feat: add whisper tasks

app/shared/db/alembic/versions/426b6bdc3360_add_job_tables.py

@@ -1,8 +1,8 @@
 """add_job_tables
 
-Revision ID: bb249ed79907
+Revision ID: 426b6bdc3360
 Revises:
-Create Date: 2023-01-17 14:30:30.920466
+Create Date: 2023-01-27 17:55:21.758828
 
 """
 import sqlalchemy as sa
@@ -10,7 +10,7 @@ from alembic import op
 from sqlalchemy.dialects import postgresql
 
 # revision identifiers, used by Alembic.
-revision = "bb249ed79907"
+revision = "426b6bdc3360"
 down_revision = None
 branch_labels = None
 depends_on = None
@@ -23,12 +23,21 @@ def upgrade() -> None:
         sa.Column("url", sa.String(length=2048), nullable=True),
         sa.Column(
             "status",
-            sa.Enum("create", "error", "processing", "success", name="jobstatus"),
+            sa.Enum("create", "processing", "error", "success", name="jobstatus"),
             nullable=False,
         ),
-        sa.Column("type", sa.Enum("transcript", name="jobtype"), nullable=False),
+        sa.Column("config", sa.JSON(none_as_null=True), nullable=True),
+        sa.Column("meta", sa.JSON(none_as_null=True), nullable=True),
         sa.Column(
-            "created_at", sa.DateTime(), server_default=sa.text("now()"), nullable=False
+            "type",
+            sa.Enum("transcript", "translation", "language_detection", name="jobtype"),
+            nullable=False,
+        ),
+        sa.Column(
+            "created_at",
+            sa.DateTime(),
+            server_default=sa.text("now()"),
+            nullable=False,
         ),
         sa.Column("updated_at", sa.DateTime(), nullable=True),
         sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),
@@ -40,10 +49,15 @@ def upgrade() -> None:
         sa.Column("job_id", postgresql.UUID(as_uuid=True), nullable=False),
         sa.Column("data", sa.JSON(none_as_null=True), nullable=True),
         sa.Column(
-            "type", sa.Enum("raw_transcript", name="artifacttype"), nullable=False
+            "type",
+            sa.Enum("raw_transcript", name="artifacttype"),
+            nullable=False,
         ),
         sa.Column(
-            "created_at", sa.DateTime(), server_default=sa.text("now()"), nullable=False
+            "created_at",
+            sa.DateTime(),
+            server_default=sa.text("now()"),
+            nullable=False,
         ),
         sa.Column("updated_at", sa.DateTime(), nullable=True),
         sa.Column("id", postgresql.UUID(as_uuid=True), nullable=False),

app/shared/db/alembic/versions/684a5e546314_add_job_meta_field.py

@@ -1,27 +0,0 @@
-"""add_job_meta_field
-
-Revision ID: 684a5e546314
-Revises: bb249ed79907
-Create Date: 2023-01-18 13:38:07.692830
-
-"""
-import sqlalchemy as sa
-from alembic import op
-
-# revision identifiers, used by Alembic.
-revision = "684a5e546314"
-down_revision = "bb249ed79907"
-branch_labels = None
-depends_on = None
-
-
-def upgrade() -> None:
-    # ### commands auto generated by Alembic - please adjust! ###
-    op.add_column("jobs", sa.Column("meta", sa.JSON(none_as_null=True), nullable=True))
-    # ### end Alembic commands ###
-
-
-def downgrade() -> None:
-    # ### commands auto generated by Alembic - please adjust! ###
-    op.drop_column("jobs", "meta")
-    # ### end Alembic commands ###

app/shared/db/dtos.py

@@ -1,6 +1,6 @@
 import enum
 from datetime import datetime
-from typing import Any, Optional
+from typing import Any, List, Optional
 from uuid import UUID
 
 from pydantic import AnyHttpUrl, BaseModel
@@ -21,6 +21,8 @@ class ArtifactType(str, enum.Enum):
 
 class JobType(str, enum.Enum):
     transcript = "transcript"
+    translation = "translation"
+    language_detection = "language_detection"
 
 
 class JobStatus(str, enum.Enum):
@@ -30,8 +32,12 @@ class JobStatus(str, enum.Enum):
     success = "success"
 
 
-class JobMeta(BaseModel):
+class JobConfig(BaseModel):
     language: Optional[str]
+
+
+class JobMeta(BaseModel):
+    error: Optional[str]
     task_id: Optional[UUID]
 
 
@@ -40,10 +46,11 @@ class Job(WithDbFields):
     type: JobType
     url: AnyHttpUrl
     meta: Optional[JobMeta]
+    config: Optional[JobConfig]
 
 
 class Artifact(WithDbFields):
     # TODO: narrow type
-    data: Optional[Any]
+    data: Optional[List[Any]]
     job_id: UUID
     type: ArtifactType

app/shared/db/models.py

@@ -35,6 +35,7 @@ class Job(Base, WithStandardFields):
 
     url = Column(String(length=2048))
     status = Column(Enum(JobStatus), nullable=False)
+    config = Column(JSON(none_as_null=True))
     meta = Column(JSON(none_as_null=True))
     type = Column(Enum(JobType), nullable=False)