backend/core/parser.py

"""
parser.py
---------
Extracts raw text from PDF files using PyPDF2.
Falls back gracefully if a page has no extractable text.
"""

import io
import PyPDF2


def extract_text_from_pdf(file_bytes: bytes) -> str:
    """
    Given the raw bytes of a PDF, return all extracted text as a single string.
    Each page is separated by a newline for downstream parsing.
    """
    text_parts = []
    try:
        reader = PyPDF2.PdfReader(io.BytesIO(file_bytes))
        for page in reader.pages:
            page_text = page.extract_text()
            if page_text:
                text_parts.append(page_text.strip())
    except Exception as e:
        # If parsing fails entirely, return empty string and let caller handle it
        print(f"[Parser] Error reading PDF: {e}")
    return "\n".join(text_parts)


def extract_text_from_string(text: str) -> str:
    """
    Pass-through for plain text resumes (optional text-paste input).
    Applies light normalization.
    """
    return text.strip()