Upload capital_flow_analyzer.py
Browse files- capital_flow_analyzer.py +142 -21
capital_flow_analyzer.py
CHANGED
@@ -30,28 +30,149 @@ class CapitalFlowAnalyzer:
|
|
30 |
return cached_data
|
31 |
|
32 |
# 从akshare获取数据
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
try:
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
55 |
# 缓存结果
|
56 |
self.data_cache[cache_key] = (datetime.now(), result)
|
57 |
|
|
|
30 |
return cached_data
|
31 |
|
32 |
# 从akshare获取数据
|
33 |
+
try:
|
34 |
+
concept_data = ak.stock_fund_flow_concept(symbol=period)
|
35 |
+
|
36 |
+
# 记录实际的列名,以便调试
|
37 |
+
self.logger.info(f"Actual columns: {list(concept_data.columns)}")
|
38 |
+
self.logger.info(f"Column count: {len(concept_data.columns)}")
|
39 |
+
|
40 |
+
# 处理数据
|
41 |
+
result = []
|
42 |
+
for _, row in concept_data.iterrows():
|
43 |
+
try:
|
44 |
+
# 使用更灵活的方式获取列数据,适应不同的列名结构
|
45 |
+
item = {}
|
46 |
+
|
47 |
+
# 尝试多种可能的列名
|
48 |
+
if "序号" in concept_data.columns:
|
49 |
+
item["rank"] = int(row.get("序号", 0))
|
50 |
+
elif "排名" in concept_data.columns:
|
51 |
+
item["rank"] = int(row.get("排名", 0))
|
52 |
+
else:
|
53 |
+
item["rank"] = 0
|
54 |
+
|
55 |
+
# 行业/概念名称
|
56 |
+
if "行业" in concept_data.columns:
|
57 |
+
item["sector"] = row.get("行业", "")
|
58 |
+
elif "概念名称" in concept_data.columns:
|
59 |
+
item["sector"] = row.get("概念名称", "")
|
60 |
+
elif "名称" in concept_data.columns:
|
61 |
+
item["sector"] = row.get("名称", "")
|
62 |
+
else:
|
63 |
+
# 尝试找到包含"名称"的列
|
64 |
+
name_cols = [col for col in concept_data.columns if "名称" in col or "行业" in col]
|
65 |
+
if name_cols:
|
66 |
+
item["sector"] = row.get(name_cols[0], "")
|
67 |
+
else:
|
68 |
+
item["sector"] = ""
|
69 |
+
|
70 |
+
# 公司家数
|
71 |
+
if "公司家数" in concept_data.columns:
|
72 |
+
item["company_count"] = int(row.get("公司家数", 0))
|
73 |
+
else:
|
74 |
+
item["company_count"] = 0
|
75 |
+
|
76 |
+
# 行业指数
|
77 |
+
if "行业指数" in concept_data.columns:
|
78 |
+
item["sector_index"] = float(row.get("行业指数", 0))
|
79 |
+
elif "概念指数" in concept_data.columns:
|
80 |
+
item["sector_index"] = float(row.get("概念指数", 0))
|
81 |
+
else:
|
82 |
+
item["sector_index"] = 0
|
83 |
+
|
84 |
+
# 涨跌幅
|
85 |
+
if "阶段涨跌幅" in concept_data.columns:
|
86 |
+
item["change_percent"] = self._parse_percent(row.get("阶段涨跌幅", "0%"))
|
87 |
+
elif "涨跌幅" in concept_data.columns:
|
88 |
+
item["change_percent"] = self._parse_percent(row.get("涨跌幅", "0%"))
|
89 |
+
else:
|
90 |
+
# 尝试找到包含"涨跌"的列
|
91 |
+
change_cols = [col for col in concept_data.columns if "涨跌" in col]
|
92 |
+
if change_cols:
|
93 |
+
item["change_percent"] = self._parse_percent(row.get(change_cols[0], "0%"))
|
94 |
+
else:
|
95 |
+
item["change_percent"] = 0
|
96 |
+
|
97 |
+
# 资金流向数据
|
98 |
+
# 尝试多种可能的列名组合
|
99 |
+
if "流入资金" in concept_data.columns and "流出资金" in concept_data.columns:
|
100 |
+
item["inflow"] = float(row.get("流入资金", 0))
|
101 |
+
item["outflow"] = float(row.get("流出资金", 0))
