You need to enable JavaScript to run this app. 如何用Beautiful Soup提取XML中内嵌于标签的职业等特征值
优惠活动
大模型
产品
解决方案
定价
更多

如何用Beautiful Soup提取XML中内嵌于标签的职业等特征值

解决思路

你遇到的问题是不同字段的存储位置不一致:gender、given这类字段的值存在interp标签的value属性中,而occupation字段的值存在rs标签的内部文本中,现有代码统一拿value属性自然取不到职业值,调整逻辑适配两种存储规则即可。

修改后的代码

你可以在最后取值时增加兼容逻辑:优先取标签的value属性,属性不存在时自动取标签内部的文本内容,同时可以调整职业标签的匹配规则适配XML里的join关联逻辑,完整修改如下:

case_feature_keys = ['year', 'offenceCategory', 'offenceSubcategory']
person_feature_keys = ['gender', 'age', 'occupation', 'given']
outcome_key = 'verdictCategory'

def get_tag_value(tag):
    # 通用取值函数:优先取value属性,没有则取内部文本,自动去首尾空白
    if not tag:
        return None
    val = tag.get("value")
    if val is not None:
        return val
    return tag.get_text(strip=True)

def get_person_features(trial_account, person_type: str):
    person_features = {}
    # 先收集所有人物和职业的关联关系
    occu_joins = trial_account.find_all("join", result="persNameOccupation")
    occu_map = {}
    for join in occu_joins:
        person_id, occu_id = join.get("targets", "").split()
        occu_map[occu_id] = person_id
    
    for key in person_feature_keys:
        matches = []
        all_tags = trial_account.find_all(type=key)
        if key != "occupation":
            # 普通属性:在persName下,匹配人物类型
            matches = [x for x in all_tags if person_type in x.parent.attrs.get("type", "")]
        else:
            # 职业属性:通过join关联匹配对应人物类型
            for tag in all_tags:
                tag_id = tag.get("id")
                linked_person_id = occu_map.get(tag_id)
                if linked_person_id:
                    person_tag = trial_account.find(id=linked_person_id)
                    if person_tag and person_type in person_tag.attrs.get("type", ""):
                        matches.append(tag)
        if matches:
            person_features[person_type + "_" + key] = matches[0]
    return person_features

def process_trial_account(trial_account) -> dict:
    """
    Takes in a single account and returns a dictionary representing a row of the table.
    """
    case_features = {key: trial_account.find(type=key) for key in case_feature_keys}
    defendant_features = get_person_features(trial_account, 'defendant')
    victim_features = get_person_features(trial_account, 'victim')
    outcome = trial_account.find(type=outcome_key)

    features = {**case_features, **defendant_features, **victim_features, "outcome": outcome or {}}
    # 用通用函数取值
    return {key: get_tag_value(value) for key, value in features.items()}

调整说明

  • 新增了get_tag_value通用取值函数,自动适配属性存储和文本存储两种格式,不需要单独给职业字段写特殊判断
  • 补充了职业字段的关联匹配逻辑,通过XML里的join标签匹配职业对应的人物身份,避免原来直接判断父节点属性匹配不到职业标签的问题

内容的提问来源于stack exchange,提问作者Yaz

相关产品推荐
方舟 Agent Plan

超全模态模型 × Harness 升级,最新支持 Deepseek-V4.1-Flash、GLM-5.3 系列、Doubao-Seedream-5.0-pro、Kimi-K3 (部分), 限时 9.9 元起

最近更新时间:2026.10.20 16:45:04