#!/usr/bin/env python3
"""Visual comparison tool for Jan 28, 2024 photos - no dependencies."""
from http.server import HTTPServer, BaseHTTPRequestHandler
import os
import subprocess
import glob
import urllib.parse
LIBRARY = "/tank/immich/library/library/admin"
JAN2024 = f"{LIBRARY}/2024/01"
def get_sample_files_from_immich(limit=10):
"""Get IMG files from 2024/01 that are actually in Immich."""
import json
from urllib.request import Request, urlopen
files = []
try:
# Query Immich for all assets, then filter
# Use search with a broad date range for 2024/01
search_data = json.dumps({
"takenAfter": "2024-01-01T00:00:00.000Z",
"takenBefore": "2024-02-01T00:00:00.000Z",
"size": 200
}).encode()
req = Request(f"{API_URL}/search/metadata", data=search_data,
headers={"x-api-key": API_KEY, "Content-Type": "application/json"})
resp = urlopen(req)
data = json.loads(resp.read())
for item in data.get("assets", {}).get("items", []):
path = item.get("originalPath", "")
fname = path.split("/")[-1]
if "/2024/01/" in path and fname.lower().startswith("img_") and fname.lower().endswith((".jpg", ".jpeg")):
files.append({
"filename": fname,
"immich_id": item["id"],
"path": path
})
if len(files) >= limit:
break
except Exception as e:
print(f"Error: {e}")
return files
def find_matches(filename):
"""Find all files with same name in other folders."""
base = filename.rsplit('.', 1)[0].lower()
# Only strip _1024 suffix (Mylio resize indicator), keep +1 and (1) as they indicate different photos
base_clean = base.replace('_1024', '')
matches = []
for root, dirs, files in os.walk(LIBRARY):
if "/2024/01" in root:
continue
for f in files:
f_base = f.rsplit('.', 1)[0].lower()
f_clean = f_base.replace('_1024', '')
if f_clean == base_clean and f.lower().endswith(('.jpg', '.jpeg')):
matches.append(os.path.join(root, f))
return matches
API_KEY = "GsWQUTR6EXlkKp1M82jDJ3KmzhM0fMAbbIbfHDyI"
API_URL = "http://localhost:2283/api"
def get_immich_id(filename):
"""Get Immich asset ID for a file in /2024/01/."""
import json
from urllib.request import Request, urlopen
# Try both cases and without extension
base = filename.rsplit('.', 1)[0]
searches = [filename, filename.upper(), f"{base}.JPG", f"{base}.jpg"]
for search_name in searches:
try:
search_data = json.dumps({"originalFileName": search_name}).encode()
req = Request(f"{API_URL}/search/metadata", data=search_data,
headers={"x-api-key": API_KEY, "Content-Type": "application/json"})
resp = urlopen(req)
data = json.loads(resp.read())
for item in data.get("assets", {}).get("items", []):
if "/2024/01/" in item.get("originalPath", ""):
return item["id"]
except:
pass
return None
def get_file_info(filepath):
stat = os.stat(filepath)
size_kb = stat.st_size / 1024
try:
result = subprocess.run(
['exiftool', '-s', '-s', '-s', '-ImageSize', filepath],
capture_output=True, text=True, timeout=5
)
dims = result.stdout.strip() or "unknown"
except:
dims = "unknown"
try:
result = subprocess.run(
['exiftool', '-s', '-s', '-s', '-DateTimeOriginal', filepath],
capture_output=True, text=True, timeout=5
)
date = result.stdout.strip() or "No EXIF date"
except:
date = "unknown"
rel_path = filepath.replace(LIBRARY + "/", "")
folder = '/'.join(rel_path.split('/')[:-1])
filename = filepath.split('/')[-1]
return {
'path': filepath,
'rel_path': rel_path,
'size_kb': round(size_kb),
'size_mb': round(size_kb/1024, 2),
'dims': dims,
'date': date,
'folder': folder,
'immich_id': get_immich_id(filename)
}
def generate_html():
sample_files = get_sample_files_from_immich(10)
rows = []
for file_data in sample_files:
filename = file_data["filename"]
immich_id = file_data["immich_id"]
filepath = f"{JAN2024}/{filename}"
if not os.path.exists(filepath):
continue
info = get_file_info(filepath)
info['immich_id'] = immich_id # Use the ID we already have
matches = find_matches(filename)
match_infos = [get_file_info(m) for m in sorted(matches)]
match_html = ""
if match_infos:
for m in match_infos:
match_filename = m['path'].split('/')[-1]
match_html += f'''
