cd /home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/ && python3 -c " import pdfplumber with pdfplumber.open('202511221303622410.pdf') as pdf: print(f'Total pages: {len(pdf.pages)}') for i, page in enumerate(pdf.pages[:3]): print(f'--- Page {i+1} ---') text = page.extract_text() print(text[:3000] if text else 'No text') print() "

python3 -c " import pdfplumber results = [] with pdfplumber.open('/home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/202511221303622410.pdf') as pdf: total = len(pdf.pages) print(f'Total pages: {total}') for i, page in enumerate(pdf.pages[1:], start=2): # skip legend page text = page.extract_text() if not text: continue lines = text.split('\n') for line in lines: # Look for GN PwD in the line if 'GN PwD' in line: results.append(f'Page {i}: {line}') print(f'\nTotal GN PwD entries found: {len(results)}') for r in results: print(r) " 2>&1 | head -300

python3 -c " import pdfplumber with pdfplumber.open('/home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/202511221303622410.pdf') as pdf: # Check page 2 text carefully page = pdf.pages[1] text = page.extract_text() print(repr(text[:2000])) " 2>&1

python3 -c " import pdfplumber # Search across all pages for PwD-related text with pdfplumber.open('/home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/202511221303622410.pdf') as pdf: for i, page in enumerate(pdf.pages): text = page.extract_text() if text and ('PwD' in text or 'Pwd' in text or 'PWD' in text or 'Disability' in text or 'disability' in text): print(f'Page {i+1}: contains PwD/disability text') # Print the relevant lines for line in text.split('\n'): if 'PwD' in line or 'Pwd' in line or 'PWD' in line or 'Disability' in line: print(f' >>> {line}') " 2>&1 | head -200

python3 -c " import pdfplumber # Look at page 1 carefully for the legend and any PwD content with pdfplumber.open('/home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/202511221303622410.pdf') as pdf: page = pdf.pages[0] text = page.extract_text() print('PAGE 1 FULL TEXT:') print(text) print() # Check a few sample pages for unique category values print('--- Sampling page 10 ---') text10 = pdf.pages[9].extract_text() print(text10[:2000]) " 2>&1

python3 -c " import pdfplumber # Check tables on a sample page with pdfplumber.open('/home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/202511221303622410.pdf') as pdf: page = pdf.pages[1] tables = page.extract_tables() print(f'Tables found on page 2: {len(tables)}') if tables: for t in tables: for row in t: print(row) else: # Try words to understand layout words = page.extract_words() print('First 50 words with positions:') for w in words[:50]: print(w) " 2>&1

python3 << 'EOF' import pdfplumber results = [] with pdfplumber.open('/home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/202511221303622410.pdf') as pdf: total = len(pdf.pages) for i, page in enumerate(pdf.pages[1:], start=2): # skip legend page try: tables = page.extract_tables() if not tables: continue for table in tables: for row in table: if row and len(row) >= 6: allotted_cat = row[5] if row[5] else '' candidate_cat = row[6] if len(row) > 6 and row[6] else '' # GN PwD = UR PwD if 'GN PwD' in str(allotted_cat) or 'GN PwD' in str(candidate_cat): results.append({ 'sno': row[0], 'rank': row[1], 'quota': row[2], 'institute': row[3], 'course': row[4], 'allotted_cat': allotted_cat, 'candidate_cat': candidate_cat, 'remarks': row[7] if len(row) > 7 else '' }) except Exception as e: print(f"Error on page {i}: {e}") print(f"Total GN PwD entries: {len(results)}") for r in results: print(f"Rank: {r['rank']} | Institute: {r['institute'][:80]} | Course: {r['course'][:40]} | AllottedCat: {r['allotted_cat']} | CandidateCat: {r['candidate_cat']}") EOF

