hikyuu 2.6.8.4__py3-none-manylinux2014_x86_64.whl → 2.6.9__py3-none-manylinux2014_x86_64.whl

hikyuu/data/download_block.py

@@ -0,0 +1,318 @@
+#!/usr/bin/python
+# -*- coding: utf8 -*-
+#
+# Create on: 2025-10-05
+#    Author: fasiondog
+
+from hikyuu.data.common import modifiy_code
+from hikyuu.util import *
+from hikyuu.fetcher.stock.zh_block_em import *
+import os
+from datetime import datetime
+
+
+def sanitize_filename(filename: str) -> str:
+    """
+    将字符串中影响文件创建的相关目录字符替换为下划线
+
+    Args:
+        filename: 输入的文件名字符串
+
+    Returns:
+        处理后的文件名字符串，其中非法字符已被替换为下划线
+    """
+    if not filename:
+        return filename
+
+    # 需要替换的非法字符（Windows和Unix系统中的非法字符）
+    illegal_chars = '<>:"/\\|?*'
+
+    # 替换非法字符为下划线
+    sanitized = filename
+    for char in illegal_chars:
+        sanitized = sanitized.replace(char, '_')
+
+    return sanitized
+
+
+_BLOCK_SAVE_PATH = os.path.expanduser('~') + '/.hikyuu/downloads/block'
+if not os.path.exists(_BLOCK_SAVE_PATH):
+    os.makedirs(_BLOCK_SAVE_PATH)
+
+
+def read_block_from_path(block_path=_BLOCK_SAVE_PATH):
+    """
+    从指定目录下读取block信息
+
+    :param block_path: block根目录路径，默认为 _BLOCK_SAVE_PATH
+    :return: 返回格式类似于 {'block': {'category1': {'name1': [code1, code2]}}, 
+                            'block_info': {'category1: {'name1': 'code1', 'name2': ''}}
+    """
+    result = {
+        'block': {},
+        'block_info': {}
+    }
+
+    if not os.path.exists(block_path):
+        return result
+
+    # 遍历所有分类目录
+    for category in os.listdir(block_path):
+        category_path = os.path.join(block_path, category)
+        if not os.path.isdir(category_path):
+            continue
+
+        result['block'][category] = {}
+        result['block_info'][category] = {}
+
+        # 遍历分类目录下的所有.txt文件
+        for file in os.listdir(category_path):
+            if file.endswith('.txt'):
+                file_path = os.path.join(category_path, file)
+                # 文件名作为板块名称（去掉扩展名）
+                block_name = os.path.splitext(file)[0]
+
+                # 解析文件名，格式为 "code_name" 或 "name"
+                if '_' in block_name:
+                    parts = block_name.split('_', 1)  # 只分割第一个下划线
+                    code_part = parts[0]
+                    name_part = parts[1]
+                    # 判断code部分是否为纯数字或字母数字组合
+                    if code_part.replace('_', '').isalnum():
+                        result['block_info'][category][name_part] = code_part
+                        block_key = name_part
+                    else:
+                        result['block_info'][category][block_name] = ""
+                        block_key = block_name
+                else:
+                    result['block_info'][category][block_name] = ""
+                    block_key = block_name
+
+                # 读取文件中的股票代码
+                codes = []
+                with open(file_path, 'r', encoding='utf-8') as f:
+                    for line in f:
+                        code = line.strip()
+                        if code:
+                            codes.append(code)
+
+                result['block'][category][block_key] = codes
+
+    return result
+
+
+def is_file_can_download(filepath, sec_limit=5 * 24 * 60 * 60):
+    """判断文件创建时间是否超过时长限制可以下载"""
+    if not os.path.exists(filepath):
+        return True
+
+    file_mtime = os.path.getmtime(filepath)
+    file_datetime = datetime.fromtimestamp(file_mtime)
+    now = datetime.now()
+
+    # 计算时间差是否超过时间限制
+    return (now - file_datetime).total_seconds() > sec_limit
+
+
+@hku_catch()
+def save_block(stkcodes: list,  filename: str):
+    """将板块数据保存为CSV文件，格式为板块名称/code"""
