chore: something changed, commit before reorg

alembic/versions/e6f7a8b9c0d1_add_project_workspaces_table.py

@@ -0,0 +1,244 @@
+"""add project_workspaces table
+
+Revision ID: e6f7a8b9c0d1
+Revises: d5e6f7a8b9c0
+Create Date: 2026-04-07 16:00:00.000000
+
+"""
+
+from __future__ import annotations
+
+import uuid
+from datetime import datetime, timezone
+from typing import Any, Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+
+revision: str = "e6f7a8b9c0d1"
+down_revision: Union[str, None] = "d5e6f7a8b9c0"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def _utc_now() -> datetime:
+    return datetime.now(timezone.utc)
+
+
+def _parse_uuid(raw_value: object) -> str | None:
+    if raw_value is None:
+        return None
+    try:
+        return str(uuid.UUID(str(raw_value)))
+    except (TypeError, ValueError):
+        return None
+
+
+def _default_state() -> dict[str, Any]:
+    return {
+        "version": 1,
+        "phase": "INGEST",
+        "active_job": None,
+        "source_file_id": None,
+        "workspace_view": {
+            "used_file_ids": [],
+            "selected_file_id": None,
+        },
+        "silence": {
+            "status": "IDLE",
+            "settings": {
+                "min_silence_duration_ms": 200,
+                "silence_threshold_db": 16,
+                "padding_ms": 100,
+            },
+            "detect_job_id": None,
+            "detected_segments": [],
+            "reviewed_cuts": [],
+            "duration_ms": None,
+            "applied_output_file_id": None,
+        },
+        "transcription": {
+            "status": "IDLE",
+            "request": {
+                "engine": "whisper",
+                "language": None,
+                "model": "base",
+            },
+            "job_id": None,
+            "artifact_id": None,
+            "transcription_id": None,
+            "reviewed": False,
+        },
+        "captions": {
+            "status": "IDLE",
+            "preset_id": None,
+            "style_config": None,
+            "render_job_id": None,
+            "output_file_id": None,
+        },
+    }
+
+
+def _backfill_state(legacy_workspace_state: dict | None) -> dict[str, Any]:
+    state = _default_state()
+    if not isinstance(legacy_workspace_state, dict):
+        return state
+
+    wizard = legacy_workspace_state.get("wizard")
+    if not isinstance(wizard, dict):
+        wizard = {}
+
+    current_step = wizard.get("current_step")
+    step_phase_map = {
+        "upload": "INGEST",
+        "verify": "VERIFY",
+        "silence-settings": "SILENCE",
+        "processing": "SILENCE",
+        "fragments": "SILENCE",
+        "silence-apply-processing": "SILENCE",
+        "transcription-settings": "TRANSCRIPTION",
+        "transcription-processing": "TRANSCRIPTION",
+        "subtitle-revision": "TRANSCRIPTION",
+        "caption-settings": "CAPTIONS",
+        "caption-processing": "CAPTIONS",
+        "caption-result": "DONE",
+    }
+    if current_step in step_phase_map:
+        state["phase"] = step_phase_map[current_step]
+
+    source_file_id = _parse_uuid(wizard.get("primary_file_id"))
+    if source_file_id is not None:
+        state["source_file_id"] = source_file_id
+
+    silence_job_id = _parse_uuid(wizard.get("silence_job_id"))
+    if silence_job_id is not None:
+        state["silence"]["detect_job_id"] = silence_job_id
+
+    transcription_artifact_id = _parse_uuid(wizard.get("transcription_artifact_id"))
+    if transcription_artifact_id is not None:
+        state["transcription"]["artifact_id"] = transcription_artifact_id
+
+    caption_preset_id = _parse_uuid(wizard.get("caption_preset_id"))
+    if caption_preset_id is not None:
+        state["captions"]["preset_id"] = caption_preset_id
+
+    caption_style_config = wizard.get("caption_style_config")
+    if isinstance(caption_style_config, dict):
+        state["captions"]["style_config"] = caption_style_config
+
+    captioned_video_file_id = _parse_uuid(wizard.get("captioned_video_file_id"))
+    if captioned_video_file_id is not None:
+        state["captions"]["output_file_id"] = captioned_video_file_id
+        state["captions"]["status"] = "COMPLETED"
+        state["phase"] = "DONE"