python3 << 'EOF' import pdfplumber results = [] with pdfplumber.open('/home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/202511221303622410.pdf') as pdf: total = len(pdf.pages) for i, page in enumerate(pdf.pages[1:], start=2): # skip legend page try: text = page.extract_text() if not text: continue lines = text.split('\n') for line in lines: # The allotted category column text for UR PWD would be "GN PwD" if 'GN PwD' in line: results.append(f"Page {i}: {line}") except Exception as e: pass print(f"Total GN PwD entries: {len(results)}") for r in results[:50]: print(r) EOF

python3 << 'EOF' import pdfplumber results = [] PDF = '/home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/202511221303622410.pdf' # Process in batches of 100 pages BATCH = 100 TOTAL = 1301 for batch_start in range(1, TOTAL, BATCH): batch_end = min(batch_start + BATCH, TOTAL) with pdfplumber.open(PDF) as pdf: for i in range(batch_start, batch_end): try: text = pdf.pages[i].extract_text() if text and 'GN PwD' in text: for line in text.split('\n'): if 'GN PwD' in line: results.append(f"P{i+1}: {line}") except: pass print(f"Total GN PwD lines: {len(results)}") for r in results: print(r) EOF

which pdftotext && pdftotext --version 2>&1 || echo "not found"

pdftotext /home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/202511221303622410.pdf /home/daytona/workspace/allotment_text.txt && echo "Done" && wc -l /home/daytona/workspace/allotment_text.txt

grep -n "GN PwD" /home/daytona/workspace/allotment_text.txt | head -200

grep -n "PwD\|PWD\|pwd\|Person with" /home/daytona/workspace/allotment_text.txt | head -50

python3 << 'EOF' # Read the extracted text file and find GN PwD (UR PWD) entries # The pattern: "PwD" alone on a line = GN PwD (since OBC/EW/SC/ST PwD are prefixed) with open('/home/daytona/workspace/allotment_text.txt', 'r') as f: lines = f.readlines() results = [] i = 0 while i < len(lines): line = lines[i].strip() # bare "PwD" = GN PwD if line == 'PwD': # Look back up to 20 lines for context context_start = max(0, i - 20) context = [(j+1, lines[j].strip()) for j in range(context_start, i+5)] results.append({'line': i+1, 'context': context}) i += 1 print(f"Total bare PwD (GN PwD / UR PWD) occurrences: {len(results)}") print() # Display each entry with context for entry in results: print(f"=== Line {entry['line']} ===") for ln, txt in entry['context']: marker = " <<< GN PwD" if ln == entry['line'] else "" print(f" {ln}: {txt}{marker}") print() EOF

