Merge pull request #24773 from tailsu:sd/pathlike

python: accept path-like objects wherever file names are expected #24773

Merry Christmas, all 🎄

Implements #15731

Support is enabled for all arguments named `filename` or `filepath` (case-insensitive), or annotated with `CV_WRAP_FILE_PATH`.

Support is based on `PyOS_FSPath`, which is available in Python 3.6+. When running on older Python versions the arguments must have a `str` value as before.

### Pull Request Readiness Checklist

See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request

- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch
- [x] There is a reference to the original bug report and related work
- [ ] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
      Patch to opencv_extra has the same branch name.
- [ ] The feature is well documented and sample code can be built with the project CMake

modules/core/include/opencv2/core/cvdef.h

@@ -475,6 +475,8 @@ Cv64suf;
 #define CV_WRAP_MAPPABLE(mappable)
 #define CV_WRAP_PHANTOM(phantom_header)
 #define CV_WRAP_DEFAULT(val)
+/* Indicates that the function parameter has filesystem path semantic */
+#define CV_WRAP_FILE_PATH
 
 /****************************************************************************************\
 *                                  Matrix type (Mat)                                     *

modules/dnn/include/opencv2/dnn/dnn.hpp

@@ -484,7 +484,7 @@ CV__DNN_INLINE_NS_BEGIN
          *  Networks imported from Intel's Model Optimizer are launched in Intel's Inference Engine
          *  backend.
          */
-        CV_WRAP static Net readFromModelOptimizer(const String& xml, const String& bin);
+        CV_WRAP static Net readFromModelOptimizer(CV_WRAP_FILE_PATH const String& xml, CV_WRAP_FILE_PATH const String& bin);
 
         /** @brief Create a network from Intel's Model Optimizer in-memory buffers with intermediate representation (IR).
          *  @param[in] bufferModelConfig buffer with model's configuration.
@@ -517,7 +517,7 @@ CV__DNN_INLINE_NS_BEGIN
          *  @param path   path to output file with .dot extension
          *  @see dump()
          */
-        CV_WRAP void dumpToFile(const String& path);
+        CV_WRAP void dumpToFile(CV_WRAP_FILE_PATH const String& path);
         /** @brief Adds new layer to the net.
          *  @param name   unique name of the adding layer.
          *  @param type   typename of the adding layer (type must be registered in LayerRegister).
@@ -890,7 +890,7 @@ CV__DNN_INLINE_NS_BEGIN
     *  @param darknetModel path to the .weights file with learned network.
     *  @returns Network object that ready to do forward, throw an exception in failure cases.
     */
-    CV_EXPORTS_W Net readNetFromDarknet(const String &cfgFile, const String &darknetModel = String());
+    CV_EXPORTS_W Net readNetFromDarknet(CV_WRAP_FILE_PATH const String &cfgFile, CV_WRAP_FILE_PATH const String &darknetModel = String());
 
     /** @brief Reads a network model stored in <a href="https://pjreddie.com/darknet/">Darknet</a> model files.
      *  @param bufferCfg   A buffer contains a content of .cfg file with text description of the network architecture.
@@ -915,7 +915,7 @@ CV__DNN_INLINE_NS_BEGIN
       * @param caffeModel path to the .caffemodel file with learned network.
       * @returns Net object.
       */
-    CV_EXPORTS_W Net readNetFromCaffe(const String &prototxt, const String &caffeModel = String());
+    CV_EXPORTS_W Net readNetFromCaffe(CV_WRAP_FILE_PATH const String &prototxt, CV_WRAP_FILE_PATH const String &caffeModel = String());
 
     /** @brief Reads a network model stored in Caffe model in memory.
       * @param bufferProto buffer containing the content of the .prototxt file
@@ -944,7 +944,7 @@ CV__DNN_INLINE_NS_BEGIN
       *               let us make it more flexible.
       * @returns Net object.
       */
-    CV_EXPORTS_W Net readNetFromTensorflow(const String &model, const String &config = String());
+    CV_EXPORTS_W Net readNetFromTensorflow(CV_WRAP_FILE_PATH const String &model, CV_WRAP_FILE_PATH const String &config = String());
 
     /** @brief Reads a network model stored in <a href="https://www.tensorflow.org/">TensorFlow</a> framework's format.
       * @param bufferModel buffer containing the content of the pb file
@@ -969,7 +969,7 @@ CV__DNN_INLINE_NS_BEGIN
       * @param model  path to the .tflite file with binary flatbuffers description of the network architecture
       * @returns Net object.
       */
-    CV_EXPORTS_W Net readNetFromTFLite(const String &model);
+    CV_EXPORTS_W Net readNetFromTFLite(CV_WRAP_FILE_PATH const String &model);
 