+
+    active_job_id = _parse_uuid(wizard.get("active_job_id"))
+    active_job_type = wizard.get("active_job_type")
+    if active_job_id is not None and isinstance(active_job_type, str):
+        state["active_job"] = {
+            "job_id": active_job_id,
+            "job_type": active_job_type,
+        }
+        if active_job_type == "TRANSCRIPTION_GENERATE":
+            state["transcription"]["job_id"] = active_job_id
+        if active_job_type == "CAPTIONS_GENERATE":
+            state["captions"]["render_job_id"] = active_job_id
+
+    silence_settings = wizard.get("silence_settings")
+    if isinstance(silence_settings, dict):
+        state["silence"]["settings"] = {
+            "min_silence_duration_ms": silence_settings.get("min_silence_duration_ms", 200),
+            "silence_threshold_db": silence_settings.get("silence_threshold_db", 16),
+            "padding_ms": silence_settings.get("padding_ms", 100),
+        }
+        state["silence"]["status"] = "CONFIGURED"
+
+    if current_step == "processing":
+        state["silence"]["status"] = "DETECTING"
+    elif current_step == "fragments":
+        state["silence"]["status"] = "REVIEWING"
+    elif current_step == "silence-apply-processing":
+        state["silence"]["status"] = "APPLYING"
+    elif current_step == "transcription-processing":
+        state["transcription"]["status"] = "PROCESSING"
+    elif current_step == "subtitle-revision":
+        state["transcription"]["status"] = "REVIEWING"
+    elif current_step == "caption-settings":
+        state["captions"]["status"] = "CONFIGURED"
+    elif current_step == "caption-processing":
+        state["captions"]["status"] = "PROCESSING"
+    elif current_step == "caption-result":
+        state["captions"]["status"] = "COMPLETED"
+
+    used_files = legacy_workspace_state.get("used_files")
+    if isinstance(used_files, list):
+        parsed_ids: list[str] = []
+        for item in used_files:
+            if not isinstance(item, dict):
+                continue
+            parsed_id = _parse_uuid(item.get("id"))
+            if parsed_id is not None and parsed_id not in parsed_ids:
+                parsed_ids.append(parsed_id)
+        state["workspace_view"]["used_file_ids"] = parsed_ids
+        if source_file_id in parsed_ids:
+            state["workspace_view"]["selected_file_id"] = source_file_id
+
+    return state
+
+
+def upgrade() -> None:
+    op.create_table(
+        "project_workspaces",
+        sa.Column(
+            "project_id",
+            postgresql.UUID(as_uuid=True),
+            sa.ForeignKey("projects.id", ondelete="CASCADE"),
+            primary_key=True,
+        ),
+        sa.Column("revision", sa.Integer(), nullable=False, server_default=sa.text("0")),
+        sa.Column("state", postgresql.JSONB(astext_type=sa.Text()), nullable=False),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
+        sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
+    )
+
+    connection = op.get_bind()
+    projects_table = sa.table(
+        "projects",
+        sa.column("id", postgresql.UUID(as_uuid=True)),
+        sa.column("workspace_state", sa.JSON()),
+        sa.column("is_active", sa.Boolean()),
+    )
+    workspaces_table = sa.table(
+        "project_workspaces",
+        sa.column("project_id", postgresql.UUID(as_uuid=True)),
+        sa.column("revision", sa.Integer()),
+        sa.column("state", postgresql.JSONB(astext_type=sa.Text())),
+        sa.column("created_at", sa.DateTime(timezone=True)),
+        sa.column("updated_at", sa.DateTime(timezone=True)),
+    )
+
+    rows = connection.execute(
+        sa.select(projects_table.c.id, projects_table.c.workspace_state).where(
+            projects_table.c.is_active.is_(True)
+        )
+    )
+
+    now = _utc_now()
+    payloads = [
+        {
+            "project_id": row.id,
+            "revision": 0,
+            "state": _backfill_state(row.workspace_state),
+            "created_at": now,
+            "updated_at": now,
+        }
+        for row in rows
+    ]
+    if payloads:
+        connection.execute(sa.insert(workspaces_table), payloads)
+
+
+def downgrade() -> None:
+    op.drop_table("project_workspaces")

