@bill_nee/dwg-reader 1.3.0 → 1.4.0

package/lib/index.js

@@ -8,6 +8,7 @@ const MAX_DWG_BYTES = 512 * 1024 * 1024; // 512 MB 上限，超大图纸会先
 const MAX_ENTITY_DETAIL = 2000;
 const PYTHON_EXE = "D:\\dsh-home\\dsh-runtimes\\dsh-primary-runtime\\dependencies\\python\\python.exe";
 const DXF_SCRIPT = fileURLToPath(new URL("../scripts/read_dxf.py", import.meta.url));
+const OCR_SCRIPT = fileURLToPath(new URL("../scripts/ocr_image.py", import.meta.url));
 const MAX_PDF_BYTES = 100 * 1024 * 1024; // 100 MB PDF 上限
 
 let libreDwgPromise;
@@ -571,10 +572,166 @@ export function apply(ctx, config) {
   const disposeDwg = ctx.tools.register(buildReadDwgTool(ctx));
   const disposeDxf = ctx.tools.register(buildReadDxfTool(ctx));
   const disposePdf = ctx.tools.register(buildReadPdfTool(ctx));
+  const disposeImage = ctx.tools.register(buildReadImageTool(ctx));
   return () => {
     disposeDwg();
     disposeDxf();
     disposePdf();
+    disposeImage();
+  };
+}
+
+function buildReadImageTool(ctx) {
+  return {
+    name: "read_image",
+    description:
+      "读取图片/截屏文件（尺寸、格式、元数据 + Tesseract OCR 文字提取，无需安装图片查看软件）。" +
+      "支持 PNG/JPEG/WebP/BMP/TIFF；OCR 语言为简体中文+英文。",
+    parameters: {
+      type: "object",
+      additionalProperties: true,
+      properties: {
+        file_path: {
+          type: "string",
+          description: "图片文件路径（绝对路径或相对当前工作目录）。",
+        },
+        ocr: {
+          type: "boolean",
+          description: "是否执行 OCR 文字提取（默认 true）。",
+        },
+        tessdata_prefix: {
+          type: "string",
+          description: "Tesseract tessdata 目录（可选；缺省使用系统默认）。",
+        },
+      },
+      required: ["file_path"],
+    },
+    output: {
+      schema: {
+        type: "object",
+        additionalProperties: true,
+        properties: {
+          path: { type: "string", description: "解析后的文件路径。" },
+          fileName: { type: "string", description: "文件名。" },
+          fileSizeBytes: { type: "integer", description: "文件字节数。" },
+          format: { type: "string", description: "图片格式（如 PNG/JPEG）。" },
+          width: { type: "integer", description: "图片宽度（像素）。" },
+          height: { type: "integer", description: "图片高度（像素）。" },
+          mode: { type: "string", description: "颜色模式（如 RGB/RGBA）。" },
+          ocrText: { type: "string", description: "OCR 提取的文字（可能为空）。" },
+          ocrError: { type: "string", description: "OCR 错误信息；成功时为 null。" },
+          error: { type: "string", description: "错误信息；成功时为 null。" },
+        },
+      },
+      render: (args, value) => {
+        const lines = [];
+        lines.push(`<path>${value.path}</path>`);
+        lines.push(`<type>image</type>`);
+        if (value.error) {
+          lines.push(`<content>读取图片失败：${value.error}</content>`);
+          return [{ type: "text", text: lines.join("\n") }];
+        }
+        const sizeKb = (value.fileSizeBytes / 1024).toFixed(1);
+        lines.push(`<content>已解析图片：${value.fileName}（${sizeKb} KB）`);
+        lines.push(`格式：${value.format}；尺寸：${value.width}×${value.height}；模式：${value.mode}`);
+        if (value.ocrError) lines.push(`OCR 错误：${value.ocrError}`);
+        if (value.ocrText) {
+          const preview = value.ocrText.slice(0, 300);
+          lines.push(`OCR 文字（${value.ocrText.length} 字符）：${preview}${value.ocrText.length > 300 ? "…" : ""}`);
+        } else if (!value.ocrError) {
+          lines.push("OCR 文字：无（未检测到可识别文字或 OCR 未启用）");
+        }
+        lines.push(`</content>`);