     /** @brief Reads a network model stored in <a href="https://www.tensorflow.org/lite">TFLite</a> framework's format.
       * @param bufferModel buffer containing the content of the tflite file
@@ -1011,7 +1011,7 @@ CV__DNN_INLINE_NS_BEGIN
      *
      * Also some equivalents of these classes from cunn, cudnn, and fbcunn may be successfully imported.
      */
-     CV_EXPORTS_W Net readNetFromTorch(const String &model, bool isBinary = true, bool evaluate = true);
+     CV_EXPORTS_W Net readNetFromTorch(CV_WRAP_FILE_PATH const String &model, bool isBinary = true, bool evaluate = true);
 
      /**
       * @brief Read deep learning network represented in one of the supported formats.
@@ -1037,7 +1037,7 @@ CV__DNN_INLINE_NS_BEGIN
       * @ref readNetFromTorch or @ref readNetFromDarknet. An order of @p model and @p config
       * arguments does not matter.
       */
-     CV_EXPORTS_W Net readNet(const String& model, const String& config = "", const String& framework = "");
+     CV_EXPORTS_W Net readNet(CV_WRAP_FILE_PATH const String& model, CV_WRAP_FILE_PATH const String& config = "", const String& framework = "");
 
      /**
       * @brief Read deep learning network represented in one of the supported formats.
@@ -1064,7 +1064,7 @@ CV__DNN_INLINE_NS_BEGIN
      *  backend.
      */
     CV_EXPORTS_W
-    Net readNetFromModelOptimizer(const String &xml, const String &bin = "");
+    Net readNetFromModelOptimizer(CV_WRAP_FILE_PATH const String &xml, CV_WRAP_FILE_PATH const String &bin = "");
 
     /** @brief Load a network from Intel's Model Optimizer intermediate representation.
      *  @param[in] bufferModelConfig Buffer contains XML configuration with network's topology.
@@ -1093,7 +1093,7 @@ CV__DNN_INLINE_NS_BEGIN
      *  @param onnxFile path to the .onnx file with text description of the network architecture.
      *  @returns Network object that ready to do forward, throw an exception in failure cases.
      */
-    CV_EXPORTS_W Net readNetFromONNX(const String &onnxFile);
+    CV_EXPORTS_W Net readNetFromONNX(CV_WRAP_FILE_PATH const String &onnxFile);
 
     /** @brief Reads a network model from <a href="https://onnx.ai/">ONNX</a>
      *         in-memory buffer.
@@ -1116,7 +1116,7 @@ CV__DNN_INLINE_NS_BEGIN
      *  @param path to the .pb file with input tensor.
      *  @returns Mat.
      */
-    CV_EXPORTS_W Mat readTensorFromONNX(const String& path);
+    CV_EXPORTS_W Mat readTensorFromONNX(CV_WRAP_FILE_PATH const String& path);
 
     /** @brief Creates 4-dimensional blob from image. Optionally resizes and crops @p image from center,
      *  subtract @p mean values, scales values by @p scalefactor, swap Blue and Red channels.
@@ -1289,7 +1289,7 @@ CV__DNN_INLINE_NS_BEGIN
      *       is taken from NVidia's Caffe fork: https://github.com/NVIDIA/caffe.
      *       So the resulting model may be used there.
      */
-    CV_EXPORTS_W void shrinkCaffeModel(const String& src, const String& dst,
+    CV_EXPORTS_W void shrinkCaffeModel(CV_WRAP_FILE_PATH const String& src, CV_WRAP_FILE_PATH const String& dst,
                                        const std::vector<String>& layersTypes = std::vector<String>());
 
     /** @brief Create a text representation for a binary network stored in protocol buffer format.
@@ -1298,7 +1298,7 @@ CV__DNN_INLINE_NS_BEGIN
      *
      *  @note To reduce output file size, trained weights are not included.
      */
-    CV_EXPORTS_W void writeTextGraph(const String& model, const String& output);
+    CV_EXPORTS_W void writeTextGraph(CV_WRAP_FILE_PATH const String& model, CV_WRAP_FILE_PATH const String& output);
 
     /** @brief Performs non maximum suppression given boxes and corresponding scores.
 