cpv3/api/v1/router.py

@@ -10,6 +10,7 @@ from cpv3.modules.captions.router import router as captions_router
 from cpv3.modules.files.router import router as files_router
 from cpv3.modules.jobs.router import events_router, jobs_router
 from cpv3.modules.media.router import artifacts_router, media_router, mediafiles_router
+from cpv3.modules.project_workspaces.router import router as project_workspaces_router
 from cpv3.modules.projects.router import router as projects_router
 from cpv3.modules.system.router import router as system_router
 from cpv3.modules.tasks.router import router as tasks_router
@@ -29,6 +30,7 @@ api_router.include_router(users_router)
 
 # Projects
 api_router.include_router(projects_router)
+api_router.include_router(project_workspaces_router)
 
 # Files (storage module renamed)
 api_router.include_router(files_router)

cpv3/db/models.py

@@ -2,6 +2,7 @@ from cpv3.db.base import Base
 from cpv3.modules.captions.models import CaptionPreset
 from cpv3.modules.jobs.models import Job, JobEvent
 from cpv3.modules.media.models import ArtifactMediaFile, MediaFile
+from cpv3.modules.project_workspaces.models import ProjectWorkspace
 from cpv3.modules.projects.models import Project
 from cpv3.modules.files.models import File
 from cpv3.modules.transcription.models import Transcription
@@ -14,6 +15,7 @@ __all__ = [
     "CaptionPreset",
     "User",
     "Project",
+    "ProjectWorkspace",
     "File",
     "MediaFile",
     "ArtifactMediaFile",

cpv3/modules/project_workspaces/__init__.py

@@ -0,0 +1 @@
+"""Typed project workspace module."""