python3 << 'EOF' # Parse the text file to extract GN PwD (UR PWD) entries with rank and institute # Strategy: find lines with numbers near the PwD entries, look for institute names nearby with open('/home/daytona/workspace/allotment_text.txt', 'r') as f: lines = f.readlines() # Strip lines lines = [l.rstrip('\n') for l in lines] import re def is_rank(s): return bool(re.match(r'^\d+$', s.strip())) and len(s.strip()) <= 6 def looks_like_rank(s): s = s.strip() return bool(re.match(r'^\d{3,6}$', s)) # Find all "Open PwD" / "General PwD" pairs # Pattern: line N = "Open", line N+1 = "PwD", line N+2 = "Open" or "General" # Allotted Category = "Open PwD" = GN PwD # Let's find the pattern: a line "Open" followed immediately (with possible blank) by "PwD" # where the PwD line is a GN PwD entry (not OBC/EW/SC/ST PwD) results = [] i = 0 while i < len(lines): line = lines[i].strip() # Pattern for GN PwD allotted: "Open" then "PwD" (bare) - allotted category = GN PwD if line == 'Open' and i + 1 < len(lines) and lines[i+1].strip() == 'PwD': # This is an "Open PwD" allotted category entry # Look back for rank numbers (two numbers close together: SNo and Rank) # and institute name context = lines[max(0,i-40):i+10] # Find numbers that look like ranks in the context numbers = [] for j, ctx_line in enumerate(context): ctx_stripped = ctx_line.strip() if looks_like_rank(ctx_stripped): numbers.append((max(0,i-40)+j, ctx_stripped)) # Look for course names courses = [] for j, ctx_line in enumerate(context): ctx_stripped = ctx_line.strip() if ctx_stripped.startswith('M.D.') or ctx_stripped.startswith('M.S.') or ctx_stripped.startswith('MS ') or ctx_stripped.startswith('MD/') or ctx_stripped.startswith('(NBEMS'): courses.append(ctx_stripped) # Look for institute names (long lines that aren't categories) institutes = [] for j, ctx_line in enumerate(context): ctx_stripped = ctx_line.strip() if len(ctx_stripped) > 30 and not ctx_stripped.startswith('M.D.') and not ctx_stripped.startswith('M.S.') and not any(c in ctx_stripped for c in ['Allotted', 'Open', 'OBC', 'General', 'EWS', 'Quota', 'Merit']) and re.search(r'[A-Z]{3,}', ctx_stripped): institutes.append(ctx_stripped) results.append({ 'line': i+1, 'numbers': numbers[-4:] if numbers else [], 'courses': courses[-2:] if courses else [], 'institutes': institutes[-2:] if institutes else [] }) i += 1 # Deduplicate - sometimes same entry appears twice (Open PwD + General PwD) # Filter to unique entries by looking at consecutive line numbers final_results = [] prev_line = -100 for r in results: if r['line'] - prev_line > 10: # avoid duplicates within 10 lines final_results.append(r) prev_line = r['line'] print(f"Total unique GN PwD (UR PWD) allotments: {len(final_results)}") print() for idx, r in enumerate(final_results, 1): nums = [n[1] for n in r['numbers']] courses_str = ' / '.join(r['courses'][:1]) inst_str = r['institutes'][0][:70] if r['institutes'] else 'N/A' print(f"{idx:3}. Line {r['line']:7} | Ranks: {nums} | Course: {courses_str[:50]} | Institute: {inst_str}") EOF

python3 << 'EOF' # Use pdfplumber on specific pages containing GN PwD entries # Page numbers based on line numbers in text (roughly 350 lines per page) # Line 101321 ~ page 290, line 286740 ~ page 820, etc. # But text has ~450899 lines / 1301 pages = ~346 lines/page import pdfplumber PDF = '/home/daytona/workspace/attachments/6d8109ec-98b4-4c59-b803-ff1308ed430b/202511221303622410.pdf' # Approximate page numbers for GN PwD entries based on line numbers # 450899 lines / 1301 pages ≈ 346 lines per page # text line -> page: line/346 line_to_page = { 101321: int(101321/346)+1, 286740: int(286740/346)+1, 314536: int(314536/346)+1, 345619: int(345619/346)+1, 350421: int(350421/346)+1, 365520: int(365520/346)+1, 371361: int(371361/346)+1, 394517: int(394517/346)+1, 396659: int(396659/346)+1, } for tl, pg in line_to_page.items(): print(f"Text line {tl} -> estimated page {pg}") # Try reading specific pages target_pages = sorted(set(line_to_page.values())) print(f"\nTarget pages: {target_pages}") results = [] with pdfplumber.open(PDF) as pdf: for pg_num in target_pages: if pg_num > len(pdf.pages): continue page = pdf.pages[pg_num - 1] tables = page.extract_tables() for table in tables: for row in table: if row and len(row) >= 6: allotted_cat = str(row[5]) if row[5] else '' if 'GN PwD' in allotted_cat or 'PwD' in allotted_cat: results.append({ 'page': pg_num, 'rank': row[1], 'quota': row[2], 'institute': (row[3] or '')[:80], 'course': (row[4] or '')[:50], 'allotted_cat': allotted_cat, 'cand_cat': str(row[6]) if len(row) > 6 and row[6] else '' }) print(f"\nGN PwD rows found via table extraction: {len(results)}") for r in results: print(f"Page {r['page']} | Rank: {r['rank']} | Cat: {r['allotted_cat']} | Course: {r['course'][:40]} | Institute: {r['institute'][:60]}") EOF

