scripts/parsing.py

"""Parse uploaded course material into plain text.

Supports PDF, PPTX, DOCX, and plain text files.
"""

import io


def parse_pdf(data):
    """Extract text from PDF bytes using PyMuPDF."""
    import fitz  # PyMuPDF

    text_parts = []
    with fitz.open(stream=data, filetype="pdf") as document:
        for page in document:
            text_parts.append(page.get_text())
    return "\n".join(text_parts)


def parse_pptx(data):
    """Extract text from PPTX bytes, slide by slide."""
    from pptx import Presentation

    presentation = Presentation(io.BytesIO(data))
    text_parts = []
    for index, slide in enumerate(presentation.slides, start=1):
        text_parts.append(f"[Slide {index}]")
        for shape in slide.shapes:
            if shape.has_text_frame:
                for paragraph in shape.text_frame.paragraphs:
                    line = "".join(run.text for run in paragraph.runs)
                    if line.strip():
                        text_parts.append(line)
    return "\n".join(text_parts)


def parse_docx(data):
    """Extract text from DOCX bytes."""
    from docx import Document

    document = Document(io.BytesIO(data))
    return "\n".join(p.text for p in document.paragraphs if p.text.strip())


def parse_material(filename, data):
    """Pick the right parser based on file extension and return plain text."""
    lower = filename.lower()
    if lower.endswith(".pdf"):
        return parse_pdf(data)
    if lower.endswith(".pptx"):
        return parse_pptx(data)
    if lower.endswith(".docx"):
        return parse_docx(data)
    if lower.endswith(".txt") or lower.endswith(".md"):
        return data.decode("utf-8", errors="ignore")
    raise ValueError(f"Unsupported file type: {filename}")


def clean_text(text):
    """Strip extra whitespace and blank lines."""
    lines = [line.strip() for line in text.splitlines()]
    non_empty = [line for line in lines if line]
    return "\n".join(non_empty)