Streamlit

app.py

@@ -0,0 +1,78 @@
+import asyncio
+import json
+import os
+
+import httpx
+import pandas as pd
+import streamlit as st
+
+from search import fetch_all_user_details, format_register_time, search_users_api
+from takeout import (
+    API_BASE,
+    download_notes_data,
+    gen_markdown,
+    post_process,
+    write_user_bak_meta,
+)
+
+st.title("画吧作品备份工具")
+
+# ── 搜索 ──────────────────────────────────────────────
+st.header("搜索用户")
+
+keyword = st.text_input("用户名关键词", key="keyword")
+if st.button("搜索") and keyword:
+    async def _search():
+        async with httpx.AsyncClient(timeout=60) as client:
+            users = await search_users_api(client, keyword)
+            return await fetch_all_user_details(client, users)
+
+    with st.spinner("搜索中..."):
+        users = asyncio.run(_search())
+
+    exact = sorted([u for u in users if u.is_exact_match], key=lambda u: u.notes_count, reverse=True)
+    other = sorted([u for u in users if not u.is_exact_match], key=lambda u: u.notes_count, reverse=True)
+    st.session_state["search_results"] = exact + other
+
+if "search_results" in st.session_state:
+    users = st.session_state["search_results"]
+    df = pd.DataFrame([{
+        "精确匹配": "⭐" if u.is_exact_match else "",
+        "用户名": u.authorname,
+        "作品数": u.notes_count,
+        "注册时间": format_register_time(u.register_time),
+        "JID": u.jid,
+    } for u in users])
+    st.dataframe(df, width='stretch')
+
+# ── 导出 ──────────────────────────────────────────────
+st.header("导出备份")
+
+jid = st.text_input("JID", value=st.session_state.get("export_jid", ""), key="export_jid_input")
+from_local = st.checkbox("从本地 notes.json 加载（跳过 API 拉取）")
+
+if st.button("开始导出") and jid:
+    async def _export():
+        async with httpx.AsyncClient(timeout=300) as client:
+            if from_local:
+                usr_dir = jid.split("@")[0]
+                with open(f"user_backups/{usr_dir}/notes.json", "r") as f:
+                    notes = json.load(f)
+            else:
+                r = await client.get(API_BASE + "notes", params={"jid": jid})
+                notes = r.json()
+                write_user_bak_meta(jid, notes)
+            await download_notes_data(client, jid, notes)
+            gen_markdown(jid, notes)
+
+    with st.spinner("下载中（进度见终端）..."):
+        asyncio.run(_export())
+
+    with st.spinner("生成压缩包..."):
+        post_process(jid)
+
+    usr_dir = jid.split("@")[0]
+    st.success("导出完成！")
+    st.write(f"备份预览：https://huabar-takeout-preview.saveweb.org/{usr_dir}/notes.html")
+    st.write(f"备份：https://huabar-takeout-preview.saveweb.org/{usr_dir}.zip")
+    st.write("您好。请尽快下载压缩包，链接将于数周后失效。祝好。")

compose.yml

@@ -9,3 +9,23 @@ services:
     environment:
       - MONGODB_URI=${MONGODB_URI}
     restart: always
+
+  takeout_ui:
+    image: ghcr.io/astral-sh/uv:python3.13-alpine
+    container_name: huabar-takeout-ui
+    working_dir: /app
+    ports:
+      - "127.0.0.63:8501:8501"
+    volumes:
+      - .:/app
+      - uv-cache:/root/.cache/uv
+    environment:
+      - API_BASE=http://draws_index:8080/api/
+    command: sh -c "apk add --no-cache pandoc zip && uv run streamlit run app.py --server.address=0.0.0.0 --server.baseUrlPath=/ui"
+    depends_on:
+      - draws_index
+    restart: always
+    mem_limit: 1G
+
+volumes:
+  uv-cache:

pyproject.toml

@@ -7,6 +7,8 @@ requires-python = ">=3.13"
 dependencies = [
     "filetype>=1.2",
     "httpx>=0.28.1",
+    "pandas>=2.3.3",
     "pymongo>=4.15.4",
+    "streamlit>=1.55.0",
     "tqdm>=4.67.1",
 ]

search.py

@@ -1,10 +1,12 @@
 import asyncio
+import os
+
 import httpx
 from datetime import datetime, timezone, timedelta
 from typing import List
 import csv
 
-API_BASE = "http://127.0.0.63:8539/api/"
+API_BASE = os.environ.get("API_BASE", "http://127.0.0.63:8539/api/")
 
 
 class UserInfo:

takeout.py

@@ -14,7 +14,7 @@
     get_urltype
 )
 
-API_BASE = "http://127.0.0.63:8539/api/"
+API_BASE = os.environ.get("API_BASE", "http://127.0.0.63:8539/api/")
 
 def is_keyable(url: str)->bool:
     if not url:
@@ -34,16 +34,16 @@
 
 def find_existing_file(path: str) -> str | None:
     """查找 path 或 path.* 是否已存在，返回找到的路径或 None。"""
     if os.path.exists(path):
         return path
     matches = glob_mod.glob(path + ".*")
     return matches[0] if matches else None
 
 def write_user_bak_meta(jid: str, notes: list[dict]):
     usr_dir = jid.split('@')[0]
     os.makedirs(f'user_backups/{usr_dir}', exist_ok=True)
 
     with open(f'user_backups/{usr_dir}/notes.json', 'w') as f:
         f.write(json.dumps(notes, ensure_ascii=False, indent=2))
 
 async def download_to_bak(sem:asyncio.Semaphore, client:httpx.AsyncClient, url, jid, key):
@@ -51,15 +51,15 @@
     path = f'user_backups/{usr_dir}/notes_data/{key}'
     if find_existing_file(path):
         return
     os.makedirs(f'user_backups/{usr_dir}/notes_data/', exist_ok=True)
     async with sem:
         r = await client.get(url, follow_redirects=True)
         r.raise_for_status()
         with open(path, 'wb') as f:
             f.write(r.content)
         ext = detect_image_ext(path)
         if ext:
             os.rename(path, path + ext)
 
 async def get_zipname(client:httpx.AsyncClient, key:str):
     # qiniu-draw-20240127-072800.3565.zip
@@ -87,8 +87,8 @@
     usr_dir = jid.split('@')[0]
     subprocess.run(['pandoc', f'user_backups/{usr_dir}/notes.md', '--standalone', '--output', f'user_backups/{usr_dir}/notes.html'], check=True)
     # rm ${jid}.zip
     if os.path.exists(f'user_backups/{usr_dir}.zip'):
         os.unlink(f'user_backups/{usr_dir}.zip')
     # zip -r $jid $jid
     subprocess.run(['zip', '-r', f'user_backups/{usr_dir}.zip', f'user_backups/{usr_dir}'], check=True)
 
@@ -121,8 +121,8 @@
 
 def gen_markdown(jid, notes):
     usr_dir = jid.split('@')[0]
     os.makedirs(f'user_backups/{usr_dir}', exist_ok=True)
     with open(f'user_backups/{usr_dir}/notes.md', 'w') as f:
         f.write(f"""\
 # {notes[0]['payload']['authorname']} 的画吧作品备份