plugins/text/api.py

import logging
import re

from fastapi import APIRouter, Form, HTTPException, UploadFile

logger = logging.getLogger(__name__)

router = APIRouter()

plugin = None


def set_plugin_instance(plugin_instance):
    global plugin
    plugin = plugin_instance


@router.get("/status")
async def get_status():
    if plugin is None:
        return {
            "name": "text",
            "enabled": False,
            "message": "插件未加载",
        }
    return plugin.get_status()


@router.post("/process")
async def process_text(text: str = Form(""), mode: str = Form("standard")):
    if plugin is None or not plugin.enabled:
        raise HTTPException(status_code=400, detail="插件未启用")

    if not text:
        raise HTTPException(status_code=400, detail="文本内容不能为空")

    cleaned_text = clean_text(text, mode)

    return {
        "original_text": text,
        "cleaned_text": cleaned_text,
        "original_char_count": len(text),
        "original_line_count": len(text.split("\n")),
        "cleaned_char_count": len(cleaned_text),
        "cleaned_line_count": len(cleaned_text.split("\n")),
    }


@router.post("/upload")
async def upload_file(file: UploadFile):
    if plugin is None or not plugin.enabled:
        raise HTTPException(status_code=400, detail="插件未启用")

    filename = file.filename.lower()
    if not (filename.endswith(".md") or filename.endswith(".txt")):
        raise HTTPException(status_code=400, detail="只支持 .md 和 .txt 文件")

    content_bytes = await file.read()
    try:
        content = content_bytes.decode("utf-8")
    except UnicodeDecodeError:
        raise HTTPException(status_code=400, detail="文件编码不支持，请使用 UTF-8 编码")

    return {
        "filename": file.filename,
        "content": content,
    }


def clean_text(text: str, mode: str = "standard") -> str:
    if mode == "standard":
        lines = text.split("\n")
        cleaned_lines = []
        prev_line_type = None

        for line in lines:
            stripped = line.strip()
            if stripped:
                # 检查当前行类型
                is_heading = re.match(r"^#+\s", stripped) is not None
                is_numbered = re.match(r"^\d+[\.\)]\s+", stripped) is not None
                is_bulleted = re.match(r"^[\-\*]\s+", stripped) is not None

                current_line_type = (
                    "heading"
                    if is_heading
                    else (
                        "numbered"
                        if is_numbered
                        else ("bulleted" if is_bulleted else "other")
                    )
                )

                # 判断是否需要添加空行
                if prev_line_type is None:
                    # 第一行，不需要空行
                    cleaned_lines.append(line)
                elif prev_line_type == "heading" and current_line_type in [
                    "numbered",
                    "bulleted",
                    "other",
                ]:
                    # 标题后直接跟内容，不需要空行
                    cleaned_lines.append(line)
                elif current_line_type in [
                    "numbered",
                    "bulleted",
                ] and prev_line_type in ["numbered", "bulleted"]:
                    # 列表项之间，不需要空行
                    cleaned_lines.append(line)
                else:
                    # 其他情况，添加空行
                    cleaned_lines.append(line)

                prev_line_type = current_line_type
            else:
                # 空行，跳过
                pass

        result = "\n".join(cleaned_lines)
    elif mode == "compact":
        result = re.sub(r"\r\n+", " ", text)
        result = re.sub(r"\r+", " ", result)
        result = re.sub(r"\n+", " ", result)
        result = re.sub(r"[ \t\u00A0\u200B\u200C\u200D\u2060]+", " ", result)
        result = re.sub(r" +", " ", result)
    else:
        result = text

    result = re.sub(r"[，,]", "，", result)

    result = re.sub(r"(?<=\d)\.", "\x00", result)
    result = re.sub(r"[。.](?![a-zA-Z0-9])", "。", result)
    result = re.sub(r"\x00", ".", result)

    result = re.sub(r"[！!]", "！", result)
    result = re.sub(r"[？?]", "？", result)
    result = re.sub(r"[；;]", "；", result)
    result = re.sub(r"[：:]", "：", result)
    result = re.sub(r"[（(]", "（", result)
    result = re.sub(r"[）)]", "）", result)
    result = re.sub(r"[【\[]", "【", result)
    result = re.sub(r"[】\]]", "】", result)

    return result.strip()