# -*- coding: utf-8 -*- """ 文档解析模块 支持PDF、Word、Excel等金融文档解析 """ import json import re import os from pathlib import Path from typing import Dict, Any, List, Optional # 文档解析依赖检测 PDF_AVAILABLE = False DOCX_AVAILABLE = False XLSX_AVAILABLE = False try: import pypdf PDF_AVAILABLE = True except ImportError: pass try: import PyPDF2 PDF_AVAILABLE = True except ImportError: pass try: from docx import Document as DocxDocument DOCX_AVAILABLE = True except ImportError: pass try: import openpyxl XLSX_AVAILABLE = True except ImportError: pass SKILL_DATA_DIR = Path(__file__).parent.parent / "data" def parse_pdf(file_path: str) -> Dict[str, Any]: """ 解析PDF文档 Args: file_path: PDF文件路径 Returns: 解析结果 """ result = { "file_type": "pdf", "file_path": file_path, "pages": 0, "text_content": "", "tables": [], "financial_data": {} } if not PDF_AVAILABLE: result["error"] = "PyPDF2未安装，请运行: pip install PyPDF2" return result try: with open(file_path, 'rb') as f: reader = PyPDF2.PdfReader(f) result["pages"] = len(reader.pages) # 提取文本 full_text = [] for page in reader.pages: text = page.extract_text() if text: full_text.append(text) result["text_content"] = "\n\n".join(full_text) # 尝试提取财务数据 result["financial_data"] = extract_financial_numbers(result["text_content"]) except Exception as e: result["error"] = f"PDF解析失败: {e}" return result def parse_docx(file_path: str) -> Dict[str, Any]: """ 解析Word文档 Args: file_path: DOCX文件路径 Returns: 解析结果 """ result = { "file_type": "docx", "file_path": file_path, "paragraphs": [], "tables": [], "text_content": "", "financial_data": {} } if not DOCX_AVAILABLE: result["error"] = "python-docx未安装，请运行: pip install python-docx" return result try: doc = DocxDocument(file_path) # 提取段落 paragraphs = [] for para in doc.paragraphs: if para.text.strip(): paragraphs.append(para.text) result["paragraphs"] = paragraphs result["text_content"] = "\n\n".join(paragraphs) # 提取表格 tables = [] for i, table in enumerate(doc.tables): table_data = [] for row in table.rows: row_data = [cell.text.strip() for cell in row.cells] table_data.append(row_data) tables.append(table_data) result["tables"] = tables # 财务数据 result["financial_data"] = extract_financial_numbers(result["text_content"]) except Exception as e: result["error"] = f"DOCX解析失败: {e}" return result def parse_xlsx(file_path: str, sheet_name: Optional[str] = None) -> Dict[str, Any]: """ 解析Excel文档 Args: file_path: Excel文件路径 sheet_name: 指定工作表名称（默认第一个） Returns: 解析结果 """ result = { "file_type": "xlsx", "file_path": file_path, "sheets": {}, "financial_data": {} } if not XLSX_AVAILABLE: result["error"] = "openpyxl未安装，请运行: pip install openpyxl" return result try: wb = openpyxl.load_workbook(file_path, data_only=True) result["sheet_names"] = wb.sheetnames # 读取指定sheet或全部 sheets_to_read = [sheet_name] if sheet_name else wb.sheetnames for sheet in sheets_to_read: ws = wb[sheet] data = [] for row in ws.iter_rows(values_only=True): row_data = [str(cell) if cell is not None else "" for cell in row] if any(row_data): data.append(row_data) result["sheets"][sheet] = { "rows": len(data), "data": data[:100] # 限制前100行 } # 财务数据 all_text = json.dumps(result["sheets"]) result["financial_data"] = extract_financial_numbers(all_text) except Exception as e: result["error"] = f"Excel解析失败: {e}" return result def parse_txt(file_path: str) -> Dict[str, Any]: """ 解析文本文件 Args: file_path: 文本文件路径 Returns: 解析结果 """ result = { "file_type": "txt", "file_path": file_path, "text_content": "", "financial_data": {} } try: with open(file_path, 'r', encoding='utf-8') as f: result["text_content"] = f.read() result["financial_data"] = extract_financial_numbers(result["text_content"]) except Exception as e: try: # 尝试GBK编码 with open(file_path, 'r', encoding='gbk') as f: result["text_content"] = f.read() result["financial_data"] = extract_financial_numbers(result["text_content"]) except: result["error"] = f"文本解析失败: {e}" return result def extract_financial_numbers(text: str) -> Dict[str, Any]: """ 从文本中提取财务数字 Args: text: 文本内容 Returns: 财务数据字典 """ data = {} # 提取金额（亿/万） amount_patterns = [ (r'([+-]?