@@ -1403,7 +1403,7 @@ CV__DNN_INLINE_NS_BEGIN
           * @param[in] model Binary file contains trained weights.
           * @param[in] config Text file contains network configuration.
           */
-         CV_WRAP Model(const String& model, const String& config = "");
+         CV_WRAP Model(CV_WRAP_FILE_PATH const String& model, CV_WRAP_FILE_PATH const String& config = "");
 
          /**
           * @brief Create model from deep learning network.
@@ -1508,7 +1508,7 @@ CV__DNN_INLINE_NS_BEGIN
           * @param[in] model Binary file contains trained weights.
           * @param[in] config Text file contains network configuration.
           */
-          CV_WRAP ClassificationModel(const String& model, const String& config = "");
+          CV_WRAP ClassificationModel(CV_WRAP_FILE_PATH const String& model, CV_WRAP_FILE_PATH const String& config = "");
 
          /**
           * @brief Create model from deep learning network.
@@ -1558,7 +1558,7 @@ CV__DNN_INLINE_NS_BEGIN
           * @param[in] model Binary file contains trained weights.
           * @param[in] config Text file contains network configuration.
           */
-          CV_WRAP KeypointsModel(const String& model, const String& config = "");
+          CV_WRAP KeypointsModel(CV_WRAP_FILE_PATH const String& model, CV_WRAP_FILE_PATH const String& config = "");
 
          /**
           * @brief Create model from deep learning network.
@@ -1590,7 +1590,7 @@ CV__DNN_INLINE_NS_BEGIN
           * @param[in] model Binary file contains trained weights.
           * @param[in] config Text file contains network configuration.
           */
-          CV_WRAP SegmentationModel(const String& model, const String& config = "");
+          CV_WRAP SegmentationModel(CV_WRAP_FILE_PATH const String& model, CV_WRAP_FILE_PATH const String& config = "");
 
          /**
           * @brief Create model from deep learning network.
@@ -1621,7 +1621,7 @@ CV__DNN_INLINE_NS_BEGIN
           * @param[in] model Binary file contains trained weights.
           * @param[in] config Text file contains network configuration.
           */
-         CV_WRAP DetectionModel(const String& model, const String& config = "");
+         CV_WRAP DetectionModel(CV_WRAP_FILE_PATH const String& model, CV_WRAP_FILE_PATH const String& config = "");
 
          /**
           * @brief Create model from deep learning network.
@@ -1687,7 +1687,7 @@ public:
      * @param[in] config Text file contains network configuration
      */
     CV_WRAP inline
-    TextRecognitionModel(const std::string& model, const std::string& config = "")
+    TextRecognitionModel(CV_WRAP_FILE_PATH const std::string& model, CV_WRAP_FILE_PATH const std::string& config = "")
         : TextRecognitionModel(readNet(model, config)) { /* nothing */ }
 
     /**
@@ -1842,7 +1842,7 @@ public:
      * @param[in] config Text file contains network configuration.
      */
     CV_WRAP inline
-    TextDetectionModel_EAST(const std::string& model, const std::string& config = "")
+    TextDetectionModel_EAST(CV_WRAP_FILE_PATH const std::string& model, CV_WRAP_FILE_PATH const std::string& config = "")
         : TextDetectionModel_EAST(readNet(model, config)) { /* nothing */ }
 
     /**
@@ -1903,7 +1903,7 @@ public:
      * @param[in] config Text file contains network configuration.
      */
     CV_WRAP inline
-    TextDetectionModel_DB(const std::string& model, const std::string& config = "")
+    TextDetectionModel_DB(CV_WRAP_FILE_PATH const std::string& model, CV_WRAP_FILE_PATH const std::string& config = "")
         : TextDetectionModel_DB(readNet(model, config)) { /* nothing */ }
 
     CV_WRAP TextDetectionModel_DB& setBinaryThreshold(float binaryThreshold);

modules/objdetect/include/opencv2/objdetect/barcode.hpp

@@ -27,7 +27,7 @@ public:
      * @param prototxt_path prototxt file path for the super resolution model
      * @param model_path model file path for the super resolution model
      */
-    CV_WRAP BarcodeDetector(const std::string &prototxt_path, const std::string &model_path);
+    CV_WRAP BarcodeDetector(CV_WRAP_FILE_PATH const std::string &prototxt_path, CV_WRAP_FILE_PATH const std::string &model_path);
     ~BarcodeDetector();
 
     /** @brief Decodes barcode in image once it's found by the detect() method.

modules/objdetect/include/opencv2/objdetect/face.hpp

@@ -82,8 +82,8 @@ public:
      *  @param backend_id the id of backend
      *  @param target_id the id of target device
      */
-    CV_WRAP static Ptr<FaceDetectorYN> create(const String& model,
-                                              const String& config,
+    CV_WRAP static Ptr<FaceDetectorYN> create(CV_WRAP_FILE_PATH const String& model,
+                                              CV_WRAP_FILE_PATH const String& config,
                                               const Size& input_size,
                                               float score_threshold = 0.9f,
                                               float nms_threshold = 0.3f,
@@ -154,7 +154,7 @@ public:
      *  @param backend_id the id of backend
      *  @param target_id the id of target device
      */
-    CV_WRAP static Ptr<FaceRecognizerSF> create(const String& model, const String& config, int backend_id = 0, int target_id = 0);
+    CV_WRAP static Ptr<FaceRecognizerSF> create(CV_WRAP_FILE_PATH const String& model, CV_WRAP_FILE_PATH const String& config, int backend_id = 0, int target_id = 0);
 };
 
