     if s is None:
@@ -144,47 +145,82 @@ def cleaning_column(headers, bodies):
 
 
 
+def parse_page_selection(selectedPage: str, total_pages: int):
+    if not selectedPage:
+        return list(range(1, total_pages + 1))
 
-def read_pdf(path: str):
+    pages = set()
+    parts = re.split(r'[,\s]+', selectedPage.strip())
+
+    for part in parts:
+        if '-' in part:
+            try:
+                start, end = map(int, part.split('-'))
+                pages.update(range(start, end + 1))
+            except ValueError:
+                continue
+        else:
+            try:
+                pages.add(int(part))
+            except ValueError:
+                continue
+
+    valid_pages = [p for p in sorted(pages) if 1 <= p <= total_pages]
+    return valid_pages
+
+
+
+def read_pdf(path: str, page: str):
     pdf_path = path
+    selectedPage = page
     tables_data = [] 
     with pdfplumber.open(pdf_path) as pdf:
-        page = pdf.pages[0]
+        total_pages = len(pdf.pages)
-        tables = page.find_tables()
+        selected_pages = parse_page_selection(selectedPage, total_pages)
-        for i, t in enumerate(tables, start=1):
-            table = t.extract()  
-            if len(table) > 4:
-                tables_data.append(table)
 
-        print(f"\nTotal tabel valid: {len(tables_data)}\n")
+        print(f"[INFO] Total halaman PDF: {total_pages}")
+        print(f"[INFO] Halaman yang dipilih untuk dibaca: {selected_pages}")
 
-        header_only = []
+        for page_num in selected_pages:
-        body_only = []
+            pdf_page = pdf.pages[page_num - 1]  # index pdfplumber mulai dari 0
-        for tbl in tables_data:
+            tables = pdf_page.find_tables()
-            head, body = detect_header_rows(tbl)
+            print(f"[INFO] Halaman {page_num}: {len(tables)} tabel terdeteksi")
-            header_only.append(head)
-            body_only.append(body)
 
-        clean_header = []
+            for t in tables:
-        for h in header_only:
+                table = t.extract()
-            clean_header.append(merge_multiline_header(h))
+                if len(table) > 2:
+                    tables_data.append(table)
     
-        clean_body=[]
+    print(f"\nTotal tabel valid: {len(tables_data)}\n")
-        for i, raw_body in enumerate(body_only):
-            con_body = [[cell for cell in row if cell not in (None, '')] for row in raw_body]
-            cleaned = cleaning_column(clean_header[i], [con_body])
-            clean_body.append(cleaned[0])
 
-        parsed = []
+    header_only = []
-        for i, (cols, rows) in enumerate(zip(clean_header, clean_body), start=1):
+    body_only = []
-            parsed.append({
+    for tbl in tables_data:
-                "title": str(i),
+        head, body = detect_header_rows(tbl)
-                "columns": cols,
+        header_only.append(head)
-                "rows": rows
+        body_only.append(body)
-            })
 
+    clean_header = []
+    for h in header_only:
+        clean_header.append(merge_multiline_header(h))
 
-        return parsed
+    clean_body=[]
+    for i, raw_body in enumerate(body_only):
+        con_body = [[cell for cell in row if cell not in (None, '')] for row in raw_body]
+        cleaned = cleaning_column(clean_header[i], [con_body])
+        clean_body.append(cleaned[0])
+
+    parsed = []
+    for i, (cols, rows) in enumerate(zip(clean_header, clean_body), start=1):
+        parsed.append({
+            "title": str(i),
+            "columns": cols,
+            "rows": rows
+        })
+
+    clean_parsed = filter_geo_admin_column(parsed)
+    print(f"parsed{clean_parsed}")
+    return clean_parsed