\d+\.?\d*)\s*亿元', 'amount_yi'), (r'([+-]?\d+\.?\d*)\s*万元', 'amount_wan'), (r'([+-]?\d+\.?\d*)\s*%，', 'percentage'), ] amounts = [] percentages = [] for pattern, ptype in amount_patterns: matches = re.findall(pattern, text) for match in matches: try: value = float(match) if ptype == 'amount_yi': amounts.append(value) elif ptype == 'percentage': percentages.append(value) except: pass if amounts: data["amounts_yi"] = amounts[:10] # 取前10个 data["total_yi"] = sum(amounts) if percentages: data["percentages"] = percentages[:20] # 提取常见财务指标 indicators = { '净利润': r'净利润[：:]\s*([+-]?\d+\.?\d*)', '营业收入': r'营业收入[：:]\s*([+-]?\d+\.?\d*)', '总资产': r'总资产[：:]\s*([+-]?\d+\.?\d*)', '净资产': r'净资产[：:]\s*([+-]?\d+\.?\d*)', '每股收益': r'每股收益[（(]EPS[）)][：:]\s*([+-]?\d+\.?\d*)', '市盈率': r'市盈率[（(]PE[）)][：:]\s*([+-]?\d+\.?\d*)', '市净率': r'市净率[（(]PB[）)][：:]\s*([+-]?\d+\.?\d*)', } for name, pattern in indicators.items(): match = re.search(pattern, text) if match: try: data[name] = float(match.group(1)) except: pass return data def parse_document(file_path: str, sheet_name: Optional[str] = None) -> Dict[str, Any]: """ 通用文档解析接口 Args: file_path: 文件路径 sheet_name: Excel指定工作表 Returns: 解析结果 """ path = Path(file_path) suffix = path.suffix.lower() parsers = { '.pdf': parse_pdf, '.docx': parse_docx, '.doc': parse_docx, # 尝试作为docx '.xlsx': parse_xlsx, '.xls': parse_xlsx, '.txt': parse_txt, '.md': parse_txt, '.csv': parse_txt, } parser = parsers.get(suffix) if not parser: return {"error": f"不支持的文件类型: {suffix}"} if suffix == '.xlsx' or suffix == '.xls': return parser(file_path, sheet_name) else: return parser(file_path) def extract_key_info(parsed_doc: Dict[str, Any]) -> str: """ 从解析结果中提取关键信息摘要 Args: parsed_doc: parse_document的返回结果 Returns: 关键信息摘要文本 """ if "error" in parsed_doc: return f"解析错误: {parsed_doc['error']}" summary = f"【文档解析结果】\n\n" summary += f"文件类型: {parsed_doc.get('file_type', 'unknown').upper()}\n" if "pages" in parsed_doc: summary += f"页数: {parsed_doc['pages']}\n" if "paragraphs" in parsed_doc: summary += f"段落数: {len(parsed_doc['paragraphs'])}\n" if "tables" in parsed_doc: summary += f"表格数: {len(parsed_doc['tables'])}\n" if "sheets" in parsed_doc: summary += f"工作表: {', '.join(parsed_doc.get('sheet_names', []))}\n" # 财务数据摘要 fin_data = parsed_doc.get("financial_data", {}) if fin_data: summary += "\n【提取的财务数据】\n" for key, value in fin_data.items(): if key == "percentages": summary += f"百分比数据: {value[:5]}...\n" if len(value) > 5 else f"百分比数据: {value}\n" elif key == "amounts_yi": summary += f"金额数据(亿元): {value[:5]}...\n" if len(value) > 5 else f"金额数据(亿元): {value}\n" elif key == "total_yi": summary += f"总金额: {value:.2f}亿元\n" else: summary += f"{key}: {value}\n" # 文本预览 text = parsed_doc.get("text_content", "") if text: preview = text[:500].replace('\n', ' ') summary += f"\n【内容预览】\n{preview}..." return summary if __name__ == "__main__": import sys if len(sys.argv) < 2: print("用法: python document_parser.py <文件路径>") print("示例: python document_parser.py ./report.pdf") print("\n支持格式: PDF, DOCX, XLSX, TXT, MD, CSV") sys.exit(1) file_path = sys.argv[1] if not os.path.exists(file_path): print(f"文件不存在: {file_path}") sys.exit(1) print(f"正在解析: {file_path}\n") result = parse_document(file_path) print(extract_key_info(result))