import fitz # pymupdf

doc = fitz.open("/tmp/doc_page_1.png") # wait, open the pdf
doc_pdf = fitz.open(stream=open("/tmp/doc_page_1.png", "rb").read(), filetype="png") # image

pdf = fitz.open()
# Let's inspect /tmp/syyly_nhi_vo_v1.pdf or export fresh pdf
import io, os, json
from google.oauth2.credentials import Credentials
from googleapiclient.discovery import build
from googleapiclient.http import MediaIoBaseDownload

token_path = os.path.expanduser("~/.hermes/profiles/reviewer/google_token.json")
with open(token_path) as f: creds = Credentials.from_authorized_user_info(json.load(f))
drive_service = build("drive", "v3", credentials=creds)
doc_id = "1sjTe7Im51CY2Tl0SNQrm8xmO7bYweMkq44E0WXaxs1I"

request = drive_service.files().export_media(fileId=doc_id, mimeType="application/pdf")
pdf_data = request.execute()
pdf = fitz.open(stream=pdf_data, filetype="pdf")

p1 = pdf[0]
print("=== Page 1 Rect ===", p1.rect)
print("=== Drawings on Page 1 ===")
for d in p1.get_drawings():
    print("Drawing rect:", d["rect"], "items:", len(d["items"]))

print("=== Text Blocks on Page 1 ===")
blocks = p1.get_text("blocks")
for b in blocks[:15]:
    print(f"Rect: ({b[0]:.1f}, {b[1]:.1f}, {b[2]:.1f}, {b[3]:.1f}) Text: {repr(b[4].strip())}")