<change> api json format

.gitignore

@@ -7,6 +7,8 @@ z_sandbox/
 
 # databases
 app/*.sqlite
+app/klang/data_dev
+app/klang/data_test
 
 .idea/
 *.pyc

app/klang/controller.py

@@ -21,7 +21,8 @@ def get(self) :
 class ConllNameServiceResource(Resource):
     "ConllService"
 
-    def get(self, conll_name) :
+    def get(self, conll_name):
+
         conll_string = ConllService.get_by_name(conll_name)
         sentences_string = ConllService.seperate_conll_sentences(conll_string)
         sentences_audio_token = []

app/klang/data_test/John_Doe/John_Doe.intervals.conll

@@ -1,9 +1,5 @@
 # sent_id = John_Doe.intervals.conll__1
-# text = it is a test conll
-# sound_url = Veronique_Jezewski.wav
-1	it	it	_	_	_	_	_	_	AlignBegin=162330|AlignEnd=162590
-2	is	is	_	_	_	_	_	_	AlignBegin=162590|AlignEnd=162849
-3	a	a	_	_	_	_	_	_	AlignBegin=162849|AlignEnd=163109
-4	test	test	_	_	_	_	_	_	AlignBegin=163368|AlignEnd=163628
-5	que	que	_	_	_	_	_	_	AlignBegin=163628|AlignEnd=163887
-6	conll	conll	_	_	_	_	_	_	AlignBegin=163887|AlignEnd=164147
+# text = it is
+# sound_url = John_Doe.wav
+1	it	it	_	_	_	_	_	_	AlignBegin=100|AlignEnd=500
+2	is	is	_	_	_	_	_	_	AlignBegin=600|AlignEnd=1000

app/klang/service.py

@@ -8,6 +8,7 @@
     r"^\d+\t(.+?)\t.*AlignBegin=(\d+).*AlignEnd=(\d+)"
 )
 
+
 class ConllService:
     @staticmethod
     def get_path_data():
@@ -45,11 +46,27 @@ def seperate_conll_sentences(conll_string: str) -> List[str]:
 
     @staticmethod
     def sentence_to_audio_tokens(sentence_string: str):
-        audio_tokens = []
+        audio_tokens = {}
         for line in sentence_string.split("\n"):
             if line:
                 if not line.startswith("#"):
                     m = align_begin_and_end_regex.search(line)
-                    audio_tokens += [(m.group(1), int(m.group(2)), int(m.group(3)))]
-
+                    audio_token = {
+                        "token": m.group(1),
+                        "alignBegin": int(m.group(2)),
+                        "alignEnd": int(m.group(3)),
+                    }
+                    audio_tokens[int(line.split("\t")[0])] = audio_token
+
+        print(audio_tokens)
         return audio_tokens
+
+    @staticmethod
+    def process_sentences_audio_token(conll_name: str):
+        conll_string = ConllService.get_by_name(conll_name)
+        sentences_string = ConllService.seperate_conll_sentences(conll_string)
+        sentences_audio_token = []
+        for sentence_string in sentences_string:
+          audio_tokens = ConllService.sentence_to_audio_tokens(sentence_string)
+          sentences_audio_token.append(audio_tokens)
+        return sentences_audio_token

app/klang/service_test.py

@@ -13,32 +13,46 @@ def test_get_path_data():
     print("KK path_data", path_data)
     assert os.path.isdir(path_data)
 
+
 def test_get_path_conll():
     path_conll = ConllService.get_path_conll(file_name)
     assert os.path.isfile(path_conll)
 
+
 def test_read_conll():
     path_conll = ConllService.get_path_conll(file_name)
     conll_string = ConllService.read_conll(path_conll)
     assert type(conll_string) == str
 
+
 def test_get_all():
     conlls = ConllService.get_all()
     assert conlls == ["John_Doe"]
 
+
 def test_get_by_name():
     path_conll = ConllService.get_path_conll(file_name)
     conll_string = ConllService.read_conll(path_conll)
     assert conll_string == ConllService.get_by_name(file_name)
 
+
 def test_seperate_conll_sentences():
     conll_string = "# sent_id = test_sentence_1\n1\ttest_token\ntest_lemma\ntest_upos\n\n# sent_id = test_sentence_2\n1\ttest_token\ntest_lemma\ntest_upos\n\n"
     sentences = ConllService.seperate_conll_sentences(conll_string)
     assert len(sentences) == 2
 
+
 def test_sentence_to_audio_tokens():
     sentence = "# sent_id = test_sentence_1\n1\tdonc\tdonc\t_\t_\t_\t_\t_\t_\tAlignBegin=0|AlignEnd=454"
     audio_tokens = ConllService.sentence_to_audio_tokens(sentence)
-    assert audio_tokens[0][0] == "donc"
-    assert audio_tokens[0][1] == 0
-    assert audio_tokens[0][2] == 454
+    assert audio_tokens[1]["token"] == "donc"
+    assert audio_tokens[1]["alignBegin"] == 0
+    assert audio_tokens[1]["alignEnd"] == 454
+
+
+def test_process_sentences_audio_token():
+    sentences_audio_token = ConllService.process_sentences_audio_token(file_name)
+    assert sentences_audio_token == [{
+        1: {"token": "it", "alignBegin": 100, "alignEnd": 500},
+        2: {"token": "is", "alignBegin": 600, "alignEnd": 1000}
+    }]