~/

~/attachments/5edeb78c-f33d-424c-8ac3-b66c50be9e3b

PDF upload rejected malicious content how to sanitize PDF remove JavaScript embedded files official guidance

python -c "from pypdf import PdfReader; p='/tmp/workspace/attachments/5edeb78c-f33d-424c-8ac3-b66c50be9e3b/appproval.pdf'; r=PdfReader(p, strict=False); print('pages',len(r.pages)); print('encrypted',r.is_encrypted); print('metadata',r.metadata); print('root keys',list(r.trailer['/Root'].keys())); print('size',__import__('os').path.getsize(p)); [print('page',i+1,'keys',list(pg.keys())) for i,pg in enumerate(r.pages)]"

python -c "import fitz; print(fitz.__doc__.split()[1])"

mkdir -p "/tmp/workspace/pdf-cleanup/output" 66 python - <<'PY' from pypdf import PdfReader p='/tmp/workspace/attachments/5edeb78c-f33d-424c-8ac3-b66c50be9e3b/appproval.pdf' r=PdfReader(p, strict=False) for i,page in enumerate(r.pages,1): a=page.get('/Annots') print('page',i,'annotations',len(a) if a else 0) if a: for x in a: o=x.get_object(); print(' ', o.get('/Subtype'), 'action',o.get('/A'), 'AA',o.get('/AA')) raw=open(p,'rb').read() for term in (b'/JavaScript',b'/JS',b'/EmbeddedFiles',b'/OpenAction',b'/Launch',b'/RichMedia',b'/XFA'): print(term.decode(),raw.count(term)) PY

python - <<'PY' from pypdf import PdfReader, PdfWriter from pypdf.generic import NameObject src='/tmp/workspace/attachments/5edeb78c-f33d-424c-8ac3-b66c50be9e3b/appproval.pdf' out='/tmp/workspace/pdf-cleanup/output/appproval_clean.pdf' r=PdfReader(src, strict=False) w=PdfWriter() for page in r.pages: # annotations might trigger automated checks or refer to external objects; omit them if '/Annots' in page: del page[NameObject('/Annots')] w.add_page(page) # Do not copy document catalog extensions, Xodo marks, scripts, attachments, or original metadata w.add_metadata({'/Title':'Approval Document','/Producer':'PDF document clean copy'}) with open(out,'wb') as f: w.write(f) # Validate output rr=PdfReader(out, strict=False) print('created:',out) print('pages:',len(rr.pages),'encrypted:',rr.is_encrypted,'bytes:',__import__('os').path.getsize(out)) print('catalog:',list(rr.trailer['/Root'].keys())) print('metadata:',rr.metadata) print('page annotation flags:',[('/Annots' in pg) for pg in rr.pages]) raw=open(out,'rb').read() print('active-content markers:',{x.decode('ascii'):raw.count(x) for x in [b'/JavaScript',b'/JS',b'/EmbeddedFiles',b'/OpenAction',b'/Launch',b'/RichMedia',b'/XFA',b'/xodo_watermark']}) PY

site:natboard.edu.in "Malicious content detected in file"

python - <<'PY' from pypdf import PdfReader p='/tmp/workspace/pdf-cleanup/output/appproval_clean.pdf'; r=PdfReader(p) for n,pg in enumerate(r.pages,1): rs=pg.get('/Resources',{}) xo=rs.get('/XObject',{}) print('PAGE',n,'xobjects',len(xo),'content bytes',len(pg.get_contents().get_data())) for name,ref in xo.items(): o=ref.get_object(); print(name,o.get('/Subtype'),o.get('/Width'),o.get('/Height'),o.get('/Filter'),len(o.get_data()) if o.get('/Subtype')=='/Image' else '') PY

