6.13提交，语义处理测试

2025-06-13 10:11:42 +08:00
parent 026f9bd70c
commit 89b0154217
11 changed files with 2481 additions and 0 deletions
@@ -0,0 +1,221 @@
+def Dictionary_content_mapping(input_str, data, key="指标名称"):
+    import re
+
+    match = re.search(r'【(.*?)】', input_str)
+
+    if match:
+        extracted = match.group(1)
+    else:
+        return None  # 如果没有匹配到，提前返回
+
+    for i in range(len(data)):
+        if data[i].get(key) == extracted:
+            return data[i]
+
+    return None
+
+def find_target_item(input_str, data):
+    result = [None, None]
+    temp = None
+    for item in data:
+        if isinstance(item["指标描述"], dict):
+            if item["指标描述"]["指标映射"][0] == input_str:
+                result[0] = item["指标描述"]["映射规则"]
+                temp = item["指标描述"]["指标映射"][0]
+
+    for item in data:
+        if item["指标名称"] == temp:
+            result[1] = item['指标描述']
+            return result
+
+def find_target_items(ele, input_str, data):
+    result = [None, ele]
+    for item in data:
+        if isinstance(item["指标描述"], dict):
+            if item["指标名称"] == input_str or item["指标描述"]["指标映射"][0] == input_str:
+                if len(item["指标描述"]["指标映射"]) == 1:
+                    result[0] = item["指标描述"]["指标映射"][0]
+    return result
+
+def judge_exists(input_str, data):
+    for item in data:
+        if isinstance(item["指标描述"], dict):
+            if item["指标名称"] == input_str or item["指标描述"]["指标映射"][0] == input_str:
+                return True
+    return False
+
+def judge_str(ceshi, data):
+    for item in data:
+        if isinstance(item["指标描述"], str):
+            if item["指标名称"] == ceshi["指标描述"]["指标映射"][0]:
+                return True
+    return False
+
+def pre_mapping(keywords, data):
+    import re
+
+    match = re.search(r'【(.*?)】', keywords)
+
+    if match:
+        extracted = match.group(1)
+
+    for i in range(len(data)):
+        if data[i]["指标名称"] == extracted:
+            ceshi = data[i]
+            break
+
+    if isinstance(ceshi["指标描述"], str):
+        return ceshi["指标描述"]
+
+    elif isinstance(ceshi["指标描述"], dict):
+        if ceshi["指标描述"]["映射规则"] == "1":
+            temp = ceshi["指标描述"]["指标映射"][0]
+            return f"模糊查找一下【{temp}】"
+        else:
+            if len(ceshi["指标描述"]["指标映射"]) == 1:
+                temp0, temp1 = find_target_item(ceshi["指标描述"]["指标映射"][0], data)
+                return f"{temp1}，换算规则：【{temp0}】"
+
+            elif len(ceshi["指标描述"]["指标映射"]) > 1:
+                
+                
+                result = []
+                if judge_str(ceshi, data) == True:
+                    for ele in ceshi["指标描述"]["指标映射"]:
+                        for item in data:
+                            if isinstance(item["指标描述"], str) and item["指标名称"] == ele:
+                                temp1 = item["指标描述"]
+                                temp2 = ceshi["指标描述"]["映射规则"]
+                                result.append(f"{temp1}，换算规则：【{temp2}】")
+
+                else:
+                
+                    for item in ceshi["指标描述"]["指标映射"]:
+                        temp0, temp1 = find_target_items(ceshi["指标描述"]["映射规则"], item, data)
+    
+                        if temp0 is None and temp1 is None:
+                            pass
+                        elif temp0 != None and temp1 is None:
+                            if judge_exists(item, data):
+                                temp1 = item
+                                # temp0 = find_target_items(ceshi["指标描述"]["映射规则"], item, data)
+                                result.append(f"模糊查找一下【{temp0}】，换算规则：【{temp1}】")
+                            else:
+                                continue
+    
+                        elif temp0 != None and temp1 != None:
+                            if judge_exists(item, data):
+                                # temp0, temp1 = find_target_items(ceshi["指标描述"]["映射规则"], item, data)
+                                result.append(f"模糊查找一下【{temp0}】，换算规则：【{temp1}】")
+                            else:
+                                continue
+
+                return result
+
+
+def extract_concrete_info(outputs):
+    import re
+    from typing import List
+
+    """
+    从多个句子中提取第一个“【】”作为查找信息，最后一个“【】”作为换算规则，
+    返回格式为：[合并的查找句子, 换算规则]
+    """
+    prefixes = []
+    suffix = ''
+
+    for item in outputs:
+        matches = re.findall(r'【([^】]+)】', item)
+        if len(matches) >= 2:
+            prefixes.append(f"查找一下【{matches[0]}】")
+            # 假设所有换算规则一致，取第一个即可
+            if not suffix:
+                suffix = f'换算规则：【{matches[-1]}】'
+
+    if not prefixes or not suffix:
+        return []
+
+    return ['; '.join(prefixes), suffix]
+
+
+def extract_query_prefix_list(input_list):
+    import re
+    """
+    输入一个字符串列表，提取每个字符串中符合格式的前缀内容（例如：'查找一下【样式】'）
+
+    参数:
+        input_list (list[str]): 包含描述性语句的字符串列表
+
+    返回:
+        list[str]: 提取出的前缀部分列表（如 '查找一下【大板式】'）
+    """
+    pattern = r'(查找一下【[^】]+】)'
+    return [re.match(pattern, text).group(1) for text in input_list if re.match(pattern, text)]
+
+
+
+def pre_mapping2(keywords, data):
+    import re
+
+    # 提取关键字中中括号内的内容
+    match = re.search(r'【(.*?)】', keywords)
+    if not match:
+        return "未找到匹配的关键字"
+
+    extracted = match.group(1)
+
+    # 查找对应的指标项
+    ceshi = next((item for item in data if item["指标名称"] == extracted), None)
+    if not ceshi:
+        return "未找到对应的指标"
+
+    desc = ceshi.get("指标描述")
+
+    if isinstance(desc, str):
+        return f"测试：{desc}"
+
+    elif isinstance(desc, dict):
+        mapping_rule = desc.get("映射规则")
+        mappings = desc.get("指标映射", [])
+
+        # 单一映射规则为1时
+        if mapping_rule == "1":
+            return f"模糊查找一下【{mappings[0]}】"
+
+        # 仅有一个映射项
+        if len(mappings) == 1:
+            temp0, temp1 = find_target_item(mappings[0], data)
+            if "【" in temp0:
+                return f"{temp1}，换算规则：【{temp0}】"
+            else:
+                return f"查找一下【{temp1}】，换算规则：【{temp0}】"
+
+
+        # 多个映射项
+        result = []
+        if judge_str(ceshi, data):
+            for ele in mappings:
+                item = next((d for d in data if d["指标名称"] == ele and isinstance(d["指标描述"], str)), None)
+                if item:
+                    result.append(f"{item['指标描述']}，换算规则：【{mapping_rule}】")
+        else:
+            for item_name in mappings:
+                temp0, temp1 = find_target_items(mapping_rule, item_name, data)
+
+                if temp0 is None and temp1 is None:
+                    continue
+
+                if judge_exists(item_name, data):
+                    if temp1 is None:
+                        temp1 = item_name
+                    result.append(f"模糊查找一下【{temp0}】，换算规则：【{temp1}】")
+
+        return result or "未匹配到有效的映射项"
+
+    else:
+        return "不支持的指标描述格式"
+
+
+
+    
+