Jayce-Ping's picture
Add files using upload-large-folder tool
0ce0b49 verified
raw
history blame
15.7 kB
import os
import json
from datetime import datetime
from utils import divide_prompt
import argparse
from flask import Flask, render_template, request, send_from_directory, jsonify
app = Flask(__name__)
FILE_PATH = os.path.dirname(os.path.abspath(__file__))
# 设置图像目录路径
IMAGE_DIR = os.path.join(FILE_PATH, "../images_divided")
ITEMS_PER_PAGE = 1 # 每页显示的 data 项目数
# PROMPT_DATA_FILE = os.path.join(FILE_PATH, 'data/extended_data_summary_v2.jsonl')
PROMPT_DATA_FILE = os.path.join(FILE_PATH, 'data/2by2_data_summary.jsonl')
with open(PROMPT_DATA_FILE, 'r', encoding='utf-8') as f:
PROMPT_DATA = [json.loads(line) for line in f if line.strip()]
PROMPT_DATA = {item['idx']: item for item in PROMPT_DATA}
def load_metadata():
"""加载用户的元数据"""
if not os.path.exists(METADATA_FILE):
return {"last_page": None, "last_updated": None}
try:
with open(METADATA_FILE, 'r', encoding='utf-8') as f:
return json.load(f)
except:
return {"last_page": None, "last_updated": None}
def save_metadata(page):
"""保存当前页码到元数据文件"""
metadata = {
"last_page": page,
"last_updated": datetime.now().isoformat()
}
os.makedirs(os.path.dirname(METADATA_FILE), exist_ok=True)
with open(METADATA_FILE, 'w', encoding='utf-8') as f:
json.dump(metadata, f, ensure_ascii=False, indent=2)
@app.route('/')
def index():
# 读取所有文件夹
# folders = sorted([
# f for f in os.listdir(IMAGE_DIR)
# if os.path.isdir(os.path.join(IMAGE_DIR, f)) and not f.startswith('.')
# ])
folders = sorted([
idx for idx in PROMPT_DATA.keys()
if os.path.isdir(os.path.join(IMAGE_DIR, idx))
])
data = []
# 加载已有的标注记录
existing_records = load_existing_records()
existing_dict = {}
for record in existing_records:
folder = record.get('folder')
ref_image = record.get('ref_image')
compare_images = tuple(sorted(record.get('compare_images', [])))
key = (folder, ref_image, compare_images)
existing_dict[key] = record
for cnt, idx in enumerate(folders):
folder_path = os.path.join(IMAGE_DIR, idx)
images = sorted([img for img in os.listdir(folder_path) if img.endswith('.png')])
first_indices, second_indices = zip(*[
list(map(int, img.split('.')[0].split('_')))
for img in images
])
first_indices = sorted(set(first_indices))
second_indices = sorted(set(second_indices))
for i in first_indices:
for j in second_indices:
ref_img = f"{i}_{j}.png"
candidates = [
[
f"{x}_{y}.png"
for x in first_indices
]
for y in second_indices if y != j
]
items = []
for y, cand in zip([y for y in second_indices if y != j], candidates):
item = {
'ref_image': f"/images/{idx}/{ref_img}",
'compare_images': [f"/images/{idx}/{cand_img}" for cand_img in cand],
'folder': idx,
'ref_index': (i, j),
'candidate_column': y,
'prompt': divide_prompt(PROMPT_DATA[idx]['prompt'])[0] if idx in PROMPT_DATA and 'prompt' in PROMPT_DATA[idx] else '',
'existing_compare_images': [],
'existing_rank_images': [],
'is_annotated': False,
'stared': False,
'discarded': False,
'grid_issue': False
}
current_compare_images = [img.split('/')[-1] for img in item['compare_images']]
ref_filename = item['ref_image'].split('/')[-1]
lookup_key = (idx, ref_filename, tuple(sorted(current_compare_images)))
existing_data = existing_dict.get(lookup_key, {})
item['existing_compare_images'] = existing_data.get('compare_images', [])
item['existing_rank_images'] = existing_data.get('rank_images', [])
item['is_annotated'] = len(item['existing_rank_images']) > 0
item['stared'] = existing_data.get('stared', False)
item['discarded'] = existing_data.get('discarded', False)
item['grid_issue'] = existing_data.get('grid_issue', False)
items.append(item)
data.extend(items)
# 获取页码参数,优先级:URL参数 > 元数据文件 > start_idx配置 > 默认1
url_page = request.args.get('page')
if url_page:
# 如果URL明确指定了页码,使用URL的页码
page = int(url_page)
else:
metadata = load_metadata()
last_page = metadata.get('last_page', None)
if last_page:
