ForcePilot/backend/package/yuxi/channel/extensions/msteams/format.py

from __future__ import annotations

import html
import re

MAX_TEAMS_TEXT_LENGTH = 4000


def decode_html_entities(text: str) -> str:
    return html.unescape(text)


def strip_at_mentions(text: str) -> str:
    return re.sub(r"<at>[^<]*</at>", "", text).strip()


def strip_markdown_links(text: str) -> str:
    return re.sub(r"\[([^\]]*)\]\([^)]*\)", r"\1", text)


def clean_inbound_text(text: str) -> str:
    text = decode_html_entities(text)
    text = strip_at_mentions(text)
    text = text.strip()
    return text


def chunk_text(text: str, limit: int = MAX_TEAMS_TEXT_LENGTH) -> list[str]:
    if len(text) <= limit:
        return [text]

    chunks = []
    remaining = text

    while len(remaining) > limit:
        split_at = remaining.rfind("\n", 0, limit)
        if split_at == -1:
            split_at = remaining.rfind(". ", 0, limit)
        if split_at == -1:
            split_at = remaining.rfind(" ", 0, limit)
        if split_at == -1:
            split_at = limit

        chunks.append(remaining[:split_at].strip())
        remaining = remaining[split_at:].strip()

    if remaining:
        chunks.append(remaining.strip())

    return chunks


def escape_teams_markdown(text: str) -> str:
    return text.replace("*", r"\*").replace("_", r"\_").replace("~", r"\~").replace("#", r"\#")


def render_mention(name: str, user_id: str) -> str:
    return f"<at>{name}</at>"


def extract_media_urls(attachments: list[dict]) -> list[dict]:
    result = []
    for att in attachments:
        content_type = att.get("contentType", "")
        content_url = att.get("contentUrl", "")
        name = att.get("name", "")
        if content_url:
            result.append(
                {
                    "url": content_url,
                    "content_type": content_type,
                    "name": name,
                }
            )
    return result