cpv3/modules/project_workspaces/models.py

@@ -0,0 +1,30 @@
+from __future__ import annotations
+
+import uuid
+from datetime import datetime
+
+from sqlalchemy import DateTime, ForeignKey, Integer, JSON
+from sqlalchemy.dialects.postgresql import JSONB, UUID
+from sqlalchemy.orm import Mapped, mapped_column
+
+from cpv3.db.base import Base, utcnow
+
+STATE_JSON_TYPE = JSON().with_variant(JSONB(), "postgresql")
+
+
+class ProjectWorkspace(Base):
+    __tablename__ = "project_workspaces"
+
+    project_id: Mapped[uuid.UUID] = mapped_column(
+        UUID(as_uuid=True),
+        ForeignKey("projects.id", ondelete="CASCADE"),
+        primary_key=True,
+    )
+    revision: Mapped[int] = mapped_column(Integer, default=0, nullable=False)
+    state: Mapped[dict] = mapped_column(STATE_JSON_TYPE, default=dict, nullable=False)
+    created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), default=utcnow)
+    updated_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True),
+        default=utcnow,
+        onupdate=utcnow,
+    )

cpv3/modules/project_workspaces/repository.py

@@ -0,0 +1,74 @@
+from __future__ import annotations
+
+import uuid
+
+from sqlalchemy import select, update
+from sqlalchemy.exc import IntegrityError
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from cpv3.db.base import utcnow
+from cpv3.modules.project_workspaces.models import ProjectWorkspace
+
+
+class WorkspaceRevisionConflictError(RuntimeError):
+    """Raised when the optimistic workspace revision check fails."""
+
+
+class ProjectWorkspaceRepository:
+    def __init__(self, session: AsyncSession) -> None:
+        self._session = session
+
+    async def get_by_project_id(self, project_id: uuid.UUID) -> ProjectWorkspace | None:
+        result = await self._session.execute(
+            select(ProjectWorkspace).where(ProjectWorkspace.project_id == project_id)
+        )
+        return result.scalar_one_or_none()
+
+    async def create(self, *, project_id: uuid.UUID, state: dict) -> ProjectWorkspace:
+        workspace = ProjectWorkspace(project_id=project_id, revision=0, state=state)
+        self._session.add(workspace)
+        await self._session.commit()
+        await self._session.refresh(workspace)
+        return workspace
+
+    async def get_or_create(self, *, project_id: uuid.UUID, state: dict) -> ProjectWorkspace:
+        workspace = await self.get_by_project_id(project_id)
+        if workspace is not None:
+            return workspace
+
+        try:
+            return await self.create(project_id=project_id, state=state)
+        except IntegrityError:
+            await self._session.rollback()
+            workspace = await self.get_by_project_id(project_id)
+            if workspace is None:
+                raise
+            return workspace
+
+    async def update_state(
+        self,
+        *,
+        project_id: uuid.UUID,
+        expected_revision: int,
+        state: dict,
+    ) -> ProjectWorkspace:
+        stmt = (
+            update(ProjectWorkspace)
+            .where(ProjectWorkspace.project_id == project_id)
+            .where(ProjectWorkspace.revision == expected_revision)
+            .values(
+                state=state,
+                revision=expected_revision + 1,
+                updated_at=utcnow(),
+            )
+        )
+        result = await self._session.execute(stmt)
+        if result.rowcount != 1:
+            await self._session.rollback()
+            raise WorkspaceRevisionConflictError
+
+        await self._session.commit()
+        workspace = await self.get_by_project_id(project_id)
+        if workspace is None:
+            raise RuntimeError("Workspace disappeared after update")
+        return workspace

cpv3/modules/project_workspaces/router.py

@@ -0,0 +1,78 @@
+from __future__ import annotations
+
+import uuid
+
+from fastapi import APIRouter, Depends, HTTPException, status
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from cpv3.db.session import get_db
+from cpv3.infrastructure.auth import get_current_user
+from cpv3.modules.projects.service import ProjectService
+from cpv3.modules.project_workspaces.schemas import (
+    ProjectWorkspaceRead,
+    WorkflowActionRequest,
+)
+from cpv3.modules.project_workspaces.service import (
+    ProjectWorkspaceRevisionConflictError,
+    ProjectWorkspaceService,
+    ProjectWorkflowValidationError,
+)
+from cpv3.modules.users.models import User
+
+router = APIRouter(prefix="/api/projects", tags=["Project Workspaces"])
+
+
+@router.get("/{project_id}/workspace", response_model=ProjectWorkspaceRead)
+@router.get(
+    "/{project_id}/workspace/",
+    response_model=ProjectWorkspaceRead,
+    include_in_schema=False,
+)
+async def get_project_workspace(
+    project_id: uuid.UUID,
+    current_user: User = Depends(get_current_user),
+    db: AsyncSession = Depends(get_db),
+) -> ProjectWorkspaceRead:
+    project_service = ProjectService(db)
+    project = await project_service.get_project(project_id)
+    if project is None:
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Не найдено")
+
+    if not current_user.is_staff and project.owner_id != current_user.id:
+        raise HTTPException(status_code=status.HTTP_403_FORBIDDEN, detail="Доступ запрещён")
+
+    workspace_service = ProjectWorkspaceService(db)
+    return await workspace_service.get_workspace(project=project)
+
+
+@router.post("/{project_id}/workflow/actions", response_model=ProjectWorkspaceRead)
+@router.post(
+    "/{project_id}/workflow/actions/",
+    response_model=ProjectWorkspaceRead,
+    include_in_schema=False,
+)
+async def dispatch_project_workflow_action(
+    project_id: uuid.UUID,
+    body: WorkflowActionRequest,
+    current_user: User = Depends(get_current_user),
+    db: AsyncSession = Depends(get_db),
+) -> ProjectWorkspaceRead:
+    project_service = ProjectService(db)
+    project = await project_service.get_project(project_id)
+    if project is None:
+        raise HTTPException(status_code=status.HTTP_404_NOT_FOUND, detail="Не найдено")
+
+    if not current_user.is_staff and project.owner_id != current_user.id:
+        raise HTTPException(status_code=status.HTTP_403_FORBIDDEN, detail="Доступ запрещён")
+
+    workspace_service = ProjectWorkspaceService(db)
+    try:
+        return await workspace_service.apply_action(
+            project=project,
+            requester=current_user,
+            action=body,
+        )
+    except ProjectWorkspaceRevisionConflictError as exc:
+        raise HTTPException(status_code=status.HTTP_409_CONFLICT, detail=str(exc)) from exc
+    except ProjectWorkflowValidationError as exc:
+        raise HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail=str(exc)) from exc

