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

from __future__ import annotations

import re


def markdown_to_line(md_text: str) -> str:
    text = md_text

    text = re.sub(r"\*\*(.+?)\*\*", r"【\1】", text)
    text = re.sub(r"__([^_]+)__", r"【\1】", text)

    text = re.sub(r"\*([^*\n]+)\*", r"_\1_", text)
    text = re.sub(r"(?<!\w)_([^_\n]+)_(?!\w)", r"_\1_", text)

    text = re.sub(r"`([^`\n]+)`", r"[ \1 ]", text)

    text = re.sub(r"~~(.+?)~~", r"", text)

    text = re.sub(r"!\[[^\]]*\]\([^)]*\)", "", text)

    text = re.sub(r"\[([^\]]+)\]\(([^)]+)\)", r"\1\n(\2)", text)

    text = re.sub(r"^#{1,6}\s+(.+)$", r"■ \1", text, flags=re.MULTILINE)

    text = re.sub(r"^>\s+(.+)$", r"┃ \1", text, flags=re.MULTILINE)

    text = re.sub(r"^-{3,}$", "────────────", text, flags=re.MULTILINE)

    text = re.sub(r"^[*-]\s+(.+)$", r"• \1", text, flags=re.MULTILINE)

    return text.strip()


def extract_code_blocks(text: str) -> tuple[str, list[dict]]:
    code_blocks: list[dict] = []

    def _extract(match: re.Match) -> str:
        language = match.group(1) or ""
        code = match.group(2)
        code_blocks.append({"language": language, "code": code[:2000]})
        return ""

    cleaned = re.sub(r"```(\w*)\n(.*?)```", _extract, text, flags=re.DOTALL)
    return cleaned, code_blocks


def extract_flex_from_markdown(text: str) -> tuple[str, list[dict]]:
    from yuxi.channel.extensions.line.flex_templates import create_code_block_card

    text, code_blocks = extract_code_blocks(text)
    text, table_flex_messages = convert_table_to_flex(text)

    flex_messages: list[dict] = list(table_flex_messages)

    for cb in code_blocks:
        flex_messages.append(create_code_block_card(cb["code"], cb["language"]))

    return text, flex_messages


def convert_table_to_flex(text: str) -> tuple[str, list[dict]]:
    flex_messages: list[dict] = []
    lines = text.split("\n")
    result: list[str] = []
    in_table = False
    headers: list[str] = []
    rows: list[list[str]] = []

    for line in lines:
        stripped = line.strip()
        if stripped.startswith("|") and stripped.endswith("|"):
            cells = [c.strip() for c in stripped[1:-1].split("|")]
            if not in_table:
                headers = cells
                in_table = True
                continue
            if all(re.match(r"^[-:]+$", c) for c in cells):
                continue
            rows.append(cells)
        else:
            if in_table:
                if len(headers) == 2:
                    flex_messages.append(_build_receipt_card(headers, rows))
                elif len(headers) >= 3:
                    flex_messages.append(_build_flex_table(headers, rows[:10]))
                headers = []
                rows = []
                in_table = False
            result.append(line)

    if in_table:
        if len(headers) == 2:
            flex_messages.append(_build_receipt_card(headers, rows))
        elif len(headers) >= 3:
            flex_messages.append(_build_flex_table(headers, rows[:10]))

    return "\n".join(result), flex_messages


def _build_receipt_card(headers: list[str], rows: list[list[str]]) -> dict:
    body_contents = []
    for i, row in enumerate(rows):
        label = row[0] if len(row) > 0 else ""
        value = row[1] if len(row) > 1 else ""
        bg_color = "#FFFFFF" if i % 2 == 0 else "#FAFAFA"
        body_contents.append({
            "type": "box",
            "layout": "horizontal",
            "backgroundColor": bg_color,
            "contents": [
                {"type": "text", "text": label, "flex": 1, "wrap": True},
                {"type": "text", "text": value, "flex": 1, "wrap": True, "align": "end"},
            ],
        })

    return {
        "type": "flex",
        "altText": headers[0] if headers else "Receipt",
        "contents": {
            "type": "bubble",
            "body": {
                "type": "box",
                "layout": "vertical",
                "contents": body_contents,
            },
        },
    }


def _build_flex_table(headers: list[str], rows: list[list[str]]) -> dict:
    header_items = []
    for h in headers:
        header_items.append({
            "type": "text",
            "text": h,
            "weight": "bold",
            "flex": 1,
            "wrap": True,
            "size": "sm",
        })

    body_contents = [
        {
            "type": "box",
            "layout": "horizontal",
            "contents": header_items,
        },
        {"type": "separator"},
    ]

    for row in rows:
        row_items = []
        for i, h in enumerate(headers):
            value = row[i] if i < len(row) else ""
            row_items.append({
                "type": "text",
                "text": value,
                "flex": 1,
                "wrap": True,
                "size": "sm",
            })
        body_contents.append({
            "type": "box",
            "layout": "horizontal",
            "contents": row_items,
        })
        body_contents.append({"type": "separator"})

    return {
        "type": "flex",
        "altText": f"Table: {headers[0]}" if headers else "Table",
        "contents": {
            "type": "bubble",
            "body": {
                "type": "box",
                "layout": "vertical",
                "contents": body_contents[: len(body_contents) - 1] if body_contents else [],
            },
        },
    }


def chunk_text(text: str, limit: int = 5000) -> list[str]:
    if len(text) <= limit:
        return [text]

    chunks: list[str] = []
    while len(text) > limit:
        split_point = text.rfind("\n", 0, limit)
        if split_point == -1 or split_point < limit // 2:
            split_point = text.rfind(". ", 0, limit)
        if split_point == -1 or split_point < limit // 2:
            split_point = text.rfind(" ", 0, limit)
        if split_point == -1 or split_point < limit // 2:
            split_point = limit

        chunks.append(text[:split_point].strip())
        text = text[split_point:].strip()

    if text:
        chunks.append(text)
    return chunks