md fix voc and manifest error scenario

mindspore/ccsrc/minddata/dataset/engine/datasetops/source/manifest_op.cc

@@ -174,10 +174,16 @@ Status ManifestOp::ParseManifestFile() {
   }
   std::string line;
   std::set<std::string> classes;
+  uint64_t line_count = 1;
   while (getline(file_handle, line)) {
     try {
       nlohmann::json js = nlohmann::json::parse(line);
       std::string image_file_path = js.value("source", "");
+      if (image_file_path == "") {
+        file_handle.close();
+        RETURN_STATUS_UNEXPECTED("Invalid data, source is not found in Manifest file: " + file_ + " at line " +
+                                 std::to_string(line_count));
+      }
       // If image is not JPEG/PNG/GIF/BMP, drop it
       bool valid = false;
       RETURN_IF_NOT_OK(CheckImageType(image_file_path, &valid));
@@ -185,6 +191,11 @@ Status ManifestOp::ParseManifestFile() {
         continue;
       }
       std::string usage = js.value("usage", "");
+      if (usage == "") {
+        file_handle.close();
+        RETURN_STATUS_UNEXPECTED("Invalid data, usage is not found in Manifest file: " + file_ + " at line " +
+                                 std::to_string(line_count));
+      }
       (void)std::transform(usage.begin(), usage.end(), usage.begin(), ::tolower);
       if (usage != usage_) {
         continue;
@@ -197,7 +208,8 @@ Status ManifestOp::ParseManifestFile() {
         classes.insert(label_name);
         if (label_name == "") {
           file_handle.close();
-          RETURN_STATUS_UNEXPECTED("Invalid data, label name is not found in Manifest file: " + image_file_path);
+          RETURN_STATUS_UNEXPECTED("Invalid data, label name is not found in Manifest file: " + file_ + " at line " +
+                                   std::to_string(line_count));
         }
         if (class_index_.empty() || class_index_.find(label_name) != class_index_.end()) {
           if (label_index_.find(label_name) == label_index_.end()) {
@@ -209,6 +221,7 @@ Status ManifestOp::ParseManifestFile() {
       if (!labels.empty()) {
         image_labelname_.emplace_back(std::make_pair(image_file_path, labels));
       }
+      line_count++;
     } catch (const std::exception &err) {
       file_handle.close();
       RETURN_STATUS_UNEXPECTED("Invalid file, failed to parse manifest file: " + file_);

mindspore/ccsrc/minddata/dataset/engine/datasetops/source/voc_op.cc

@@ -213,6 +213,17 @@ void VOCOp::ParseNodeValue(XMLElement *bbox_node, const char *name, float *value
   }
 }
 
+Status VOCOp::CheckIfBboxValid(const float &xmin, const float &ymin, const float &xmax, const float &ymax,
+                               const std::string &path) {
+  if (!(xmin > 0 && ymin > 0 && xmax > xmin && ymax > ymin)) {
+    std::string invalid_bbox = "{" + std::to_string(static_cast<int>(xmin)) + ", " +
+                               std::to_string(static_cast<int>(ymin)) + ", " + std::to_string(static_cast<int>(xmax)) +
+                               ", " + std::to_string(static_cast<int>(ymax)) + "}";
+    RETURN_STATUS_UNEXPECTED("Invalid bndbox: " + invalid_bbox + " found in " + path);
+  }
+  return Status::OK();
+}
+
 Status VOCOp::ParseAnnotationBbox(const std::string &path) {
   if (!Path(path).Exists()) {
     RETURN_STATUS_UNEXPECTED("Invalid file, failed to open file: " + path);
@@ -245,6 +256,7 @@ Status VOCOp::ParseAnnotationBbox(const std::string &path) {
       ParseNodeValue(bbox_node, "xmax", &xmax);
       ParseNodeValue(bbox_node, "ymin", &ymin);
       ParseNodeValue(bbox_node, "ymax", &ymax);
+      RETURN_IF_NOT_OK(CheckIfBboxValid(xmin, ymin, xmax, ymax, path));
     } else {
       RETURN_STATUS_UNEXPECTED("Invalid data, bndbox dismatch in " + path);
     }

mindspore/ccsrc/minddata/dataset/engine/datasetops/source/voc_op.h

@@ -230,6 +230,15 @@ class VOCOp : public MappableLeafOp {
   // @return Status The status code returned
   Status ParseAnnotationBbox(const std::string &path);
 
+  // @param xmin - the left coordinate of bndbox
+  // @param ymin - the top coordinate of bndbox
+  // @param xmax - the right coordinate of bndbox
+  // @param ymax - the bottom coordinate of bndbox
+  // @param path - the file path of bndbox xml
+  // @return Status The status code returned
+  Status CheckIfBboxValid(const float &xmin, const float &ymin, const float &xmax, const float &ymax,
+                          const std::string &path);
+
   // @param XMLElement *bbox_node - bbox node info found in json object
   // @param const char *name - sub node name in object
   // @param float *value - value of certain sub node

