Adds Sync API tests for CaptionFile ViewSet

contentcuration/contentcuration/constants/transcription_languages.py

@@ -0,0 +1,106 @@
+# This file contains a list of transcription languages.
+# The list is in the format of (language code, language name).
+# For example, the first element in the list is ('en', 'english').
+
+
+CAPTIONS_LANGUAGES = [
+    ("en", "english"),
+    ("zh", "chinese"),
+    ("de", "german"),
+    ("es", "spanish"),
+    ("ru", "russian"),
+    ("ko", "korean"),
+    ("fr", "french"),
+    ("ja", "japanese"),
+    ("pt", "portuguese"),
+    ("tr", "turkish"),
+    ("pl", "polish"),
+    ("ca", "catalan"),
+    ("nl", "dutch"),
+    ("ar", "arabic"),
+    ("sv", "swedish"),
+    ("it", "italian"),
+    ("id", "indonesian"),
+    ("hi", "hindi"),
+    ("fi", "finnish"),
+    ("vi", "vietnamese"),
+    ("he", "hebrew"),
+    ("uk", "ukrainian"),
+    ("el", "greek"),
+    ("ms", "malay"),
+    ("cs", "czech"),
+    ("ro", "romanian"),
+    ("da", "danish"),
+    ("hu", "hungarian"),
+    ("ta", "tamil"),
+    ("no", "norwegian"),
+    ("th", "thai"),
+    ("ur", "urdu"),
+    ("hr", "croatian"),
+    ("bg", "bulgarian"),
+    ("lt", "lithuanian"),
+    ("la", "latin"),
+    ("mi", "maori"),
+    ("ml", "malayalam"),
+    ("cy", "welsh"),
+    ("sk", "slovak"),
+    ("te", "telugu"),
+    ("fa", "persian"),
+    ("lv", "latvian"),
+    ("bn", "bengali"),
+    ("sr", "serbian"),
+    ("az", "azerbaijani"),
+    ("sl", "slovenian"),
+    ("kn", "kannada"),
+    ("et", "estonian"),
+    ("mk", "macedonian"),
+    ("br", "breton"),
+    ("eu", "basque"),
+    ("is", "icelandic"),
+    ("hy", "armenian"),
+    ("ne", "nepali"),
+    ("mn", "mongolian"),
+    ("bs", "bosnian"),
+    ("kk", "kazakh"),
+    ("sq", "albanian"),
+    ("sw", "swahili"),
+    ("gl", "galician"),
+    ("mr", "marathi"),
+    ("pa", "punjabi"),
+    ("si", "sinhala"),
+    ("km", "khmer"),
+    ("sn", "shona"),
+    ("yo", "yoruba"),
+    ("so", "somali"),
+    ("af", "afrikaans"),
+    ("oc", "occitan"),
+    ("ka", "georgian"),
+    ("be", "belarusian"),
+    ("tg", "tajik"),
+    ("sd", "sindhi"),
+    ("gu", "gujarati"),
+    ("am", "amharic"),
+    ("yi", "yiddish"),
+    ("lo", "lao"),
+    ("uz", "uzbek"),
+    ("fo", "faroese"),
+    ("ht", "haitian creole"),
+    ("ps", "pashto"),
+    ("tk", "turkmen"),
+    ("nn", "nynorsk"),
+    ("mt", "maltese"),
+    ("sa", "sanskrit"),
+    ("lb", "luxembourgish"),
+    ("my", "myanmar"),
+    ("bo", "tibetan"),
+    ("tl", "tagalog"),
+    ("mg", "malagasy"),
+    ("as", "assamese"),
+    ("tt", "tatar"),
+    ("haw", "hawaiian"),
+    ("ln", "lingala"),
+    ("ha", "hausa"),
+    ("ba", "bashkir"),
+    ("jw", "javanese"),
+    ("su", "sundanese"),
+]