+    with open(filename, 'w', encoding='utf-8') as f:
+        for code in stkcodes:
+            f.write(f"{code}\n")
+    # hku_info(f"已保存至 {filename}")
+
+
+@hku_catch(ret={}, trace=False)
+def down_em_all_hybk_info():
+    """下载东财所有行业板块列表"""
+    save_path = f'{_BLOCK_SAVE_PATH}/行业板块'
+    if not os.path.exists(save_path):
+        os.makedirs(save_path)
+    blk_list = get_hybk_names()
+    time.sleep(random.uniform(1, 3))
+    total = len(blk_list)
+    for i, blk in enumerate(blk_list):
+        filename = f"{save_path}/{blk[0]}_{blk[1]}.txt"
+        if is_file_can_download(filename, 5 * 24 * 60 * 60):
+            stk_codes = get_hybk_cons_code(blk[0])
+            hku_info(f"{i+1}|{total} 获取行业板块{blk[1]}成分: {len(stk_codes)}")
+            stk_codes = [modifiy_code(code) for code in stk_codes]
+            stk_codes = [code for code in stk_codes if code is not None]
+            save_block(stk_codes, filename)
+            time.sleep(random.uniform(1, 3))
+
+
+@hku_catch(ret={}, trace=False)
+def down_em_all_gnbk_info():
+    """获取所有概念版本列表"""
+    save_path = f'{_BLOCK_SAVE_PATH}/概念板块'
+    if not os.path.exists(save_path):
+        os.makedirs(save_path)
+    blk_names = stock_board_concept_name_em()['板块名称']
+    not_need_blks = set(["昨日连板_含一字", "昨日涨停_含一字", "昨日涨停"])
+    total = len(blk_names)
+    for i, blk_name in enumerate(blk_names):
+        if blk_name in not_need_blks:
+            continue
+        filename = f"{save_path}/{sanitize_filename(blk_name)}.txt"
+        if is_file_can_download(filename, 30 * 24 * 60 * 60):
+            stk_codes = stock_board_concept_cons_em(blk_name)
+            stk_codes = stk_codes['代码'].to_list()
+            stk_codes = [modifiy_code(code) for code in stk_codes]
+            stk_codes = [code for code in stk_codes if code is not None]
+            hku_info(f"{i+1}|{total} 获取概念板块{blk_name}成分: {len(stk_codes)}")
+            save_block(stk_codes, filename)
+            time.sleep(random.uniform(1, 3))
+
+
+@hku_catch(ret={}, trace=False)
+def down_em_all_dybk_info():
+    """获取所有地域板块列表"""
+    save_path = f'{_BLOCK_SAVE_PATH}/地域板块'
+    if not os.path.exists(save_path):
+        os.makedirs(save_path)
+    blk_list = get_dybk_names()
+
+    url = "http://13.push2.eastmoney.com/api/qt/clist/get"
+    params = {
+        "pn": "1",
+        "pz": str(em_num_per_page),
+        "po": "1",
+        "np": "1",
+        "ut": "bd1d9ddb04089700cf9c27f6f7426281",
+        "fltt": "2",
+        "invt": "2",
+        "fid": "f3",
+        # "fs": f"b:{stock_board_code} f:!50",
+        "fields": "f12",
+        "_": "1626081702127",
+    }
+
+    total = len(blk_list)
+    for i, v in enumerate(blk_list):
+        blk_code, blk_name = v[0], v[1]
+        filename = f"{save_path}/{blk_name}.txt"
+        if not is_file_can_download(filename, 10 * 24 * 60 * 60):
+            continue
+
+        params["fs"] = f"b:{blk_code} f:!50"
+        params["pn"] = 1
+        time.sleep(random.uniform(1, 3))
+        r = requests.get(url, params=params, timeout=15)
+        data = r.json()
+        if data["data"] is None:
+            continue
+
+        stk_json = r.json()
+        stk_json = stk_json["data"]["diff"]
+        stk_codes = []
+
+        total_page = math.ceil(data["data"]["total"] / em_num_per_page)
+        for page in range(2, total_page + 1):
+            params["pn"] = page
+            r = requests.get(url, params=params, timeout=15)
+            stk_json = r.json()
+            stk_json = stk_json["data"]["diff"]
+            stk_codes.extend([f"{v['f12']}" for v in stk_json])
+            time.sleep(random.uniform(1, 3))
+
