!4402 [Dataset] Fix c_api bugs

Merge pull request !4402 from luoyang/c-api-pyfunc

mindspore/ccsrc/minddata/dataset/api/datasets.cc

@@ -408,6 +408,10 @@ bool ValidateCommonDatasetParams(std::string dataset_dir) {
     MS_LOG(ERROR) << "No dataset path is specified";
     return false;
   }
+  if (access(dataset_dir.c_str(), R_OK) == -1) {
+    MS_LOG(ERROR) << "No access to specified dataset path: " << dataset_dir;
+    return false;
+  }
   return true;
 }
 
@@ -1185,8 +1189,8 @@ std::vector<std::shared_ptr<DatasetOp>> RepeatDataset::Build() {
 }
 
 bool RepeatDataset::ValidateParams() {
-  if (repeat_count_ != -1 && repeat_count_ <= 0) {
-    MS_LOG(ERROR) << "Repeat: Repeat count cannot be" << repeat_count_;
+  if (repeat_count_ <= 0 && repeat_count_ != -1) {
+    MS_LOG(ERROR) << "Repeat: repeat_count should be either -1 or positive integer, repeat_count_: " << repeat_count_;
     return false;
   }
 
@@ -1232,7 +1236,7 @@ std::vector<std::shared_ptr<DatasetOp>> SkipDataset::Build() {
 // Function to validate the parameters for SkipDataset
 bool SkipDataset::ValidateParams() {
   if (skip_count_ <= -1) {
-    MS_LOG(ERROR) << "Skip: Invalid input, skip_count: " << skip_count_;
+    MS_LOG(ERROR) << "Skip: skip_count should not be negative, skip_count: " << skip_count_;
     return false;
   }
 
@@ -1253,8 +1257,8 @@ std::vector<std::shared_ptr<DatasetOp>> TakeDataset::Build() {
 
 // Function to validate the parameters for TakeDataset
 bool TakeDataset::ValidateParams() {
-  if (take_count_ < -1) {
-    MS_LOG(ERROR) << "Take: Invalid input, take_count: " << take_count_;
+  if (take_count_ < 0 && take_count_ != -1) {
+    MS_LOG(ERROR) << "Take: take_count should be either -1 or positive integer, take_count: " << take_count_;
     return false;
   }
 

mindspore/ccsrc/minddata/dataset/api/iterator.cc

@@ -25,7 +25,7 @@ namespace api {
 void Iterator::GetNextRow(TensorMap *row) {
   Status rc = iterator_->GetNextAsMap(row);
   if (rc.IsError()) {
-    MS_LOG(ERROR) << "GetNextRow: Failed to get next row.";
+    MS_LOG(ERROR) << "GetNextRow: Failed to get next row. Error status: " << rc;
     row->clear();
   }
 }
@@ -35,7 +35,7 @@ void Iterator::GetNextRow(TensorVec *row) {
   TensorRow tensor_row;
   Status rc = iterator_->FetchNextTensorRow(&tensor_row);
   if (rc.IsError()) {
-    MS_LOG(ERROR) << "GetNextRow: Failed to get next row.";
+    MS_LOG(ERROR) << "GetNextRow: Failed to get next row. Error status: " << rc;
     row->clear();
   }
   // Generate a vector as return

mindspore/ccsrc/minddata/dataset/include/datasets.h

@@ -17,6 +17,7 @@
 #ifndef MINDSPORE_CCSRC_MINDDATA_DATASET_INCLUDE_DATASETS_H_
 #define MINDSPORE_CCSRC_MINDDATA_DATASET_INCLUDE_DATASETS_H_
 
+#include <unistd.h>
 #include <vector>
 #include <memory>
 #include <set>

tests/ut/cpp/dataset/c_api_dataset_ops_test.cc

@@ -437,22 +437,22 @@ TEST_F(MindDataTestPipeline, TestRepeatDefault) {
   // Create an ImageFolder Dataset
   std::string folder_path = datasets_root_path_ + "/testPK/data/";
   std::shared_ptr <Dataset> ds = ImageFolder(folder_path, true, RandomSampler(false, 10));
-  EXPECT_NE(ds,nullptr);
+  EXPECT_NE(ds, nullptr);
 
   // Create a Repeat operation on ds
   // Default value of repeat count is -1, expected to repeat infinitely
   ds = ds->Repeat();
-  EXPECT_NE(ds,nullptr);
+  EXPECT_NE(ds, nullptr);
 
   // Create a Batch operation on ds
   int32_t batch_size = 1;
   ds = ds->Batch(batch_size);
-  EXPECT_NE(ds,nullptr);
+  EXPECT_NE(ds, nullptr);
 