python3 << 'EOF' # Parse allotment_text.txt more carefully to extract GN PwD entries # The PDF column layout: SNo | Rank | Quota | Institute | Course | AllottedCat | CandidateCat | Remarks # In text, each row appears as alternating columns. Numbers that appear in pairs near # the PwD entries are (SNo, NEET_PG_rank) # Key insight: In pdftotext output for this table, the columns interleave like: # SNo1, SNo2, Rank1, Rank2, Quota1, Institute1, Course1, AllottedCat1, CandCat1, ... # This is complex. Let's use a different approach: # Find consecutive pairs of numbers where 2nd is larger = SNo + NEET Rank with open('/home/daytona/workspace/allotment_text.txt', 'r') as f: content = f.read() # Split into pages using the "Page No." marker pages_raw = content.split('Page No.') print(f"Total page splits: {len(pages_raw)}") import re # For each page, find GN PwD entries and extract data all_entries = [] for pg_idx, page_text in enumerate(pages_raw[1:], start=1): # skip first (legend) if 'PwD' not in page_text: continue lines = page_text.split('\n') # Find "Open" followed by "PwD" (GN PwD) for i, line in enumerate(lines): if line.strip() == 'Open' and i+1 < len(lines) and lines[i+1].strip() == 'PwD': # Extract surrounding context start = max(0, i-60) ctx = lines[start:i+15] # Find all numbers in context nums = [] for j, cl in enumerate(ctx): cl_s = cl.strip() if re.match(r'^\d{4,6}$', cl_s): nums.append(cl_s) # Find course (M.D./M.S./NBEMS lines) courses = [cl.strip() for cl in ctx if cl.strip().startswith(('M.D.', 'M.S.', 'MS (', 'MD/', '(NBEMS'))] # Find long institute-like lines institutes = [] for cl in ctx: cl_s = cl.strip() if (len(cl_s) > 25 and re.search(r'[A-Z]{4}', cl_s) and not any(kw in cl_s for kw in ['Page No', 'NEET', 'Allotment', 'Round', 'Category', 'Abbrev', 'Allotted', 'All India', 'DNB Quota', 'Non-Resident', 'Self-Financed', 'Delhi University', 'IP University', 'Minority'])): institutes.append(cl_s) # The NEET PG rank is typically the larger 5-digit number neet_ranks = [n for n in nums if len(n) >= 5] entry = { 'page': pg_idx, 'neet_ranks': neet_ranks[-4:] if neet_ranks else [], 'course': courses[-1] if courses else '', 'institute': institutes[-1][:80] if institutes else '' } all_entries.append(entry) # Deduplicate by page + approximate position # Some entries show up twice from adjacent pages print(f"\nRaw GN PwD entries: {len(all_entries)}") # Filter: remove entries where neet_ranks are empty or look like pincodes def is_plausible_rank(r): # NEET PG ranks go up to ~1,90,000 return 1000 <= int(r) <= 200000 filtered = [] for e in all_entries: good_ranks = [r for r in e['neet_ranks'] if is_plausible_rank(r)] e['good_ranks'] = good_ranks filtered.append(e) print(f"\nTotal GN PwD allotment entries found: {len(filtered)}") print("\n{:4} {:8} {:8} {:<50} {}".format("No.", "Page", "Rank(s)", "Course", "Institute")) print("-"*130) for i, e in enumerate(filtered, 1): ranks_str = ', '.join(e['good_ranks']) if e['good_ranks'] else 'unclear' print(f"{i:3}. Pg {e['page']:4} | {ranks_str:20} | {e['course'][:45]:45} | {e['institute'][:60]}") EOF