# 如果元数据文件中有记录,使用元数据中的页码
page = last_page
elif not app.config.get('START_IDX_USED', False):
# 如果start_idx还没被使用过,使用start_idx
page = app.config.get('START_IDX', 1)
app.config['START_IDX_USED'] = True
else:
# 否则使用默认页码1
page = 1
# 保存当前页码到元数据
save_metadata(page)
start = (page - 1) * ITEMS_PER_PAGE
end = start + ITEMS_PER_PAGE
paginated_data = data[start:end]
return render_template(
'index.html',
data=paginated_data,
page=page,
total_pages=(len(data) + ITEMS_PER_PAGE - 1) // ITEMS_PER_PAGE
)
@app.route('/update_page', methods=['POST'])
def update_page():
"""更新当前页码到元数据文件"""
try:
data = request.get_json()
page = data.get('page')
if page:
save_metadata(int(page))
return jsonify({'status': 'success'})
return jsonify({'status': 'error', 'message': '无效的页码'}), 400
except Exception as e:
return jsonify({'status': 'error', 'message': str(e)}), 500
@app.route('/sort', methods=['POST'])
def sort_images():
try:
sorted_images = request.form.getlist('sorted_images')
if not sorted_images:
return jsonify({'status': 'error', 'message': '没有排序数据'}), 400
ranking = []
for item in sorted_images:
rank, image_path = item.split(':', 1)
ranking.append({'rank': int(rank), 'image_path': image_path})
ranking.sort(key=lambda x: x['rank'])
if ranking:
first_image = ranking[0]['image_path']
parts = first_image.split('/')
folder = parts[2]
ref_image = request.form.get('ref_image', '')
ref_filename = ref_image.split('/')[-1]
all_compare_images = request.form.get('all_compare_images', '')
if all_compare_images:
all_compare_images = json.loads(all_compare_images)
else:
all_compare_images = []
ranked_filenames = [item['image_path'].split('/')[-1] for item in ranking]
record = {
'folder': folder,
'ref_image': ref_filename,
'compare_images': all_compare_images,
'rank_images': ranked_filenames,
'timestamp': datetime.now().isoformat()
}
existing_records = load_existing_records()
updated = False
for i, existing in enumerate(existing_records):
if (existing.get('folder') == folder and
existing.get('ref_image') == ref_filename and
tuple(sorted(existing.get('compare_images', []))) == tuple(sorted(all_compare_images))):
record['stared'] = existing.get('stared', False)
record['discarded'] = existing.get('discarded', False)
record['grid_issue'] = existing.get('grid_issue', False)
existing_records[i] = record
updated = True
break
if not updated:
record['stared'] = False
record['discarded'] = False
record['grid_issue'] = False
existing_records.append(record)
save_records(existing_records)
return jsonify({'status': 'success','message': f"排序已{'更新' if updated else '保存'}!",'record': record})
except Exception as e:
return jsonify({'status': 'error', 'message': str(e)}), 500
@app.route('/star', methods=['POST'])
def star_image():
try:
data = request.get_json()
folder = data.get("folder")
ref_image = data.get("ref_image")
stared = data.get("stared", False)
compare_images = data.get("compare_images", [])
rank_images = data.get("rank_images", [])
existing_records = load_existing_records()
updated = False
for record in existing_records:
if (record.get("folder") == folder and
record.get("ref_image") == ref_image and
tuple(sorted(record.get("compare_images", []))) == tuple(sorted(compare_images))):
record["stared"] = stared
if rank_images:
record["rank_images"] = rank_images
record["timestamp"] = datetime.now().isoformat()
updated = True
break
if not updated:
new_record = {
"folder": folder,
"ref_image": ref_image,
"compare_images": compare_images,
"rank_images": rank_images,
"stared": stared,
"timestamp": datetime.now().isoformat()
}
existing_records.append(new_record)
save_records(existing_records)
return jsonify({"status": "success", "message": f"{'已收藏' if stared else '已取消收藏'}"})
except Exception as e:
print(f"Error in star_image: {str(e)}")
return jsonify({"status": "error", "message": str(e)}), 500
@app.route('/discard', methods=['POST'])
def discard_image():
try:
