Download 4-HSLS/parse_recordfile.py from yijiexu/LLMValues: direct link, hf CLI and curl.
- Browser
- Download file 5.22 kB
-
https://huggingface.co/datasets/yijiexu/LLMValues/resolve/main/4-HSLS/parse_recordfile.py
- Command line
-
hf download hf://datasets/yijiexu/LLMValues/4-HSLS/parse_recordfile.py
-
curl -L -o parse_recordfile.py https://huggingface.co/datasets/yijiexu/LLMValues/resolve/main/4-HSLS/parse_recordfile.py
5.22 kB
| import json | |
| import re | |
| import os | |
| def parse_codebook(codebook_file): | |
| """ | |
| 解析 codebook.jsonl 文件。 | |
| Args: | |
| codebook_file (str): codebook.jsonl 文件的路径。 | |
| Returns: | |
| dict: 一个字典,键是变量名(name),值是对应的元数据。 | |
| """ | |
| codebook_data = {} | |
| with open(codebook_file, 'r', encoding='utf-8') as f: | |
| for line in f: | |
| try: | |
| data = json.loads(line) | |
| # 确保 'name' 键存在 | |
| if 'name' in data: | |
| codebook_data[data['name']] = data | |
| except json.JSONDecodeError: | |
| print(f"警告: 无法解析行: {line}") | |
| return codebook_data | |
| def parse_variable_labels(layout_file, | |
| start_marker="/* Variable Value Labels */"): | |
| """ | |
| 解析包含变量值标签的文本文件。 | |
| Args: | |
| layout_file (str): RecordFileLayout... .txt 文件的路径。 | |
| start_marker (str): 开始解析的标记。 | |
| Returns: | |
| dict: 一个字典,键是变量名,值是响应选项的列表。 | |
| """ | |
| value_labels = {} | |
| current_variable = None | |
| parsing_started = False | |
| with open(layout_file, 'r', encoding='utf-8', errors='ignore') as f: | |
| for line in f: | |
| # 只有在找到起始标记后才开始解析 | |
| if not parsing_started: | |
| if start_marker in line: | |
| parsing_started = True | |
| continue | |
| # 去除行首行尾的空白字符 | |
| line = line.strip() | |
| # 如果行为空,则跳过 | |
| if not line: | |
| continue | |
| # 如果行不是以数字或'-'开头,则假定它是一个新的变量名 | |
| if not (line[0].isdigit() or line.startswith('-')): | |
| current_variable = line | |
| if current_variable not in value_labels: | |
| value_labels[current_variable] = [] | |
| # 否则,解析值和标签 | |
| elif current_variable: | |
| # 使用正则表达式匹配 "code = 'text'" 格式 | |
| match = re.match(r'^(-?\d+\.?\d*)\s*=\s*"(.*)"', line) | |
| if match: | |
| code_str, text = match.groups() | |
| # 将code转换为整数或浮点数 | |
| try: | |
| code = int(code_str) | |
| except ValueError: | |
| code = float(code_str) | |
| value_labels[current_variable].append({ | |
| "code": code, | |
| "text": text | |
| }) | |
| return value_labels | |
| def main(): | |
| """ | |
| 主函数,执行整个数据处理流程。 | |
| """ | |
| # --- 文件路径配置 --- | |
| # 请确保这些文件与脚本在同一目录下,或者提供完整路径 | |
| codebook_filename = "codebook.jsonl" | |
| layout_filename = "RecordFileLayout_HSLS_17_PETS_SR_STUDENT.txt" | |
| output_filename = "record_file.jsonl" | |
| # 检查所需文件是否存在 | |
| if not os.path.exists(codebook_filename): | |
| print(f"错误: 文件 '{codebook_filename}' 未找到。请确保文件存在于当前目录。") | |
| return | |
| if not os.path.exists(layout_filename): | |
| print(f"错误: 文件 '{layout_filename}' 未找到。请确保文件存在于当前目录。") | |
| return | |
| # 步骤 1: 解析 codebook | |
| print(f"正在解析 '{codebook_filename}'...") | |
| codebook = parse_codebook(codebook_filename) | |
| print(f"成功解析 {len(codebook)} 个变量的元数据。") | |
| # 步骤 2: 解析变量标签 | |
| print(f"正在解析 '{layout_filename}'...") | |
| value_labels = parse_variable_labels(layout_filename) | |
| print(f"成功解析 {len(value_labels)} 个变量的响应选项。") | |
| # 步骤 3: 合并数据并生成输出 | |
| print("正在合并数据并生成输出文件...") | |
| final_output = [] | |
| index_counter = 1 | |
| for var_name, var_info in codebook.items(): | |
| # question_text 使用 'label' 字段 | |
| question_text = var_info.get('label', 'N/A') | |
| # response_options 来自解析的 value_labels | |
| # 如果找不到,则留空列表 | |
| response_options = value_labels.get(var_name, []) | |
| # 按照您期望的格式创建条目 | |
| # 注意: 'section' 和 'type' 字段在源文件中不可用,因此在此省略。 | |
| # 如果需要,可以手动添加或修改此处的逻辑。 | |
| # 'question_id' 使用变量名 `name` | |
| output_item = { | |
| "index": index_counter, | |
| "question_id": var_name, | |
| "question_text": question_text, | |
| "description": var_info.get('description', ''), | |
| "response_options": response_options | |
| } | |
| final_output.append(output_item) | |
| index_counter += 1 | |
| # 步骤 4: 将结果写入输出文件 | |
| with open(output_filename, 'w', encoding='utf-8') as f: | |
| for item in final_output: | |
| f.write(json.dumps(item, ensure_ascii=False) + '\n') | |
| print(f"\n处理完成!结果已保存到 '{output_filename}'。") | |
| print(f"共生成 {len(final_output)} 条记录。") | |
| if __name__ == "__main__": | |
| main() | |