cpv3/modules/project_workspaces/schemas.py

@@ -0,0 +1,471 @@
+from __future__ import annotations
+
+from enum import StrEnum
+from typing import Annotated, Literal, get_args
+from uuid import UUID
+
+from pydantic import AliasChoices, Field, model_validator
+
+from cpv3.common.schemas import Schema
+from cpv3.modules.jobs.schemas import JobTypeEnum
+
+
+WORKFLOW_VERSION = 1
+VALID_JOB_TYPES = set(get_args(JobTypeEnum))
+
+WorkspaceScreenEnum = Literal[
+    "upload",
+    "verify",
+    "silence-settings",
+    "processing",
+    "fragments",
+    "silence-apply-processing",
+    "transcription-settings",
+    "transcription-processing",
+    "subtitle-revision",
+    "caption-settings",
+    "caption-processing",
+    "caption-result",
+]
+
+
+class WorkflowPhase(StrEnum):
+    INGEST = "INGEST"
+    VERIFY = "VERIFY"
+    SILENCE = "SILENCE"
+    TRANSCRIPTION = "TRANSCRIPTION"
+    CAPTIONS = "CAPTIONS"
+    DONE = "DONE"
+
+
+class SilenceWorkflowStatus(StrEnum):
+    IDLE = "IDLE"
+    CONFIGURED = "CONFIGURED"
+    DETECTING = "DETECTING"
+    REVIEWING = "REVIEWING"
+    APPLYING = "APPLYING"
+    COMPLETED = "COMPLETED"
+    SKIPPED = "SKIPPED"
+
+
+class TranscriptionWorkflowStatus(StrEnum):
+    IDLE = "IDLE"
+    PROCESSING = "PROCESSING"
+    REVIEWING = "REVIEWING"
+    COMPLETED = "COMPLETED"
+
+
+class CaptionsWorkflowStatus(StrEnum):
+    IDLE = "IDLE"
+    CONFIGURED = "CONFIGURED"
+    PROCESSING = "PROCESSING"
+    COMPLETED = "COMPLETED"
+
+
+class ActiveJobState(Schema):
+    job_id: UUID
+    job_type: JobTypeEnum
+
+
+class WorkspaceViewState(Schema):
+    used_file_ids: list[UUID] = Field(default_factory=list)
+    selected_file_id: UUID | None = None
+
+
+class SilenceSettingsState(Schema):
+    min_silence_duration_ms: int = 200
+    silence_threshold_db: int = 16
+    padding_ms: int = 100
+
+
+class CutRegionState(Schema):
+    start_ms: int
+    end_ms: int
+
+
+class SilenceState(Schema):
+    status: SilenceWorkflowStatus = SilenceWorkflowStatus.IDLE
+    settings: SilenceSettingsState = Field(default_factory=SilenceSettingsState)
+    detect_job_id: UUID | None = None
+    detected_segments: list[CutRegionState] = Field(default_factory=list)
+    reviewed_cuts: list[CutRegionState] = Field(
+        default_factory=list,
+        validation_alias=AliasChoices("reviewed_cuts", "cut_regions"),
+        serialization_alias="reviewed_cuts",
+    )
+    duration_ms: int | None = None
+    applied_output_file_id: UUID | None = Field(
+        default=None,
+        validation_alias=AliasChoices("applied_output_file_id", "output_file_id"),
+        serialization_alias="applied_output_file_id",
+    )
+
+
+class TranscriptionRequestState(Schema):
+    engine: Literal["whisper", "google", "salutespeech"] = "whisper"
+    language: str | None = None
+    model: str = "base"
+
+
+class TranscriptionState(Schema):
+    status: TranscriptionWorkflowStatus = TranscriptionWorkflowStatus.IDLE
+    request: TranscriptionRequestState = Field(default_factory=TranscriptionRequestState)
+    job_id: UUID | None = None
+    artifact_id: UUID | None = None
+    transcription_id: UUID | None = None
+    reviewed: bool = False
+
+
+class CaptionsState(Schema):
+    status: CaptionsWorkflowStatus = CaptionsWorkflowStatus.IDLE
+    preset_id: UUID | None = None
+    style_config: dict | None = None
+    render_job_id: UUID | None = Field(
+        default=None,
+        validation_alias=AliasChoices("render_job_id", "job_id"),
+        serialization_alias="render_job_id",
+    )
+    output_file_id: UUID | None = None
+
+
+class ProjectWorkspaceState(Schema):
+    version: int = WORKFLOW_VERSION
+    phase: WorkflowPhase = WorkflowPhase.INGEST
+    active_job: ActiveJobState | None = None
+    source_file_id: UUID | None = None
+    workspace_view: WorkspaceViewState = Field(default_factory=WorkspaceViewState)
+    silence: SilenceState = Field(default_factory=SilenceState)
+    transcription: TranscriptionState = Field(default_factory=TranscriptionState)
+    captions: CaptionsState = Field(default_factory=CaptionsState)
+
+
+class ProjectWorkspaceRead(Schema):
+    project_id: UUID
+    revision: int
+    version: int
+    phase: WorkflowPhase
+    current_screen: WorkspaceScreenEnum
+    active_job: ActiveJobState | None
+    source_file_id: UUID | None
+    workspace_view: WorkspaceViewState
+    silence: SilenceState
+    transcription: TranscriptionState
+    captions: CaptionsState
+
+
+class WorkflowActionBase(Schema):
+    type: str
+    revision: int
+
+
+class SetSourceFileAction(WorkflowActionBase):
+    type: Literal["SET_SOURCE_FILE"]
+    file_id: UUID = Field(
+        validation_alias=AliasChoices("file_id", "source_file_id"),
+        serialization_alias="file_id",
+    )
+
+
+class ResetSourceFileAction(WorkflowActionBase):
+    type: Literal["RESET_SOURCE_FILE"]
+
+
+class StartMediaConvertAction(WorkflowActionBase):
+    type: Literal["START_MEDIA_CONVERT"]
+    output_format: str = "mp4"
+    out_folder: str = "output_files"
+
+
+class ConfirmVerifyAction(WorkflowActionBase):
+    type: Literal["CONFIRM_VERIFY"]
+
+
+class SetSilenceSettingsAction(WorkflowActionBase):
+    type: Literal["SET_SILENCE_SETTINGS"]
+    settings: SilenceSettingsState = Field(default_factory=SilenceSettingsState)
+
+    @model_validator(mode="before")
+    @classmethod