+        return [{ type: "text", text: lines.join("\n") }];
+      },
+      presentationMeta: () => null,
+    },
+    isConcurrencySafe: () => false,
+    presentCall: (args) => ({
+      card: "generic",
+      title: `读取图片 ${args.file_path || "(未指定路径)"}`,
+      kind: "read",
+      locations: [{ path: args.file_path || "" }],
+    }),
+    presentResult: (args, result) => {
+      if (result && result.isError) return undefined;
+      return {
+        card: "generic",
+        title: `图片解析完成：${args.file_path || ""}`,
+        kind: "read",
+        locations: [{ path: args.file_path || "" }],
+      };
+    },
+    async execute(args, exec) {
+      if (typeof args !== "object" || args === null) {
+        throw new Error('invalid arguments: expected an object with a "file_path" string');
+      }
+      if (typeof args.file_path !== "string" || !args.file_path.trim()) {
+        throw new Error('invalid arguments: "file_path" must be a non-empty string');
+      }
+      if (args.ocr !== undefined && typeof args.ocr !== "boolean") {
+        throw new Error('invalid arguments: "ocr" must be a boolean');
+      }
+      if (args.tessdata_prefix !== undefined && typeof args.tessdata_prefix !== "string") {
+        throw new Error('invalid arguments: "tessdata_prefix" must be a string');
+      }
+
+      const resolveOptions = await sessionResolveOptions(ctx, exec);
+      const target = await ctx.fs.resolve(args.file_path.trim(), resolveOptions);
+      const info = await ctx.fs.stat(target, exec.signal);
+      const displayPath = ctx.fs.processPath(target);
+      if (!info) {
+        ctx.emit("fs/observed", target, { kind: "absent" }, exec);
+        throw new Error(`cannot read "${displayPath}": not found`);
+      }
+      if (info.type !== "file") {
+        throw new Error(`cannot read "${displayPath}": not a regular file`);
+      }
+
+      const pythonArgs = [
+        OCR_SCRIPT,
+        args.file_path.trim(),
+        "D:\\Tools\\Tesseract-OCR\\tesseract.exe",
+      ];
+
+      const result = await new Promise((resolve, reject) => {
+        const proc = spawn(PYTHON_EXE, pythonArgs, { stdio: ["pipe", "pipe", "pipe"] });
+        let stdout = "";
+        let stderr = "";
+        proc.stdout.on("data", (d) => (stdout += d.toString()));
+        proc.stderr.on("data", (d) => (stderr += d.toString()));
+        proc.on("close", (code) => {
+          if (code !== 0) {
+            reject(new Error(`Python 进程退出码 ${code}: ${stderr || stdout}`));
+            return;
+          }
+          try {
+            resolve(JSON.parse(stdout));
+          } catch (err) {
+            reject(new Error(`OCR 结果解析失败: ${err.message}; stdout=${stdout.slice(0, 200)}`));
+          }
+        });
+        proc.on("error", (err) => reject(new Error(`无法启动 Python 进程: ${err.message}`)));
+      });
+
+      if (result.error) {
+        throw new Error(`cannot read "${displayPath}": ${result.error}`);
+      }
+
+      ctx.emit("fs/observed", target, { kind: "present", version: info.version }, exec);
+
+      return toJson({
+        path: displayPath,
+        fileName: displayPath.split(/[\\/]/).pop(),
+        fileSizeBytes: result.fileSizeBytes ?? info.size !== undefined ? info.size : 0,
+        format: result.format ?? "",
+        width: result.width ?? null,
+        height: result.height ?? null,
+        mode: result.mode ?? null,
+        ocrText: result.ocrText ?? "",
+        ocrError: result.ocrError ?? null,
+      });
+    },
   };
 }
 