contentcuration/contentcuration/migrations/0143_captioncue_captionfile.py

@@ -0,0 +1,37 @@
+# Generated by Django 3.2.14 on 2023-06-26 17:33
+
+import contentcuration.models
+from django.db import migrations, models
+import django.db.models.deletion
+import uuid
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('contentcuration', '0142_add_task_signature'),
+    ]
+
+    operations = [
+        migrations.CreateModel(
+            name='CaptionFile',
+            fields=[
+                ('id', contentcuration.models.UUIDField(default=uuid.uuid4, max_length=32, primary_key=True, serialize=False)),
+                ('file_id', contentcuration.models.UUIDField(default=uuid.uuid4, max_length=32)),
+                ('language', models.CharField(choices=[('en', 'english'), ('zh', 'chinese'), ('de', 'german'), ('es', 'spanish'), ('ru', 'russian'), ('ko', 'korean'), ('fr', 'french'), ('ja', 'japanese'), ('pt', 'portuguese'), ('tr', 'turkish'), ('pl', 'polish'), ('ca', 'catalan'), ('nl', 'dutch'), ('ar', 'arabic'), ('sv', 'swedish'), ('it', 'italian'), ('id', 'indonesian'), ('hi', 'hindi'), ('fi', 'finnish'), ('vi', 'vietnamese'), ('he', 'hebrew'), ('uk', 'ukrainian'), ('el', 'greek'), ('ms', 'malay'), ('cs', 'czech'), ('ro', 'romanian'), ('da', 'danish'), ('hu', 'hungarian'), ('ta', 'tamil'), ('no', 'norwegian'), ('th', 'thai'), ('ur', 'urdu'), ('hr', 'croatian'), ('bg', 'bulgarian'), ('lt', 'lithuanian'), ('la', 'latin'), ('mi', 'maori'), ('ml', 'malayalam'), ('cy', 'welsh'), ('sk', 'slovak'), ('te', 'telugu'), ('fa', 'persian'), ('lv', 'latvian'), ('bn', 'bengali'), ('sr', 'serbian'), ('az', 'azerbaijani'), ('sl', 'slovenian'), ('kn', 'kannada'), ('et', 'estonian'), ('mk', 'macedonian'), ('br', 'breton'), ('eu', 'basque'), ('is', 'icelandic'), ('hy', 'armenian'), ('ne', 'nepali'), ('mn', 'mongolian'), ('bs', 'bosnian'), ('kk', 'kazakh'), ('sq', 'albanian'), ('sw', 'swahili'), ('gl', 'galician'), ('mr', 'marathi'), ('pa', 'punjabi'), ('si', 'sinhala'), ('km', 'khmer'), ('sn', 'shona'), ('yo', 'yoruba'), ('so', 'somali'), ('af', 'afrikaans'), ('oc', 'occitan'), ('ka', 'georgian'), ('be', 'belarusian'), ('tg', 'tajik'), ('sd', 'sindhi'), ('gu', 'gujarati'), ('am', 'amharic'), ('yi', 'yiddish'), ('lo', 'lao'), ('uz', 'uzbek'), ('fo', 'faroese'), ('ht', 'haitian creole'), ('ps', 'pashto'), ('tk', 'turkmen'), ('nn', 'nynorsk'), ('mt', 'maltese'), ('sa', 'sanskrit'), ('lb', 'luxembourgish'), ('my', 'myanmar'), ('bo', 'tibetan'), ('tl', 'tagalog'), ('mg', 'malagasy'), ('as', 'assamese'), ('tt', 'tatar'), ('haw', 'hawaiian'), ('ln', 'lingala'), ('ha', 'hausa'), ('ba', 'bashkir'), ('jw', 'javanese'), ('su', 'sundanese')], max_length=3)),
+            ],
+            options={
+                'unique_together': {('file_id', 'language')},
+            },
+        ),
+        migrations.CreateModel(
+            name='CaptionCue',
+            fields=[
+                ('id', contentcuration.models.UUIDField(default=uuid.uuid4, max_length=32, primary_key=True, serialize=False)),
+                ('text', models.TextField()),
+                ('starttime', models.FloatField()),
+                ('endtime', models.FloatField()),
+                ('caption_file', models.ForeignKey(on_delete=django.db.models.deletion.CASCADE, related_name='caption_cue', to='contentcuration.captionfile')),
+            ],
+        ),
+    ]

contentcuration/contentcuration/models.py

@@ -2,6 +2,7 @@
 import json
 import logging
 import os
+from typing import Any
 import urllib.parse
 import uuid
 from datetime import datetime
@@ -68,6 +69,7 @@
 from contentcuration.constants import channel_history
 from contentcuration.constants import completion_criteria
 from contentcuration.constants import user_history
+from contentcuration.constants.transcription_languages import CAPTIONS_LANGUAGES
 from contentcuration.constants.contentnode import kind_activity_map
 from contentcuration.db.models.expressions import Array
 from contentcuration.db.models.functions import ArrayRemove
@@ -2057,15 +2059,42 @@ def __str__(self):
         return self.ietf_name()
 
 
-class Caption(models.Model):
+class CaptionFile(models.Model):
     """
-    Model to store captions and support intermediary changes
+    Represents a caption file record.
+
+    - file_id: The identifier of related file in Google Cloud Storage.
+    - language: The language of the caption file.
+    """
+    id = UUIDField(primary_key=True, default=uuid.uuid4)
+    file_id = UUIDField(default=uuid.uuid4, max_length=36)
+    language = models.CharField(choices=CAPTIONS_LANGUAGES, max_length=3)
+
+    class Meta:
+        unique_together = ['file_id', 'language']
+
+    def __str__(self):
+        return "{file_id} -> {language}".format(file_id=self.file_id, language=self.language)
+
+
+class CaptionCue(models.Model):
+    """
+    Represents a caption cue in a VTT file.
+
+    - text: The caption text.
+    - starttime: The start time of the cue in seconds.
+    - endtime: The end time of the cue in seconds.
+    - caption_file (Foreign Key): The related caption file.
     """
     id = UUIDField(primary_key=True, default=uuid.uuid4)
-    caption = models.JSONField()
-    language = models.CharField(max_length=10)
-    # file_id = models.CharField(unique=True, max_length=32)
-
+    text = models.TextField(null=False)
+    starttime = models.FloatField(null=False)
+    endtime = models.FloatField(null=False)
+    caption_file = models.ForeignKey(CaptionFile, related_name="caption_cue", on_delete=models.CASCADE)
+
+    def __str__(self):
+        return "text: {text}, start_time: {starttime}, end_time: {endtime}".format(text=self.text, starttime=self.starttime, endtime=self.endtime)
+
 