+    def normalize_settings(cls, data: object) -> object:
+        if not isinstance(data, dict) or "settings" in data:
+            return data
+
+        return {
+            **data,
+            "settings": {
+                "min_silence_duration_ms": data.get("min_silence_duration_ms", 200),
+                "silence_threshold_db": data.get("silence_threshold_db", 16),
+                "padding_ms": data.get("padding_ms", 100),
+            },
+        }
+
+
+class StartSilenceDetectAction(WorkflowActionBase):
+    type: Literal["START_SILENCE_DETECT"]
+
+
+class SetSilenceCutsAction(WorkflowActionBase):
+    type: Literal["SET_SILENCE_CUTS"]
+    cuts: list[CutRegionState] = Field(
+        validation_alias=AliasChoices("cuts", "reviewed_cuts", "cut_regions"),
+    )
+
+
+class SkipSilenceApplyAction(WorkflowActionBase):
+    type: Literal["SKIP_SILENCE_APPLY"]
+
+
+class StartSilenceApplyAction(WorkflowActionBase):
+    type: Literal["START_SILENCE_APPLY"]
+    cuts: list[CutRegionState] | None = None
+    out_folder: str = "output_files"
+    output_name: str | None = None
+
+
+class ReopenSilenceReviewAction(WorkflowActionBase):
+    type: Literal["REOPEN_SILENCE_REVIEW"]
+
+
+class StartTranscriptionAction(WorkflowActionBase):
+    type: Literal["START_TRANSCRIPTION"]
+    engine: Literal["whisper", "google", "salutespeech"] = "whisper"
+    language: str | None = None
+    model: str = "base"
+    request: TranscriptionRequestState | None = None
+
+    @model_validator(mode="after")
+    def normalize_request(self) -> "StartTranscriptionAction":
+        if self.request is None:
+            self.request = TranscriptionRequestState(
+                engine=self.engine,
+                language=self.language,
+                model=self.model,
+            )
+            return self
+
+        self.engine = self.request.engine
+        self.language = self.request.language
+        self.model = self.request.model
+        return self
+
+
+class ReopenTranscriptionConfigAction(WorkflowActionBase):
+    type: Literal["REOPEN_TRANSCRIPTION_CONFIG"]
+
+
+class MarkTranscriptionReviewedAction(WorkflowActionBase):
+    type: Literal["MARK_TRANSCRIPTION_REVIEWED"]
+
+
+class SelectCaptionPresetAction(WorkflowActionBase):
+    type: Literal["SELECT_CAPTION_PRESET"]
+    preset_id: UUID | None = None
+    style_config: dict | None = None
+
+
+class StartCaptionRenderAction(WorkflowActionBase):
+    type: Literal["START_CAPTION_RENDER"]
+    folder: str = "output_files"
+
+
+class ReopenCaptionConfigAction(WorkflowActionBase):
+    type: Literal["REOPEN_CAPTION_CONFIG"]
+
+
+class SetWorkspaceViewAction(WorkflowActionBase):
+    type: Literal["SET_WORKSPACE_VIEW"]
+    workspace_view: WorkspaceViewState
+
+    @model_validator(mode="before")
+    @classmethod
+    def normalize_workspace_view(cls, data: object) -> object:
+        if not isinstance(data, dict) or "workspace_view" in data:
+            return data
+
+        return {
+            **data,
+            "workspace_view": {
+                "used_file_ids": data.get("used_file_ids", []),
+                "selected_file_id": data.get("selected_file_id"),
+            },
+        }
+
+
+WorkflowActionRequest = Annotated[
+    (
+        SetSourceFileAction
+        | ResetSourceFileAction
+        | StartMediaConvertAction
+        | ConfirmVerifyAction
+        | SetSilenceSettingsAction
+        | StartSilenceDetectAction
+        | SetSilenceCutsAction
+        | SkipSilenceApplyAction
+        | StartSilenceApplyAction
+        | ReopenSilenceReviewAction
+        | StartTranscriptionAction
+        | ReopenTranscriptionConfigAction
+        | MarkTranscriptionReviewedAction
+        | SelectCaptionPresetAction
+        | StartCaptionRenderAction
+        | ReopenCaptionConfigAction
+        | SetWorkspaceViewAction
+    ),
+    Field(discriminator="type"),
+]
+
+
+def build_default_workspace_state() -> ProjectWorkspaceState:
+    return ProjectWorkspaceState()
+
+
+def build_workspace_state_from_legacy(
+    legacy_workspace_state: dict | None,
+) -> ProjectWorkspaceState:
+    state = build_default_workspace_state()
+    if not isinstance(legacy_workspace_state, dict):
+        return state
+
+    wizard = legacy_workspace_state.get("wizard")
+    if not isinstance(wizard, dict):
+        wizard = {}
+
+    source_file_id = _parse_uuid(wizard.get("primary_file_id"))
+    if source_file_id is not None:
+        state.source_file_id = source_file_id
+
+    used_file_ids: list[UUID] = []
+    used_files = legacy_workspace_state.get("used_files")
+    if isinstance(used_files, list):
+        for item in used_files:
+            if not isinstance(item, dict):
+                continue
+            file_id = _parse_uuid(item.get("id"))
+            if file_id is not None and file_id not in used_file_ids:
+                used_file_ids.append(file_id)
+
+    if source_file_id is not None and source_file_id not in used_file_ids:
+        used_file_ids.insert(0, source_file_id)
+
+    state.workspace_view.used_file_ids = used_file_ids
+    if source_file_id is not None and source_file_id in used_file_ids:
+        state.workspace_view.selected_file_id = source_file_id
+
+    active_job_id = _parse_uuid(wizard.get("active_job_id"))
+    active_job_type = wizard.get("active_job_type")
