from __future__ import annotations

import re

ZULIP_MAX_MESSAGE_LENGTH = 10000


def truncate_zulip_content(content: str, max_length: int = ZULIP_MAX_MESSAGE_LENGTH) -> str:
    if len(content) <= max_length:
        return content
    suffix = "\n\n... *(content truncated)*"
    return content[: max_length - len(suffix)] + suffix


def clean_mentions(content: str, bot_name: str | None = None) -> str:
    result = content
    if bot_name:
        result = re.sub(rf"@(?:(?:\*\*)|(__)){re.escape(bot_name)}(?:(?:\*\*)|(__))", "", result)
    return result.strip()


def safe_split_zulip_content(text: str, chunk_limit: int = ZULIP_MAX_MESSAGE_LENGTH) -> list[str]:
    if len(text) <= chunk_limit:
        return [text]

    chunks: list[str] = []
    remaining = text
    while len(remaining) > chunk_limit:
        split_at = _find_paragraph_break(remaining, chunk_limit)
        chunks.append(remaining[:split_at].strip())
        remaining = remaining[split_at:].strip()
    if remaining:
        chunks.append(remaining)
    return chunks


def sanitize_zulip_text(text: str) -> str:
    return text


def zulip_markdown_to_native(md_text: str) -> str:
    return md_text


def zulip_native_to_markdown(native_content: str) -> str:
    return native_content


def _find_paragraph_break(text: str, limit: int) -> int:
    search_text = text[:limit]
    for sep in ["\n\n", "\n"]:
        idx = search_text.rfind(sep)
        if idx > limit * 0.3:
            return idx + len(sep)

    last_space = search_text.rfind(" ")
    if last_space > limit * 0.7:
        return last_space + 1

    return limit