 ASSESSMENT_ID_INDEX_NAME = "assessment_id_idx"
 

contentcuration/contentcuration/tests/viewsets/test_caption.py

@@ -0,0 +1,147 @@
+from __future__ import absolute_import
+
+import uuid
+
+from django.urls import reverse
+
+from contentcuration.models import CaptionCue, CaptionFile
+from contentcuration.tests.base import StudioAPITestCase
+from contentcuration.tests import testdata
+from contentcuration.tests.viewsets.base import SyncTestMixin
+from contentcuration.tests.viewsets.base import generate_create_event
+from contentcuration.tests.viewsets.base import generate_update_event
+from contentcuration.tests.viewsets.base import generate_delete_event
+from contentcuration.viewsets.sync.constants import CAPTION_FILE
+
+# class CRUDTestCase(StudioAPITestCase):
+
+class SyncTestCase(SyncTestMixin, StudioAPITestCase):
+
+    @property
+    def caption_file_metadata(self):
+        return {
+            "file_id": uuid.uuid4().hex,
+            "language": "en",
+        }
+
+    @property
+    def same_file_different_language_metadata(self):
+        id = uuid.uuid4().hex
+        return [
+            {
+                "file_id": id,
+                "language": "en",
+            },
+            {
+                "file_id": id,
+                "language": "ru",
+            }
+        ]
+
+    @property
+    def caption_file_db_metadata(self):
+        return {
+            "file_id": uuid.uuid4().hex,
+            "language": "en",
+        }
+
+
+    def setUp(self):
+        super(SyncTestCase, self).setUp()
+        self.channel = testdata.channel()
+        self.user = testdata.user()
+        self.channel.editors.add(self.user)
+
+
+    def test_create_caption(self):
+        self.client.force_authenticate(user=self.user)
+        caption_file = self.caption_file_metadata
+
+        response = self.sync_changes(
+            [
+                generate_create_event(
+                    uuid.uuid4().hex,
+                    CAPTION_FILE,
+                    caption_file,
+                    channel_id=self.channel.id,
+                )
+            ],
+        )
+        self.assertEqual(response.status_code, 200, response.content)
+
+        try:
+            caption_file_db = CaptionFile.objects.get(
+                file_id=caption_file["file_id"],
+                language=caption_file["language"],
+            )
+        except CaptionFile.DoesNotExist:
+            self.fail("caption file was not created")
+
+        # Check the values of the object in the PostgreSQL
+        self.assertEqual(caption_file_db.file_id, caption_file["file_id"])
+        self.assertEqual(caption_file_db.language, caption_file["language"])
+
+    def test_delete_caption_file(self):
+        self.client.force_authenticate(user=self.user)
+        caption_file = self.caption_file_metadata
+        pk = uuid.uuid4().hex
+        response = self.sync_changes(
+            [
+                generate_create_event(
+                    pk,
+                    CAPTION_FILE,
+                    caption_file,
+                    channel_id=self.channel.id
+                )
+            ]
+        )
+        self.assertEqual(response.status_code, 200, response.content)
+
+        # Delete the caption file
+        response = self.sync_changes(
+            [
+                generate_delete_event(
+                    pk,
+                    CAPTION_FILE,
+                    channel_id=self.channel.id
+                )
+            ]
+        )
+
+        self.assertEqual(response.status_code, 200, response.content)
+
+        with self.assertRaises(CaptionFile.DoesNotExist):
+            caption_file_db = CaptionFile.objects.get(
+                file_id=caption_file['file_id'],
+                language=caption_file['language']
+            )
+
+
+    def test_delete_file_with_same_file_id_different_language(self):
+        self.client.force_authenticate(user=self.user)
+        obj = self.same_file_different_language_metadata
+
+        caption_file_1 = CaptionFile.objects.create(
+            **obj[0]
+        )
+        caption_file_2 = CaptionFile.objects.create(
+            **obj[1]
+        )
+
+        response = self.sync_changes(
+            [
+                generate_delete_event(
+                    caption_file_2.pk,
+                    CAPTION_FILE,
+                    channel_id=self.channel.id,
+                )
+            ]
+        )
+
+        self.assertEqual(response.status_code, 200, response.content)
+
+        with self.assertRaises(CaptionFile.DoesNotExist):
+            caption_file_db = CaptionFile.objects.get(
+                file_id=caption_file_2.file_id,
+                language=caption_file_2.language
+            )