!5671 modify voc doc & fix buildvocab bug

Merge pull request !5671 from luoyang/c-api-pyfunc

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

@@ -324,7 +324,7 @@ std::shared_ptr<Vocab> Dataset::BuildVocab(const std::vector<std::string> &colum
   // Finish building vocab by triggering GetNextRow
   std::unordered_map<std::string, std::shared_ptr<Tensor>> row;
   iter->GetNextRow(&row);
-  if (vocab == nullptr) {
+  if (vocab->vocab().empty()) {
     MS_LOG(ERROR) << "Fail to build vocab.";
     return nullptr;
   }

mindspore/ccsrc/minddata/dataset/text/vocab.h

@@ -107,6 +107,9 @@ class Vocab {
   // @param std::string & word - word to be added will skip if word already exists
   void append_word(const std::string &word);
 
+  // return a read-only vocab
+  const std::unordered_map<WordType, WordIdType> vocab() { return word2id_; }
+
   // destructor
   ~Vocab() = default;
 

mindspore/dataset/engine/datasets.py

@@ -4457,8 +4457,8 @@ class VOCDataset(MappableDataset):
         task (str): Set the task type of reading voc data, now only support "Segmentation" or "Detection"
             (default="Segmentation").
         mode (str): Set the data list txt file to be readed (default="train").
-        class_indexing (dict, optional): A str-to-int mapping from label name to index
-            (default=None, the folder names will be sorted alphabetically and each
+        class_indexing (dict, optional): A str-to-int mapping from label name to index, only valid in
+            "Detection" task (default=None, the folder names will be sorted alphabetically and each
             class will be given a unique index starting from 0).
         num_samples (int, optional): The number of images to be included in the dataset
             (default=None, all images).

tests/ut/cpp/dataset/c_api_dataset_vocab.cc

@@ -252,3 +252,17 @@ TEST_F(MindDataTestPipeline, TestVocabFromDatasetFail2) {
                                                 std::numeric_limits<int64_t>::max(), {"<pad>", "<unk>"}, true);
   EXPECT_EQ(vocab, nullptr);
 }
+
+TEST_F(MindDataTestPipeline, TestVocabFromDatasetFail3) {
+  MS_LOG(INFO) << "Doing MindDataTestPipeline-TestVocabFromDatasetFail3.";
+
+  // Create a TextFile dataset
+  std::string data_file = datasets_root_path_ + "/testVocab/words.txt";
+  std::shared_ptr<Dataset> ds = TextFile({data_file}, 0, ShuffleMode::kFalse);
+  EXPECT_NE(ds, nullptr);
+
+  // Create vocab from dataset
+  // Expected failure: column name does not exist in ds
+  std::shared_ptr<Vocab> vocab = ds->BuildVocab({"ColumnNotExist"});
+  EXPECT_EQ(vocab, nullptr);
+}