+    if active_job_id is not None and active_job_type in VALID_JOB_TYPES:
+        state.active_job = ActiveJobState(
+            job_id=active_job_id,
+            job_type=active_job_type,
+        )
+
+    silence_job_id = _parse_uuid(wizard.get("silence_job_id"))
+    if silence_job_id is not None:
+        state.silence.detect_job_id = silence_job_id
+
+    transcription_artifact_id = _parse_uuid(wizard.get("transcription_artifact_id"))
+    if transcription_artifact_id is not None:
+        state.transcription.artifact_id = transcription_artifact_id
+
+    caption_preset_id = _parse_uuid(wizard.get("caption_preset_id"))
+    if caption_preset_id is not None:
+        state.captions.preset_id = caption_preset_id
+
+    caption_style_config = wizard.get("caption_style_config")
+    if isinstance(caption_style_config, dict):
+        state.captions.style_config = caption_style_config
+
+    captioned_video_file_id = _parse_uuid(wizard.get("captioned_video_file_id"))
+    if captioned_video_file_id is not None:
+        state.captions.output_file_id = captioned_video_file_id
+
+    silence_settings = wizard.get("silence_settings")
+    if isinstance(silence_settings, dict):
+        state.silence.settings = SilenceSettingsState.model_validate(silence_settings)
+        state.silence.status = SilenceWorkflowStatus.CONFIGURED
+
+    current_step = wizard.get("current_step")
+    step_phase_map = {
+        "upload": WorkflowPhase.INGEST,
+        "verify": WorkflowPhase.VERIFY,
+        "silence-settings": WorkflowPhase.SILENCE,
+        "processing": WorkflowPhase.SILENCE,
+        "fragments": WorkflowPhase.SILENCE,
+        "silence-apply-processing": WorkflowPhase.SILENCE,
+        "transcription-settings": WorkflowPhase.TRANSCRIPTION,
+        "transcription-processing": WorkflowPhase.TRANSCRIPTION,
+        "subtitle-revision": WorkflowPhase.TRANSCRIPTION,
+        "caption-settings": WorkflowPhase.CAPTIONS,
+        "caption-processing": WorkflowPhase.CAPTIONS,
+        "caption-result": WorkflowPhase.DONE,
+    }
+    if current_step in step_phase_map:
+        state.phase = step_phase_map[current_step]
+
+    if current_step == "processing":
+        state.silence.status = SilenceWorkflowStatus.DETECTING
+    elif current_step == "fragments":
+        state.silence.status = SilenceWorkflowStatus.REVIEWING
+    elif current_step == "silence-apply-processing":
+        state.silence.status = SilenceWorkflowStatus.APPLYING
+    elif current_step == "transcription-processing":
+        state.transcription.status = TranscriptionWorkflowStatus.PROCESSING
+    elif current_step == "subtitle-revision":
+        state.transcription.status = TranscriptionWorkflowStatus.REVIEWING
+    elif current_step == "caption-settings":
+        state.captions.status = CaptionsWorkflowStatus.CONFIGURED
+    elif current_step == "caption-processing":
+        state.captions.status = CaptionsWorkflowStatus.PROCESSING
+    elif current_step == "caption-result":
+        state.captions.status = CaptionsWorkflowStatus.COMPLETED
+
+    if state.active_job is not None:
+        if state.active_job.job_type == "MEDIA_CONVERT":
+            state.phase = WorkflowPhase.VERIFY
+        elif state.active_job.job_type == "SILENCE_DETECT":
+            state.phase = WorkflowPhase.SILENCE
+            state.silence.status = SilenceWorkflowStatus.DETECTING
+            state.silence.detect_job_id = state.active_job.job_id
+        elif state.active_job.job_type == "SILENCE_APPLY":
+            state.phase = WorkflowPhase.SILENCE
+            state.silence.status = SilenceWorkflowStatus.APPLYING
+        elif state.active_job.job_type == "TRANSCRIPTION_GENERATE":
+            state.phase = WorkflowPhase.TRANSCRIPTION
+            state.transcription.status = TranscriptionWorkflowStatus.PROCESSING
+            state.transcription.job_id = state.active_job.job_id
+        elif state.active_job.job_type == "CAPTIONS_GENERATE":
+            state.phase = WorkflowPhase.CAPTIONS
+            state.captions.status = CaptionsWorkflowStatus.PROCESSING
+            state.captions.render_job_id = state.active_job.job_id
+
+    if captioned_video_file_id is not None:
+        state.phase = WorkflowPhase.DONE
+        state.captions.status = CaptionsWorkflowStatus.COMPLETED
+    elif transcription_artifact_id is not None and (
+        state.transcription.status == TranscriptionWorkflowStatus.IDLE
+    ):
+        state.phase = WorkflowPhase.TRANSCRIPTION
+        state.transcription.status = TranscriptionWorkflowStatus.REVIEWING
+    elif silence_job_id is not None and state.silence.status == SilenceWorkflowStatus.IDLE:
+        state.phase = WorkflowPhase.SILENCE
+        state.silence.status = SilenceWorkflowStatus.REVIEWING
+    elif source_file_id is not None and state.phase == WorkflowPhase.INGEST:
+        state.phase = WorkflowPhase.VERIFY
+
+    return state
+
+
+def _parse_uuid(value: object) -> UUID | None:
+    if value is None:
+        return None
+    try:
+        return UUID(str(value))
+    except (TypeError, ValueError):
+        return None
+
+
+# Backward-compatible aliases used by existing tests and frontend hand-written types.
+TaskWorkflowActiveJob = ActiveJobState
+SilenceSettingsPayload = SilenceSettingsState
+WorkflowSilenceState = SilenceState
+WorkflowTranscriptionRequest = TranscriptionRequestState