 //! @}

modules/python/src2/cv2.hpp

@@ -45,17 +45,20 @@ class ArgInfo
 private:
     static const uint32_t arg_outputarg_flag     = 0x1;
     static const uint32_t arg_arithm_op_src_flag = 0x2;
+    static const uint32_t arg_pathlike_flag      = 0x4;
 
 public:
     const char* name;
     bool outputarg;
     bool arithm_op_src;
+    bool pathlike;
     // more fields may be added if necessary
 
     ArgInfo(const char* name_, uint32_t arg_) :
         name(name_),
         outputarg((arg_ & arg_outputarg_flag) != 0),
-        arithm_op_src((arg_ & arg_arithm_op_src_flag) != 0) {}
+        arithm_op_src((arg_ & arg_arithm_op_src_flag) != 0),
+        pathlike((arg_ & arg_pathlike_flag) != 0) {}
 
 private:
     ArgInfo(const ArgInfo&) = delete;

modules/python/src2/cv2_convert.cpp

@@ -701,6 +701,18 @@ bool pyopencv_to(PyObject* obj, String &value, const ArgInfo& info)
         return true;
     }
     std::string str;
+
+#if ((PY_VERSION_HEX >= 0x03060000) && !defined(Py_LIMITED_API)) || (Py_LIMITED_API >= 0x03060000)
+    if (info.pathlike)
+    {
+        obj = PyOS_FSPath(obj);
+        if (PyErr_Occurred())
+        {
+            failmsg("Expected '%s' to be a str or path-like object", info.name);
+            return false;
+        }
+    }
+#endif
     if (getUnicodeString(obj, str))
     {
         value = str;

modules/python/src2/gen2.py

@@ -500,6 +500,10 @@ class ArgInfo(object):
     def outputarg(self):
         return '/O' in self._modifiers or '/IO' in self._modifiers
 
+    @property
+    def pathlike(self):
+        return '/PATH' in self._modifiers
+
     @property
     def returnarg(self):
         return self.outputarg
@@ -523,6 +527,7 @@ class ArgInfo(object):
     def crepr(self):
         arg  = 0x01 if self.outputarg else 0x0
         arg += 0x02 if self.arithm_op_src_arg else 0x0
+        arg += 0x04 if self.pathlike else 0x0
         return "ArgInfo(\"%s\", %d)" % (self.name, arg)
 
 

modules/python/src2/hdr_parser.py

@@ -90,6 +90,10 @@ class CppHeaderParser(object):
             modlist.append("/IO")
             arg_str = arg_str.replace("CV_IN_OUT", "")
 
+        if "CV_WRAP_FILE_PATH" in arg_str:
+            modlist.append("/PATH")
+            arg_str = arg_str.replace("CV_WRAP_FILE_PATH", "")
+
         isarray = False
         npos = arg_str.find("CV_CARRAY")
         if npos >= 0:
@@ -627,6 +631,8 @@ class CppHeaderParser(object):
                                                              ("noArray", arg_type)]).strip()
                     if '/IO' in modlist and '/O' in modlist:
                         modlist.remove('/O')
+                    if (arg_name.lower() == 'filename' or arg_name.lower() == 'filepath') and '/PATH' not in modlist:
+                        modlist.append('/PATH')
                     args.append([arg_type, arg_name, defval, modlist])
                 npos = arg_start-1
 

modules/python/test/test_pathlike.py

@@ -0,0 +1,38 @@
+from tests_common import NewOpenCVTests, unittest
+import cv2 as cv
+import os
+
+
+def import_path():
+    import sys
+    if sys.version_info[0] < 3 or sys.version_info[1] < 6:
+        raise unittest.SkipTest('Python 3.6+ required')
+
+    from pathlib import Path
+    return Path
+
+
+class CanPassPathLike(NewOpenCVTests):
+    def test_pathlib_path(self):
+        Path = import_path()
+
+        img_path = self.find_file('cv/imgproc/stuff.jpg', [os.environ.get('OPENCV_TEST_DATA_PATH')])
+
+        image_from_str = cv.imread(img_path)
+        self.assertIsNotNone(image_from_str)
+
+        image_from_path = cv.imread(Path(img_path))
+        self.assertIsNotNone(image_from_path)
+
+
+    def test_type_mismatch(self):
+        import_path() # checks python version
+
+        with self.assertRaises(TypeError) as context:
+            cv.imread(123)
+
+        self.assertTrue('str or path-like' in str(context.exception))
+
+
+if __name__ == '__main__':
+    NewOpenCVTests.bootstrap()