MibooGram/metadata/scripts/metadatas.py

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

154 lines
6.0 KiB
Python
Raw Permalink Normal View History

from contextlib import contextmanager
2023-06-06 09:23:15 +00:00
from typing import Iterator, Dict
from aiofiles import open as async_open
from httpx import URL, AsyncClient, RemoteProtocolError, Response
from utils.const import AMBR_HOST, PROJECT_ROOT
from utils.log import logger
try:
import ujson as jsonlib
except ImportError:
import json as jsonlib
2023-10-07 15:15:30 +00:00
__all__ = [
"update_metadata_from_ambr",
"update_metadata_from_github",
"RESOURCE_DEFAULT_PATH",
"RESOURCE_FAST_URL",
"RESOURCE_FightPropRule_URL",
]
2023-02-15 08:33:49 +00:00
RESOURCE_REPO = "PaiGramTeam/PaiGram_Resources"
RESOURCE_BRANCH = "remote"
2022-11-03 10:33:13 +00:00
RESOURCE_ROOT = "Resources"
RESOURCE_DEFAULT_PATH = f"{RESOURCE_REPO}/{RESOURCE_BRANCH}/{RESOURCE_ROOT}/"
2023-06-06 13:05:23 +00:00
RESOURCE_FAST_URL = f"https://genshin-res.paimon.vip/{RESOURCE_ROOT}/"
2023-10-07 15:15:30 +00:00
RESOURCE_FightPropRule_URL = "https://fightproprule.paimon.vip/"
client = AsyncClient()
2023-06-06 09:23:15 +00:00
async def fix_metadata_from_ambr(json_data: Dict[str, Dict], data_type: str):
if data_type == "weapon":
need_append_ids = [11304]
need_attr = ["id", "rank", "type", "name", "icon", "route"]
for wid in need_append_ids:
url = AMBR_HOST.join(f"v2/chs/{data_type}/{wid}")
response = await client.get(url)
json_data_ = jsonlib.loads(response.text)["data"]
2023-06-06 09:23:15 +00:00
json_data[str(json_data_["id"])] = {k: json_data_[k] for k in need_attr}
async def update_metadata_from_ambr(overwrite: bool = True):
result = []
targets = ["material", "weapon", "avatar", "reliquary"]
for target in targets:
path = PROJECT_ROOT.joinpath(f"metadata/data/{target}.json")
if not overwrite and path.exists():
continue
url = AMBR_HOST.join(f"v2/chs/{target}")
path.parent.mkdir(parents=True, exist_ok=True)
response = await client.get(url)
json_data = jsonlib.loads(response.text)["data"]["items"]
2023-06-06 09:23:15 +00:00
await fix_metadata_from_ambr(json_data, target)
async with async_open(path, mode="w", encoding="utf-8") as file:
data = jsonlib.dumps(json_data, ensure_ascii=False, indent=4)
await file.write(data)
result.append(json_data)
return result
@contextmanager
async def stream_request(method, url) -> Iterator[Response]:
2022-12-25 12:01:18 +00:00
async with client.stream(method=method, url=url) as response:
yield response
# noinspection PyShadowingNames
async def update_metadata_from_github(overwrite: bool = True):
path = PROJECT_ROOT.joinpath("metadata/data/namecard.json")
if not overwrite and path.exists():
return
hosts = [
2023-06-06 13:05:23 +00:00
URL(RESOURCE_FAST_URL),
2022-11-03 10:33:13 +00:00
URL(f"https://raw.fastgit.org/{RESOURCE_DEFAULT_PATH}"),
URL(f"https://raw.githubusercontent.com/{RESOURCE_DEFAULT_PATH}"),
]
for num, host in enumerate(hosts):
try:
text_map_url = host.join("TextMap/TextMapCHS.json")
material_url = host.join("ExcelBinOutput/MaterialExcelConfigData.json")
material_json_data = []
async with client.stream("GET", material_url) as response:
started = False
cell = []
async for line in response.aiter_lines():
2023-10-01 12:14:39 +00:00
if line == " {":
started = True
continue
2023-10-01 12:14:39 +00:00
if line in [" },", " }"]:
started = False
if any("MATERIAL_NAMECARD" in x for x in cell):
material_json_data.append(jsonlib.loads("{" + "".join(cell) + "}"))
cell = []
continue
if started:
if "materialType" in line and "MATERIAL_NAMECARD" not in line:
cell = []
started = False
continue
2023-10-01 12:14:39 +00:00
cell.append(line.strip(" "))
string_ids = []
for namecard_data in material_json_data:
string_ids.append(str(namecard_data["nameTextMapHash"]))
string_ids.append(str(namecard_data["descTextMapHash"]))
text_map_json_data = {}
async with client.stream("GET", text_map_url) as response:
async for line in response.aiter_lines():
splits = line.split(":")
string_id = splits[0].strip(' "')
if string_id in string_ids:
2022-11-03 10:33:13 +00:00
text_map_json_data[string_id] = splits[1].strip('\n ,"')
string_ids.remove(string_id)
if not string_ids:
break
data = {}
for namecard_data in material_json_data:
name = text_map_json_data[str(namecard_data["nameTextMapHash"])]
icon = namecard_data["icon"]
navbar = namecard_data["picPath"][0]
banner = namecard_data["picPath"][1]
rank = namecard_data["rankLevel"]
description = text_map_json_data[str(namecard_data["descTextMapHash"])].replace("\\n", "\n")
data.update(
{
str(namecard_data["id"]): {
"id": namecard_data["id"],
"name": name,
"rank": rank,
"icon": icon,
"navbar": navbar,
"profile": banner,
"description": description,
}
}
)
async with async_open(path, mode="w", encoding="utf-8") as file:
data = jsonlib.dumps(data, ensure_ascii=False, indent=4)
await file.write(data)
return data
2022-12-25 13:26:08 +00:00
except RemoteProtocolError as exc:
logger.warning("在从 %s 下载元数据的过程中遇到了错误: %s", host, str(exc))
continue
2022-12-25 13:26:08 +00:00
except Exception as exc:
if num != len(hosts) - 1:
2022-12-25 13:26:08 +00:00
logger.error("在从 %s 下载元数据的过程中遇到了错误: %s", host, str(exc))
continue
2022-12-25 13:26:08 +00:00
raise exc