+        stk_codes = [modifiy_code(code) for code in stk_codes]
+        stk_codes = [code for code in stk_codes if code is not None]
+        save_block(stk_codes, filename)
+        hku_info(f'{i+1}|{total} 获取地域板块{blk_name}成分: {len(stk_codes)}')
+
+
+@hku_catch(ret={}, trace=False)
+def download_all_zsbk_info():
+    """获取所有指数成分股列表"""
+    save_path = f'{_BLOCK_SAVE_PATH}/指数板块'
+    if not os.path.exists(save_path):
+        os.makedirs(save_path)
+
+    blk_info = ak.index_stock_info()
+    time.sleep(random.uniform(1, 3))
+
+    blk_info['index_code'] = blk_info['index_code'].astype(str)  # 确保是字符串类型
+    df_000 = blk_info[blk_info['index_code'].str.startswith('000')].reset_index(drop=True)  # 000前缀
+    df_399 = blk_info[blk_info['index_code'].str.startswith('399')].reset_index(drop=True)  # 399前缀
+
+    # 2. 交替合并两个DataFrame
+    merged_rows = []
+    max_length = max(len(df_000), len(df_399))  # 取两个DataFrame的最大长度
+
+    for i in range(max_length):
+        # 先加000前缀的行（如果存在）
+        if i < len(df_000):
+            merged_rows.append(df_000.iloc[i])
+        # 再加399前缀的行（如果存在）
+        if i < len(df_399):
+            merged_rows.append(df_399.iloc[i])
+
+    # 3. 转换为DataFrame
+    blk_info = pd.DataFrame(merged_rows).reset_index(drop=True)
+
+    not_need_blks = set(["000012", "000013", "000022", "000061", "000101", "000188",
+                         "000817", "000847", "000849", "000850", "000851", "000853",
+                         "000854", "000856", "000857", "000858", "000923", "000973",
+                         "000974", "000996", "000997", "000999", "399415", "399416"])
+
+    failed_sina = 0
+    blk_set = {}
+    blk_codes = blk_info["index_code"]
+    blk_names = blk_info["display_name"]
+    total = len(blk_codes)
+    for i in range(total):
+        blk_name = blk_names[i]
+        blk_code = blk_codes[i]
+        # 沪深指数有重复，避免深指覆盖
+        if blk_name in blk_set or blk_code in not_need_blks:
+            continue
+
+        market_block_code = f'SH{blk_code}' if blk_code.startswith('0') else f'SZ{blk_code}'
+        filename = f"{save_path}/{market_block_code}_{blk_name}.txt"
+        if not is_file_can_download(filename, 10 * 24 * 60 * 60):
+            continue
+
+        try:
+            if blk_code[:3] == "399":
+                if failed_sina >= 5:
+                    continue
+                stk_codes = ak.index_stock_cons(symbol=blk_code)
+                stk_codes = stk_codes['品种代码'].to_list()
+            else:
+                stk_codes = ak.index_stock_cons_csindex(symbol=blk_code)
+                stk_codes = stk_codes['成分券代码'].to_list()
+
+            stk_codes = [modifiy_code(code) for code in stk_codes]
+            stk_codes = [code for code in stk_codes if code is not None]
+            save_block(stk_codes, filename)
+            hku_info("{}|{} 获取指数板块 {}|{} 成分: {}", i, total, blk_code, blk_name, len(stk_codes))
+            blk_set[blk_name] = 1
+        except KeyboardInterrupt:
+            break
+        except Exception as e:
+            print(f"Failed! {i}, {blk_code}, {blk_name} {str(e)}")
+            if blk_code.startswith("399"):
+                failed_sina += 1
+            # raise e
+        time.sleep(random.uniform(1, 3))
+
+
+def download_block_info():
+    down_em_all_hybk_info()
+    # down_em_all_dybk_info()
+    # down_em_all_gnbk_info()
+    download_all_zsbk_info()
+    
+
+if __name__ == "__main__":
+    # down_em_all_hybk_info()
+    # down_em_all_dybk_info()
+    # down_em_all_gnbk_info()
+    # download_all_zsbk_info()
+
+    ret = read_block_from_path()
+    # print(ret['block'])
+    print(ret['block_info'])
+    print(tuple(ret['block'].keys()))

