You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
70 lines
2.1 KiB
70 lines
2.1 KiB
"""东方财富 CHOICE API 十大股东明细抓取工具。
|
|
- 输入: 股票代码列表 (如 ['603233.SH'])
|
|
- 输出: pandas DataFrame (按报告期末的十大股东明细)
|
|
- 不需要登录, 不需要 cookie, 无文件落盘
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
import pandas as pd
|
|
|
|
from utils._emchoice import fetch_report
|
|
|
|
|
|
_REPORT_NAME = "RPT_DMSK_HOLDERS"
|
|
# 列清单沿用 docs/api_docs.md 抓包值
|
|
_COLUMNS = (
|
|
"SECUCODE,SECURITY_CODE,ORG_CODE,SECURITY_TYPE_CODE,END_DATE,RANK,"
|
|
"HOLDER_CODE,HOLDER_NAME,HOLD_NUM,HOLD_RATIO,HOLD_NUM_CHANGE,"
|
|
"HOLD_RATIO_CHANGE,DIRECTION,SHARES_TYPE,HOLDER_NATURE,REPORT_TYPE,"
|
|
"HOLD_CHANGE,NOTICE_DATE,HOLD_RATIO_YOY,REPORT_DATE_NAME,REFERENCE_MARKET_CAP"
|
|
)
|
|
|
|
|
|
def _build_params(
|
|
secucode: str,
|
|
start_date: str,
|
|
end_date: str,
|
|
) -> dict[str, str]:
|
|
"""构造 RPT_DMSK_HOLDERS 查询参数."""
|
|
return {
|
|
"reportName": _REPORT_NAME,
|
|
"columns": _COLUMNS,
|
|
"quoteColumns": "",
|
|
"filter": f'(SECUCODE="{secucode}")(END_DATE>=\'{start_date}\')(END_DATE<=\'{end_date}\')',
|
|
"pageNumber": "",
|
|
"pageSize": "",
|
|
"sortTypes": "-1,1",
|
|
"sortColumns": "END_DATE,RANK",
|
|
"client": "SW",
|
|
"source": "CHOICE",
|
|
}
|
|
|
|
|
|
def _fetch_one(secucode: str, years: int) -> pd.DataFrame:
|
|
"""拉一只股票的十大股东明细."""
|
|
today = pd.Timestamp.today()
|
|
start = (today - pd.DateOffset(years=years)).strftime("%Y-%m-%d")
|
|
end = today.strftime("%Y-%m-%d")
|
|
|
|
rows = fetch_report(_build_params(secucode, start, end))
|
|
df = pd.DataFrame(rows)
|
|
if not df.empty:
|
|
df["SECUCODE"] = secucode
|
|
return df
|
|
|
|
|
|
def get_holders(secucodes: list[str], years: int = 3) -> pd.DataFrame:
|
|
"""抓取指定股票的十大股东明细并按股票合并。"""
|
|
codes = [code.strip() for code in secucodes if code.strip()]
|
|
if not codes:
|
|
return pd.DataFrame()
|
|
|
|
frames: list[pd.DataFrame] = []
|
|
for code in codes:
|
|
df = _fetch_one(code, years)
|
|
if not df.empty:
|
|
frames.append(df)
|
|
|
|
if not frames:
|
|
return pd.DataFrame()
|
|
return pd.concat(frames, ignore_index=True)
|
|
|