|
102 |
+
item["net_flow"] = float(row.get("净额", 0))
|
103 |
+
elif "主力净流入-净额" in concept_data.columns:
|
104 |
+
item["inflow"] = 0 # 这种情况下可能没有单独的流入流出数据
|
105 |
+
item["outflow"] = 0
|
106 |
+
item["net_flow"] = float(row.get("主力净流入-净额", 0))
|
107 |
+
else:
|
108 |
+
# 尝试找到包含特定关键词的列
|
109 |
+
inflow_cols = [col for col in concept_data.columns if "流入" in col and "净" not in col]
|
110 |
+
outflow_cols = [col for col in concept_data.columns if "流出" in col and "净" not in col]
|
111 |
+
net_cols = [col for col in concept_data.columns if "净" in col or "净额" in col]
|
112 |
+
|
113 |
+
if inflow_cols:
|
114 |
+
item["inflow"] = float(row.get(inflow_cols[0], 0))
|
115 |
+
else:
|
116 |
+
item["inflow"] = 0
|
117 |
+
|
118 |
+
if outflow_cols:
|
119 |
+
item["outflow"] = float(row.get(outflow_cols[0], 0))
|
120 |
+
else:
|
121 |
+
item["outflow"] = 0
|
122 |
+
|
123 |
+
if net_cols:
|
124 |
+
item["net_flow"] = float(row.get(net_cols[0], 0))
|
125 |
+
else:
|
126 |
+
# 如果没有净额列,尝试计算
|
127 |
+
item["net_flow"] = item["inflow"] - item["outflow"]
|
128 |
+
|
129 |
+
result.append(item)
|
130 |
+
except Exception as e:
|
131 |
+
self.logger.warning(f"Error processing row in concept fund flow: {str(e)}")
|
132 |
+
continue
|
133 |
+
|
134 |
+
except ValueError as e:
|
135 |
+
self.logger.error(f"ValueError in stock_fund_flow_concept: {str(e)}")
|
136 |
+
# 如果出现ValueError(可能是列名不匹配),尝试不重命名列直接处理数据
|
137 |
try:
|
138 |
+
concept_data = ak.stock_fund_flow_concept(symbol=period)
|
139 |
+
self.logger.info(f"Retrying with original columns: {list(concept_data.columns)}")
|
140 |
+
|
141 |
+
# 简化处理逻辑,直接使用原始列名
|
142 |
+
result = []
|
143 |
+
for _, row in concept_data.iterrows():
|
144 |
+
try:
|
145 |
+
item = {}
|
146 |
+
# 尝试将每一列的数据都添加到结果中
|
147 |
+
for col in concept_data.columns:
|
148 |
+
try:
|
149 |
+
# 尝试转换为数值类型
|
150 |
+
if isinstance(row[col], str) and '%' in row[col]:
|
151 |
+
item[col] = self._parse_percent(row[col])
|
152 |
+
else:
|
153 |
+
try:
|
154 |
+
item[col] = float(row[col])
|
155 |
+
except (ValueError, TypeError):
|
156 |
+
item[col] = row[col]
|
157 |
+
except Exception:
|
158 |
+
item[col] = row[col]
|
159 |
+
|
160 |
+
# 添加一些标准化的键,以便与现有代码兼容
|
161 |
+
if "名称" in item or "概念名称" in item or "行业" in item:
|
162 |
+
item["sector"] = item.get("名称", item.get("概念名称", item.get("行业", "")))
|
163 |
+
|
164 |
+
if "涨跌幅" in item:
|
165 |
+
item["change_percent"] = item["涨跌幅"]
|
166 |
+
|
167 |
+
result.append(item)
|
168 |
+
except Exception as e:
|
169 |
+
self.logger.warning(f"Error processing row with original columns: {str(e)}")
|
170 |
+
continue
|
171 |
+
except Exception as e2:
|
172 |
+
self.logger.error(f"Failed to process with original columns: {str(e2)}")
|
173 |
+
# 如果再次失败,返回模拟数据
|
174 |
+
result = self._generate_mock_concept_fund_flow(period)
|
175 |
+
|
176 |
# 缓存结果
|
177 |
self.data_cache[cache_key] = (datetime.now(), result)
|
178 |
|