# ============================================================
# Project: EDGAR Research Database
# Description: SEC EDGAR Research Infrastructure
#
# Authors:
# - Manel Kammoun — manel.kammoun@uqo.ca
# - Djerrycharli Tandjambianda — djerrycharli.tandjambianda@uqo.ca
# - Simon-Pierre Boucher — simon-pierre.boucher@uqo.ca
#
# Institution:
# Université du Québec en Outaouais (UQO)
#
# License:
# Research / Academic Use
# ============================================================

"""Word-window chunker (re-created 2026-10-27; the original module was lost).

Produces the same record layout as the existing ``filing_chunks`` table:
``chunk_id`` (uuid4), ``section_id``, ``chunk_text``, ``chunk_order``,
``word_count``, ``char_count``, ``start_char``, ``end_char``.
Chunks are ``chunk_size`` words long with ``overlap`` words shared between
consecutive chunks; a trailing chunk shorter than ``min_chunk_size`` words is
merged into the previous one.
"""

from __future__ import annotations

import uuid


class TextChunker:
    def __init__(
        self,
        chunk_size: int = 512,
        overlap: int = 64,
        min_chunk_size: int = 50,
    ) -> None:
        if overlap >= chunk_size:
            raise ValueError("overlap must be smaller than chunk_size")
        self.chunk_size = chunk_size
        self.overlap = overlap
        self.min_chunk_size = min_chunk_size

    def chunk_text(self, text: str, section_id: str | None = None) -> list[dict]:
        words = text.split()
        if not words:
            return []

        step = self.chunk_size - self.overlap
        windows: list[list[str]] = []
        start = 0
        while start < len(words):
            window = words[start:start + self.chunk_size]
            if windows and len(window) < self.min_chunk_size:
                # fold a tiny tail into the previous window
                tail = window[self.overlap:] if len(window) > self.overlap else []
                windows[-1].extend(tail)
                break
            windows.append(window)
            if start + self.chunk_size >= len(words):
                break
            start += step

        chunks: list[dict] = []
        cursor = 0
        for order, window in enumerate(windows):
            chunk_text = " ".join(window)
            chunks.append({
                "chunk_id": str(uuid.uuid4()),
                "section_id": section_id,
                "chunk_text": chunk_text,
                "chunk_order": order,
                "word_count": len(window),
                "char_count": len(chunk_text),
                "start_char": cursor,
                "end_char": cursor + len(chunk_text),
            })
            cursor += len(chunk_text) + 1
        return chunks

    def chunk_section(self, section: dict) -> list[dict]:
        text = section.get("text") or ""
        return self.chunk_text(text, section_id=section.get("section_id"))