~/pdf-cleanup

python - <<'PY' from pypdf import PdfReader from PIL import Image from io import BytesIO from pathlib import Path src=Path('/tmp/workspace/attachments/5edeb78c-f33d-424c-8ac3-b66c50be9e3b/appproval.pdf') outdir=Path('/tmp/workspace/pdf-cleanup/output/raster_pages') outdir.mkdir(parents=True, exist_ok=True) r=PdfReader(str(src), strict=False) images=[] for n,pg in enumerate(r.pages,1): xobj=pg['/Resources']['/XObject'] # Main full-page scan is highest-pixel image candidates=[] for _,ref in xobj.items(): ob=ref.get_object() if ob.get('/Subtype') == '/Image': candidates.append((int(ob.get('/Width',0))*int(ob.get('/Height',0)), ob)) ob=max(candidates,key=lambda z:z[0])[1] im=Image.open(BytesIO(ob.get_data())).convert('RGB') # preserve practical legibility while reduce file size im.thumbnail((1650, 2350), Image.Resampling.LANCZOS) jpg=outdir/f'page-{n}.jpg' im.save(jpg,'JPEG',quality=88,optimize=True,progressive=False) images.append(Image.open(jpg).convert('RGB')) # Standard image-only PDF. No links/forms/scripts/annotations/metadata. out=Path('/tmp/workspace/pdf-cleanup/output/approval_upload.pdf') images[0].save(out, 'PDF', save_all=True, append_images=images[1:], resolution=200.0, title='', author='', subject='', keywords='') print(out, out.stat().st_size) # basic static inspection rr=PdfReader(str(out),strict=False) print('pages',len(rr.pages),'root',list(rr.trailer['/Root'].keys()),'metadata',rr.metadata) raw=out.read_bytes() print({m.decode('ascii'):raw.count(m) for m in [b'/JavaScript',b'/JS',b'/EmbeddedFiles',b'/OpenAction',b'/Launch',b'/RichMedia',b'/XFA',b'/Annots',b'/xodo_watermark']}) PY

python - <<'PY' from reportlab.pdfgen import canvas from reportlab.lib.utils import ImageReader from reportlab.lib.pagesizes import A4 from pathlib import Path from PIL import Image outdir=Path('/tmp/workspace/pdf-cleanup/output') imgs=sorted((outdir/'raster_pages').glob('page-*.jpg'), key=lambda p:int(p.stem.split('-')[1])) out=outdir/'approval_rebuilt.pdf' c=canvas.Canvas(str(out), pagesize=A4, pageCompression=1, invariant=1) c.setTitle('Approval Document') c.setAuthor('') c.setSubject('') W,H=A4 for img_path in imgs: im=Image.open(img_path); iw,ih=im.size scale=min(W/iw,H/ih) dw,dh=iw*scale,ih*scale c.drawImage(ImageReader(str(img_path)),(W-dw)/2,(H-dh)/2,width=dw,height=dh,mask=None) c.showPage() c.save() print(out, out.stat().st_size) PY python - <<'PY' from pypdf import PdfReader p='/tmp/workspace/pdf-cleanup/output/approval_rebuilt.pdf' r=PdfReader(p, strict=False) print('pages',len(r.pages), 'root',list(r.trailer['/Root'].keys()), 'metadata',r.metadata) print('annots',[p.get('/Annots') for p in r.pages]) raw=open(p,'rb').read() print({key:raw.count(key.encode()) for key in ['/JavaScript','/OpenAction','/Launch','/EmbeddedFiles','/RichMedia','/XFA','/Annots','/xodo_watermark']}) PY

~/

NBEMS exam portal contact technical support document upload malicious content detected file

~/attachments/5edeb78c-f33d-424c-8ac3-b66c50be9e3b

python - <<'PY' from pypdf import PdfReader from pathlib import Path p=Path('/tmp/workspace/attachments/5edeb78c-f33d-424c-8ac3-b66c50be9e3b/Approval (1).pdf') r=PdfReader(str(p),strict=False) print('size',p.stat().st_size,'pages',len(r.pages),'encrypted',r.is_encrypted) print('root',list(r.trailer['/Root'].keys())) print('metadata',r.metadata) raw=p.read_bytes() print({x.decode('ascii'):raw.count(x) for x in [b'/JavaScript',b'/OpenAction',b'/Launch',b'/EmbeddedFiles',b'/RichMedia',b'/XFA',b'/Annots',b'/xodo_watermark']}) for n,page in enumerate(r.pages,1): print(n,'annots',bool(page.get('/Annots')),'MB',page.mediabox) PY