hikyuu/data/em_block_to_clickhouse.py

@@ -4,85 +4,32 @@
 # Create on: 20240102
 #    Author: fasiondog
 
-from concurrent.futures import ThreadPoolExecutor
-from hikyuu.data.common import MARKET, get_stk_code_name_list
 from hikyuu.util import *
 from hikyuu.fetcher.stock.zh_block_em import *
+from hikyuu.data.download_block import *
 
 
-def em_import_block_to_clickhouse(connect, code_market_dict, categorys=('行业板块', '指数板块')):
-    all_block_info = {}
-
-    with ThreadPoolExecutor(4) as executor:
-        if '行业板块' in categorys:
-            t1 = executor.submit(get_all_hybk_info, code_market_dict)
-
-        if '概念板块' in categorys:
-            t2 = executor.submit(get_all_gnbk_info, code_market_dict)
-
-        if '地域板块' in categorys:
-            t3 = executor.submit(get_all_dybk_info, code_market_dict)
-
-        if '指数板块' in categorys:
-            t4 = executor.submit(get_all_zsbk_info, code_market_dict)
-
-        success_fetch_hy = False
-        if '行业板块' in categorys:
-            x = t1.result()
-            hku_info("获取行业板块信息完毕")
-            if x:
-                all_block_info["行业板块"] = x
-                success_fetch_hy = True
-
-        success_fetch_gn = False
-        if '概念板块' in categorys:
-            x = t2.result()
-            hku_info("获取概念板块信息完毕")
-            if x:
-                all_block_info["概念板块"] = x
-                success_fetch_gn = True
-
-        success_fetch_dy = False
-        if '地域板块' in categorys:
-            x = t3.result()
-            hku_info("获取地域板块信息完毕")
-            if x:
-                all_block_info["地域板块"] = x
-                success_fetch_dy = True
-
-        success_fetch_zs = False
-        if '指数板块' in categorys:
-            x = t4.result()
-            hku_info("获取指数板块信息完毕")
-            if x:
-                all_block_info["指数板块"] = x
-                success_fetch_zs = True
-
-    blks = []
-    if success_fetch_hy:
-        blks.append('行业板块')
-    if success_fetch_gn:
-        blks.append('概念板块')
-    if success_fetch_dy:
-        blks.append('地域板块')
-    if success_fetch_zs:
-        blks.append('指数板块')
+def em_import_block_to_clickhouse(connect):
+    download_block_info()
+
+    blocks_info = read_block_from_path()
+    blks = blocks_info['block']
+    blks_info = blocks_info['block_info']
 
     if not blks:
-        return
+        return 0
 
     hku_info("更新数据库")
-    if len(blks) == 1:
-        sql = f"delete from hku_base.block where category in ('{blks[0]}')"
-    else:
-        sql = f"delete from hku_base.block where category in {tuple(blks)}"
+    sql = f"delete from hku_base.block where category in {tuple(blks.keys())}"
+    connect.command(sql)
+    sql = f"delete from hku_base.blockindex where category in {tuple(blks.keys())}"
     connect.command(sql)
 
     insert_records = []
 
-    for category in all_block_info:
-        for name in all_block_info[category]:
-            for code in all_block_info[category][name]:
+    for category, blocks in blks.items():
+        for name, codes in blocks.items():
+            for code in codes:
                 insert_records.append((category, name, code))
 
     if insert_records:
@@ -91,6 +38,17 @@ def em_import_block_to_clickhouse(connect, code_market_dict, categorys=('行业
                                            data=insert_records)
         connect.insert(context=ic)
 
+    index_records = []
+    for category, blocks in blks_info.items():
+        for name, index_code in blocks.items():
+            if len(index_code) == 8:
+                index_records.append((category, name, index_code))
+    if index_records:
+        ic = connect.create_insert_context(table='blockindex', database='hku_base',
+                                           column_names=['category', 'name', 'index_code'],
+                                           data=index_records)
+        connect.insert(context=ic)
+
     return len(insert_records)
 
 
@@ -109,12 +67,6 @@ if __name__ == "__main__":
     client = clickhouse_connect.get_client(
         host=host, username=user, password=password)
 
-    x = get_stk_code_name_list(MARKET.SH)
-    code_market_dict = {}
-    for v in x:
-        code_market_dict[v["code"]] = MARKET.SH
-    # print(code_market_dict)
-
-    em_import_block_to_clickhouse(client, code_market_dict, categorys=('行业板块', ))
+    em_import_block_to_clickhouse(client)
 
     client.close()

hikyuu/data/em_block_to_mysql.py

@@ -4,93 +4,49 @@
 # Create on: 20240102
 #    Author: fasiondog
 
-from concurrent.futures import ThreadPoolExecutor
-from hikyuu.data.common import MARKET, get_stk_code_name_list
+
 from hikyuu.util import *
 from hikyuu.fetcher.stock.zh_block_em import *
+from hikyuu.data.download_block import *
+
 