tests/ut/data/dataset/testManifestData/invalidNoSource.manifest

@@ -0,0 +1,6 @@
+{"source":"../data/dataset/testManifestData/train/1.JPEG", "usage":"TRAIN","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "dog","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}
+{"usage":"TRAIN","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "cat","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}
+{"source":"../data/dataset/testManifestData/train/1.JPEG", "usage":"TRAIN","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "cat","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}
+{"source":"../data/dataset/testManifestData/train/1.JPEG", "usage":"TRAIN","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "cat","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"},{"type": "modelarts/image_classification","name": "flower","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}
+{"source":"../data/dataset/testManifestData/eval/1.JPEG", "usage":"EVAL","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "cat","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}
+{"source":"../data/dataset/testManifestData/eval/2.JPEG", "usage":"EVAL","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "dog","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}

tests/ut/data/dataset/testManifestData/invalidNoUsage.manifest

@@ -0,0 +1,6 @@
+{"source":"../data/dataset/testManifestData/train/1.JPEG","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "dog","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}
+{"source":"../data/dataset/testManifestData/train/1.JPEG", "usage":"TRAIN","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "cat","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}
+{"source":"../data/dataset/testManifestData/train/1.JPEG", "usage":"TRAIN","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "cat","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}
+{"source":"../data/dataset/testManifestData/train/1.JPEG", "usage":"TRAIN","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "cat","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"},{"type": "modelarts/image_classification","name": "flower","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}
+{"source":"../data/dataset/testManifestData/eval/1.JPEG", "usage":"EVAL","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "cat","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}
+{"source":"../data/dataset/testManifestData/eval/2.JPEG", "usage":"EVAL","id":"0162005993f8065ef47eefb59d1e4970","annotation": [{"type": "modelarts/image_classification","name": "dog","property": {"color":"white","kind":"Persian cat"},"hard":"true","hard-coefficient":0.8,"annotated-by":"human","creation-time":"2019-01-23 11:30:30"}],"inference-loc":"/path/to/inference-output"}

tests/ut/data/dataset/testVOC2012/Annotations/33.xml

@@ -24,18 +24,6 @@
 			<ymax>242</ymax>
 		</bndbox>
 	</object>
-	<object>
-		<name>person</name>
-		<pose>Left</pose>
-		<truncated>0</truncated>
-		<difficult>0</difficult>
-		<bndbox>
-			<xmin>425</xmin>
-			<ymin>239</ymin>
-			<xmax>445</xmax>
-			<ymax>235</ymax>
-		</bndbox>
-	</object>
 	<object>
 		<name>person</name>
 		<pose>Left</pose>

tests/ut/data/dataset/testVOC2012/Annotations/invalidbbox.xml

@@ -0,0 +1,39 @@
+<annotation>
+	<folder>VOC2012</folder>
+	<filename>33.jpg</filename>
+	<source>
+		<database>simulate VOC2007 Database</database>
+		<annotation>simulate VOC2007</annotation>
+		<image>flickr</image>
+	</source>
+	<size>
+		<width>500</width>
+		<height>366</height>
+		<depth>3</depth>
+	</size>
+	<segmented>1</segmented>
+	<object>
+		<name>person</name>
+		<pose>Unspecified</pose>
+		<truncated>0</truncated>
+		<difficult>0</difficult>
+		<bndbox>
+			<xmin>8</xmin>
+			<ymin>121</ymin>
+			<xmax>471</xmax>
+			<ymax>242</ymax>
+		</bndbox>
+	</object>
+	<object>
+		<name>person</name>
+		<pose>Left</pose>
+		<truncated>1</truncated>
+		<difficult>0</difficult>
+		<bndbox>
+			<xmin>321</xmin>
+			<ymin>121</ymin>
+			<xmax>421</xmax>
+                        <ymax>120</ymax>
+		</bndbox>
+	</object>
+</annotation>

tests/ut/data/dataset/testVOC2012/ImageSets/Main/xmlinvalidbbox.txt

@@ -0,0 +1 @@
+invalidbbox

tests/ut/python/dataset/test_datasets_manifestop.py

@@ -160,6 +160,24 @@ def test_manifest_dataset_exception():
     except RuntimeError as e:
         assert "map operation: [PyFunc] failed. The corresponding data files" in str(e)
 
+    NO_SOURCE_DATA_FILE = "../data/dataset/testManifestData/invalidNoSource.manifest"
+    try:
+        data = ds.ManifestDataset(NO_SOURCE_DATA_FILE)
+        for _ in data.__iter__():
+            pass
+        assert False
+    except RuntimeError as e:
+        assert "Invalid data, source is not found in Manifest file" in str(e)
+
+    NO_USAGE_DATA_FILE = "../data/dataset/testManifestData/invalidNoUsage.manifest"
+    try:
+        data = ds.ManifestDataset(NO_USAGE_DATA_FILE)
+        for _ in data.__iter__():
+            pass
+        assert False
+    except RuntimeError as e:
+        assert "Invalid data, usage is not found in Manifest file" in str(e)
+
 
 if __name__ == '__main__':
     test_manifest_dataset_train()

tests/ut/python/dataset/test_datasets_voc.py

@@ -181,6 +181,14 @@ def test_voc_exception():
     except RuntimeError:
         pass
 
+    try:
+        data7 = ds.VOCDataset(DATA_DIR, task="Detection", usage="xmlinvalidbbox")
+        for _ in data7.create_dict_iterator(num_epochs=1):
+            pass
+        assert False
+    except RuntimeError as e:
+        assert "Invalid bndbox: {321, 121, 421, 120}" in str(e)
+
     def exception_func(item):
         raise Exception("Error occur!")