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@bill_nee/dwg-reader",
-  "version": "1.3.0",
-  "description": "直接读取 AutoCAD DWG/DXF 与 PDF 文件的 DSH 插件（DWG 基于 libredwg-web 0.8.2 WASM，DXF 基于 Python ezdxf 桥接，PDF 基于 pdf-parse，GPL-3.0）。",
+  "version": "1.4.0",
+  "description": "直接读取 AutoCAD DWG/DXF、PDF 与图片/截屏的 DSH 插件（DWG 基于 libredwg-web 0.8.2 WASM，DXF 基于 Python ezdxf 桥接，PDF 基于 pdf-parse，图片基于 Pillow+Tesseract OCR，GPL-3.0）。",
   "license": "GPL-3.0-only",
   "author": "MEB 工程造价项目助手 <meb@example.com>",
   "repository": {

package/scripts/ocr_image.py

@@ -0,0 +1,70 @@
+#!/usr/bin/env python3
+"""对图片进行 OCR 文字提取（供 DSH dwg-reader 插件调用，直接调用 tesseract 命令行）。"""
+import sys
+import json
+import os
+import subprocess
+import tempfile
+
+IMAGE_PATH = sys.argv[1] if len(sys.argv) > 1 else None
+TESSERACT_CMD = sys.argv[2] if len(sys.argv) > 2 else r"D:\Tools\Tesseract-OCR\tesseract.exe"
+
+
+def main():
+    if not IMAGE_PATH or not os.path.exists(IMAGE_PATH):
+        print(json.dumps({"error": f"文件不存在: {IMAGE_PATH}"}))
+        sys.exit(1)
+
+    try:
+        from PIL import Image
+    except ImportError as exc:
+        print(json.dumps({"error": f"Pillow 未安装: {exc}"}))
+        sys.exit(1)
+
+    try:
+        image = Image.open(IMAGE_PATH)
+    except Exception as exc:
+        print(json.dumps({"error": f"无法打开图片: {exc}"}))
+        sys.exit(1)
+
+    width, height = image.size
+    fmt = image.format
+    mode = image.mode
+
+    # 调用 tesseract 命令行做 OCR
+    ocr_text = ""
+    ocr_error = None
+    try:
+        with tempfile.TemporaryDirectory() as tmpdir:
+            out_base = os.path.join(tmpdir, "ocr_out")
+            cmd = [TESSERACT_CMD, IMAGE_PATH, out_base, "-l", "chi_sim+eng"]
+            proc = subprocess.run(cmd, capture_output=True, text=True, timeout=120)
+            if proc.returncode != 0:
+                ocr_error = proc.stderr.strip() or f"tesseract 退出码 {proc.returncode}"
+            else:
+                out_txt = out_base + ".txt"
+                if os.path.exists(out_txt):
+                    with open(out_txt, "r", encoding="utf-8", errors="replace") as f:
+                        ocr_text = f.read().strip()
+    except Exception as exc:
+        ocr_error = str(exc)
+
+    result = {
+        "path": os.path.abspath(IMAGE_PATH),
+        "fileName": os.path.basename(IMAGE_PATH),
+        "fileSizeBytes": os.path.getsize(IMAGE_PATH),
+        "format": fmt,
+        "width": width,
+        "height": height,
+        "mode": mode,
+        "ocrText": ocr_text,
+        "ocrError": ocr_error,
+    }
+    # Windows 控制台默认 GBK，改为 UTF-8 输出到 stdout.buffer 避免编码错误
+    sys.stdout.buffer.write(json.dumps(result, ensure_ascii=False).encode("utf-8"))
+    sys.stdout.buffer.write(b"\n")
+    sys.stdout.buffer.flush()
+
+
+if __name__ == "__main__":
+    main()