FinRAG-MCP

from pathlib import Path import pandas as pd from pypdf import PdfReader from docx import Document def read_txt_md(path: Path) -> str: return path.read_text(encoding="utf-8", errors="ignore") def read_pdf(path: Path) -> str: text = [] reader = PdfReader(str(path)) for page in reader.pages: text.append(page.extract_text() or "") return "\n".join(text) def read_docx(path: Path) -> str: doc = Document(str(path)) return "\n".join(p.text for p in doc.paragraphs) def read_csv(path: Path, max_rows: int = 100) -> str: df = pd.read_csv(path).head(max_rows) # compact textual form lines = [", ".join(map(str, df.columns.tolist()))] for _, row in df.iterrows(): lines.append(", ".join(map(lambda x: str(x), row.tolist()))) return "\n".join(lines) def load_file(path: Path) -> str: suf = path.suffix.lower() if suf in [".txt", ".md"]: return read_txt_md(path) if suf == ".pdf": return read_pdf(path) if suf in [".docx", ".doc"]: return read_docx(path) if suf == ".csv": return read_csv(path) # fallback: try text return read_txt_md(path)