data = request.get_json()
folder = data.get("folder")
ref_image = data.get("ref_image")
discarded = data.get("discarded", False)
compare_images = data.get("compare_images", [])
rank_images = data.get("rank_images", [])
existing_records = load_existing_records()
updated = False
for record in existing_records:
if (record.get("folder") == folder and
record.get("ref_image") == ref_image and
tuple(sorted(record.get("compare_images", []))) == tuple(sorted(compare_images))):
record["discarded"] = discarded
if rank_images:
record["rank_images"] = rank_images
record["timestamp"] = datetime.now().isoformat()
updated = True
break
if not updated:
new_record = {
"folder": folder,
"ref_image": ref_image,
"compare_images": compare_images,
"rank_images": rank_images,
"discarded": discarded,
"timestamp": datetime.now().isoformat()
}
existing_records.append(new_record)
save_records(existing_records)
return jsonify({"status": "success", "message": f"{'已丢弃' if discarded else '已取消丢弃'}"})
except Exception as e:
print(f"Error in discard_image: {str(e)}")
return jsonify({"status": "error", "message": str(e)}), 500
def load_existing_records():
if not os.path.exists(DATA_FILE):
return []
records = []
with open(DATA_FILE, 'r', encoding='utf-8') as f:
for line in f:
try:
records.append(json.loads(line))
except json.JSONDecodeError:
continue
return records
def save_records(records):
with open(DATA_FILE, 'w', encoding='utf-8') as f:
for record in records:
f.write(json.dumps(record, ensure_ascii=False) + '\n')
@app.route('/grid_issue', methods=['POST'])
def grid_issue_image():
try:
data = request.get_json()
folder = data.get("folder")
ref_image = data.get("ref_image")
grid_issue = data.get("grid_issue", False)
compare_images = data.get("compare_images", [])
rank_images = data.get("rank_images", [])
existing_records = load_existing_records()
updated = False
for record in existing_records:
if (record.get("folder") == folder and
record.get("ref_image") == ref_image and
tuple(sorted(record.get("compare_images", []))) == tuple(sorted(compare_images))):
record["grid_issue"] = grid_issue
if rank_images:
record["rank_images"] = rank_images
record["timestamp"] = datetime.now().isoformat()
updated = True
break
if not updated:
new_record = {
"folder": folder,
"ref_image": ref_image,
"compare_images": compare_images,
"rank_images": rank_images,
"grid_issue": grid_issue,
"timestamp": datetime.now().isoformat()
}
existing_records.append(new_record)
save_records(existing_records)
return jsonify({"status": "success", "message": f"{'已标记网格问题' if grid_issue else '已取消网格问题标记'}"})
except Exception as e:
print(f"Error in grid_issue_image: {str(e)}")
return jsonify({"status": "error", "message": str(e)}), 500
@app.route('/images/<path:filename>')
def images(filename):
return send_from_directory(IMAGE_DIR, filename)
@app.route('/view_data')
def view_data():
existing_records = load_existing_records()
return jsonify(existing_records)
def parse_args():
parser = argparse.ArgumentParser(description="Run the annotation server.")
parser.add_argument('--port', type=str, default='5000', help='Host address to run the server on')
parser.add_argument('--user', type=str, default='user', help='User name for the session')
parser.add_argument('--start_idx', type=int, default=1, help='Starting page index for initialization')
return parser.parse_args()
if __name__ == '__main__':
args = parse_args()
user = args.user
port = int(args.port)
start_idx = args.start_idx
global DATA_FILE, METADATA_FILE
DATA_FILE = os.path.join(FILE_PATH, "data", f"annotation_{user}.jsonl")
METADATA_FILE = os.path.join(FILE_PATH, "data", f"metadata_{user}.json")
# 将start_idx存储到app配置中
app.config['START_IDX'] = start_idx
print(f"启动服务器:")
print(f" 用户: {user}")
print(f" 端口: {port}")
print(f" 起始页码: {start_idx}")
print(f" 数据文件: {DATA_FILE}")
print(f" 元数据文件: {METADATA_FILE}")
app.run(debug=True, port=port)