   // Create an iterator over the result of the above dataset
   // This will trigger the creation of the Execution Tree and launch it.
   std::shared_ptr <Iterator> iter = ds->CreateIterator();
-  EXPECT_NE(iter,nullptr);
+  EXPECT_NE(iter, nullptr);
 
   // Iterate the dataset and get each row
   std::unordered_map <std::string, std::shared_ptr<Tensor>> row;
@@ -460,14 +460,16 @@ TEST_F(MindDataTestPipeline, TestRepeatDefault) {
   uint64_t i = 0;
   while (row.size()!= 0) {
     // manually stop
-    if(i==100){break;}
+    if (i == 100) {
+      break;
+    }
     i++;
     auto image = row["image"];
     MS_LOG(INFO)<< "Tensor image shape: " << image->shape();
     iter->GetNextRow(&row);
   }
 
-  EXPECT_EQ(i,100);
+  EXPECT_EQ(i, 100);
   // Manually terminate the pipeline
   iter->Stop();
 }
@@ -478,22 +480,22 @@ TEST_F(MindDataTestPipeline, TestRepeatOne) {
   // Create an ImageFolder Dataset
   std::string folder_path = datasets_root_path_ + "/testPK/data/";
   std::shared_ptr <Dataset> ds = ImageFolder(folder_path, true, RandomSampler(false, 10));
-  EXPECT_NE(ds,nullptr);
+  EXPECT_NE(ds, nullptr);
 
   // Create a Repeat operation on ds
   int32_t repeat_num = 1;
   ds = ds->Repeat(repeat_num);
-  EXPECT_NE(ds,nullptr);
+  EXPECT_NE(ds, nullptr);
 
   // Create a Batch operation on ds
   int32_t batch_size = 1;
   ds = ds->Batch(batch_size);
-  EXPECT_NE(ds,nullptr);
+  EXPECT_NE(ds, nullptr);
 
   // Create an iterator over the result of the above dataset
   // This will trigger the creation of the Execution Tree and launch it.
   std::shared_ptr <Iterator> iter = ds->CreateIterator();
-  EXPECT_NE(iter,nullptr);
+  EXPECT_NE(iter, nullptr);
 
   // Iterate the dataset and get each row
   std::unordered_map <std::string, std::shared_ptr<Tensor>> row;
@@ -506,13 +508,27 @@ TEST_F(MindDataTestPipeline, TestRepeatOne) {
     iter->GetNextRow(&row);
   }
 
-  EXPECT_EQ(i,10);
+  EXPECT_EQ(i, 10);
   // Manually terminate the pipeline
   iter->Stop();
 }
 
-TEST_F(MindDataTestPipeline, TestRepeatFail) {
-  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestRepeatFail.";
+TEST_F(MindDataTestPipeline, TestRepeatFail1) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestRepeatFail1.";
+
+  // Create an ImageFolder Dataset
+  std::string folder_path = datasets_root_path_ + "/testPK/data/";
+  std::shared_ptr<Dataset> ds = ImageFolder(folder_path, true, RandomSampler(false, 10));
+  EXPECT_NE(ds, nullptr);
+
+  // Create a Repeat operation on ds
+  int32_t repeat_num = 0;
+  ds = ds->Repeat(repeat_num);
+  EXPECT_EQ(ds, nullptr);
+}
+
+TEST_F(MindDataTestPipeline, TestRepeatFail2) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestRepeatFail2.";
   // This case is expected to fail because the repeat count is invalid (<-1 && !=0).
 
   // Create an ImageFolder Dataset

tests/ut/cpp/dataset/c_api_datasets_test.cc

@@ -133,3 +133,25 @@ TEST_F(MindDataTestPipeline, TestMnistFail1) {
   std::shared_ptr<Dataset> ds = Mnist("", RandomSampler(false, 10));
   EXPECT_EQ(ds, nullptr);
 }
+
+TEST_F(MindDataTestPipeline, TestImageFolderFail2) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestImageFolderFail2.";
+
+  // Create an ImageFolder Dataset
+  std::string folder_path = datasets_root_path_ + "/testPK/data/";
+  std::shared_ptr<Dataset> ds = ImageFolder(folder_path, true, RandomSampler(false, 2), {".JGP"});
+  EXPECT_NE(ds, nullptr);
+
+  // Create an iterator over the result of the above dataset
+  // This will trigger the creation of the Execution Tree and launch it.
+  std::shared_ptr<Iterator> iter = ds->CreateIterator();
+  EXPECT_NE(iter, nullptr);
+
+  // Iterate the dataset and get each row
+  std::unordered_map<std::string, std::shared_ptr<Tensor>> row;
+  iter->GetNextRow(&row);
+  EXPECT_EQ(row.size(), 0);
+
+  // Manually terminate the pipeline
+  iter->Stop();
+}