+def em_import_block_to_mysql(connect):
+    download_block_info()
 
-def em_import_block_to_mysql(connect, code_market_dict, categorys=('行业板块', '概念板块', '地域板块', '指数板块')):
-    all_block_info = {}
-
-    with ThreadPoolExecutor(4) as executor:
-        if '行业板块' in categorys:
-            t1 = executor.submit(get_all_hybk_info, code_market_dict)
-
-        if '概念板块' in categorys:
-            t2 = executor.submit(get_all_gnbk_info, code_market_dict)
-
-        if '地域板块' in categorys:
-            t3 = executor.submit(get_all_dybk_info, code_market_dict)
-
-        if '指数板块' in categorys:
-            t4 = executor.submit(get_all_zsbk_info, code_market_dict)
-
-        success_fetch_hy = False
-        if '行业板块' in categorys:
-            x = t1.result()
-            hku_info("获取行业板块信息完毕")
-            if x:
-                all_block_info["行业板块"] = x
-                success_fetch_hy = True
-
-        success_fetch_gn = False
-        if '概念板块' in categorys:
-            x = t2.result()
-            hku_info("获取概念板块信息完毕")
-            if x:
-                all_block_info["概念板块"] = x
-                success_fetch_gn = True
-
-        success_fetch_dy = False
-        if '地域板块' in categorys:
-            x = t3.result()
-            hku_info("获取地域板块信息完毕")
-            if x:
-                all_block_info["地域板块"] = x
-                success_fetch_dy = True
-
-        success_fetch_zs = False
-        if '指数板块' in categorys:
-            x = t4.result()
-            hku_info("获取指数板块信息完毕")
-            if x:
-                all_block_info["指数板块"] = x
-                success_fetch_zs = True
-
-    blks = []
-    if success_fetch_hy:
-        blks.append('行业板块')
-    if success_fetch_gn:
-        blks.append('概念板块')
-    if success_fetch_dy:
-        blks.append('地域板块')
-    if success_fetch_zs:
-        blks.append('指数板块')
+    blocks_info = read_block_from_path()
+    blks = blocks_info['block']
+    blks_info = blocks_info['block_info']
 
     if not blks:
-        return
+        return 0
 
     hku_info("更新数据库")
     cur = connect.cursor()
-    if len(blks) == 1:
-        sql = f"delete from hku_base.block where category in ('{blks[0]}')"
-    else:
-        sql = f"delete from hku_base.block where category in {tuple(blks)}"
+    sql = f"delete from hku_base.block where category in {tuple(blks.keys())}"
+    cur.execute(sql)
+    sql = f"delete from hku_base.BlockIndex where category in {tuple(blks.keys())}"
     cur.execute(sql)
 
     insert_records = []
 
-    for category in all_block_info:
-        for name in all_block_info[category]:
-            for code in all_block_info[category][name]:
+    for category, blocks in blks.items():
+        for name, codes in blocks.items():
+            for code in codes:
                 insert_records.append((category, name, code))
 
     if insert_records:
         sql = "insert into hku_base.block (category, name, market_code) values (%s,%s,%s)"
-        hku_info(f"insert block records: {len(insert_records)}")
         cur.executemany(sql, insert_records)
 
+    index_records = []
+    for category, blocks in blks_info.items():
+        for name, index_code in blocks.items():
+            if len(index_code) == 8:
+                index_records.append((category, name, index_code))
+    if index_records:
+        sql = "insert into hku_base.BlockIndex (category, name, market_code) values (%s,%s,%s)"
+        cur.executemany(sql, index_records)
+
     connect.commit()
     cur.close()
     return len(insert_records)
@@ -111,11 +67,5 @@ if __name__ == "__main__":
     connect = mysql.connector.connect(user=usr, password=pwd, host=host, port=port)
     create_database(connect)
 
-    x = get_stk_code_name_list(MARKET.SH)
-    code_market_dict = {}
-    for v in x:
-        code_market_dict[v["code"]] = MARKET.SH
-    # print(code_market_dict)
-
-    em_import_block_to_mysql(connect, code_market_dict)
+    em_import_block_to_mysql(connect)
     connect.close()