cpv3/modules/project_workspaces/service.py

@@ -0,0 +1,824 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from uuid import UUID
+
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from cpv3.modules.captions.models import CaptionPreset
+from cpv3.modules.captions.repository import CaptionPresetRepository
+from cpv3.modules.files.models import File
+from cpv3.modules.files.repository import FileRepository
+from cpv3.modules.jobs.models import Job
+from cpv3.modules.media.repository import ArtifactRepository
+from cpv3.modules.project_workspaces.models import ProjectWorkspace
+from cpv3.modules.project_workspaces.repository import (
+    ProjectWorkspaceRepository,
+    WorkspaceRevisionConflictError as RepositoryWorkspaceRevisionConflictError,
+)
+from cpv3.modules.project_workspaces.schemas import (
+    ActiveJobState,
+    CaptionsState,
+    CaptionsWorkflowStatus,
+    ConfirmVerifyAction,
+    CutRegionState,
+    MarkTranscriptionReviewedAction,
+    ProjectWorkspaceRead,
+    ProjectWorkspaceState,
+    ReopenCaptionConfigAction,
+    ReopenSilenceReviewAction,
+    ReopenTranscriptionConfigAction,
+    ResetSourceFileAction,
+    SelectCaptionPresetAction,
+    SetSilenceCutsAction,
+    SetSilenceSettingsAction,
+    SetSourceFileAction,
+    SetWorkspaceViewAction,
+    SilenceState,
+    SilenceWorkflowStatus,
+    SkipSilenceApplyAction,
+    StartCaptionRenderAction,
+    StartMediaConvertAction,
+    StartSilenceApplyAction,
+    StartSilenceDetectAction,
+    StartTranscriptionAction,
+    TranscriptionState,
+    TranscriptionWorkflowStatus,
+    WorkflowActionRequest,
+    WorkflowPhase,
+    WorkspaceScreenEnum,
+    WorkspaceViewState,
+    build_workspace_state_from_legacy,
+)
+from cpv3.modules.projects.models import Project
+from cpv3.modules.projects.repository import ProjectRepository
+from cpv3.modules.tasks.schemas import (
+    CaptionsGenerateRequest,
+    MediaConvertRequest,
+    SilenceApplyRequest,
+    SilenceDetectRequest,
+    TranscriptionGenerateRequest,
+)
+from cpv3.modules.transcription.repository import TranscriptionRepository
+from cpv3.modules.users.models import User
+
+
+class ProjectWorkspaceRevisionConflictError(RuntimeError):
+    pass
+
+
+class ProjectWorkflowValidationError(ValueError):
+    pass
+
+
+WorkspaceRevisionConflictError = ProjectWorkspaceRevisionConflictError
+
+
+@dataclass(slots=True)
+class _WorkspaceItemValidationResult:
+    item_id: UUID
+    file: File | None = None
+
+
+class ProjectWorkspaceService:
+    def __init__(self, session: AsyncSession) -> None:
+        self._session = session
+        self._repo = ProjectWorkspaceRepository(session)
+        self._project_repo = ProjectRepository(session)
+        self._file_repo = FileRepository(session)
+        self._artifact_repo = ArtifactRepository(session)
+        self._transcription_repo = TranscriptionRepository(session)
+        self._caption_preset_repo = CaptionPresetRepository(session)
+        self._task_service_factory = self._build_task_service
+
+    async def create_for_project(self, project: Project) -> ProjectWorkspaceRead:
+        workspace = await self._get_or_create_workspace(project=project)
+        return self._to_read_model(workspace)
+
+    async def get_workspace(self, *, project: Project) -> ProjectWorkspaceRead:
+        workspace = await self._get_or_create_workspace(project=project)
+        return self._to_read_model(workspace)
+
+    async def apply_action(
+        self,
+        *,
+        project: Project,
+        requester: User,
+        action: WorkflowActionRequest,
+    ) -> ProjectWorkspaceRead:
+        workspace = await self._get_or_create_workspace(project=project)
+        if workspace.revision != action.revision:
+            raise ProjectWorkspaceRevisionConflictError("Версия рабочего пространства устарела")
+
+        state = ProjectWorkspaceState.model_validate(workspace.state)
+        next_state = await self._apply_action_to_state(
+            project=project,
+            requester=requester,
+            state=state,
+            action=action,
+        )
+        workspace = await self._save_workspace_state(
+            project_id=project.id,
+            expected_revision=workspace.revision,
+            state=next_state,
+        )
+        return self._to_read_model(workspace)
+
+    async def handle_job_update(self, *, job: Job) -> ProjectWorkspaceRead | None:
+        if job.project_id is None:
+            return None
+
+        project = await self._project_repo.get_by_id(job.project_id)
+        if project is None:
+            return None
+
+        workspace = await self._get_or_create_workspace(project=project)
+        state = ProjectWorkspaceState.model_validate(workspace.state)
+        next_state = self._apply_job_event_to_state(state, job)
+
+        if next_state.model_dump(mode="json") == state.model_dump(mode="json"):
+            return self._to_read_model(workspace)
+
+        try:
+            workspace = await self._save_workspace_state(
+                project_id=project.id,
+                expected_revision=workspace.revision,
+                state=next_state,
+            )
+        except ProjectWorkspaceRevisionConflictError:
+            workspace = await self._get_or_create_workspace(project=project)
+
+        return self._to_read_model(workspace)
+
+    async def apply_job_update(
+        self,
+        *,
+        project: Project,
+        job: Job,
+    ) -> ProjectWorkspaceRead | None:
+        workspace = await self._get_or_create_workspace(project=project)
+        state = ProjectWorkspaceState.model_validate(workspace.state)
+        next_state = self._apply_job_event_to_state(state, job)
+
+        if next_state.model_dump(mode="json") == state.model_dump(mode="json"):
+            return self._to_read_model(workspace)
+
+        try:
+            workspace = await self._save_workspace_state(
+                project_id=project.id,
+                expected_revision=workspace.revision,
+                state=next_state,
+            )
+        except ProjectWorkspaceRevisionConflictError:
+            workspace = await self._get_or_create_workspace(project=project)
+
+        return self._to_read_model(workspace)
+
+    async def apply_job_event(self, job: Job) -> None:
+        await self.handle_job_update(job=job)
+
+    async def _get_or_create_workspace(self, *, project: Project) -> ProjectWorkspace:
+        workspace = await self._repo.get_by_project_id(project.id)
+        if workspace is not None:
+            return workspace
+
+        initial_state = build_workspace_state_from_legacy(getattr(project, "workspace_state", None))
+        state_payload = initial_state.model_dump(mode="json")
+        get_or_create = getattr(self._repo, "get_or_create", None)
+        if callable(get_or_create):
+            return await get_or_create(project_id=project.id, state=state_payload)
+        return await self._repo.create(project_id=project.id, state=state_payload)
+
+    async def _save_workspace_state(
+        self,
+        *,
+        project_id: UUID,
+        expected_revision: int,
+        state: ProjectWorkspaceState,
+    ) -> ProjectWorkspace:
+        try:
+            return await self._repo.update_state(
+                project_id=project_id,
+                expected_revision=expected_revision,
+                state=state.model_dump(mode="json"),
+            )
+        except RepositoryWorkspaceRevisionConflictError as exc:
+            raise ProjectWorkspaceRevisionConflictError(
+                "Версия рабочего пространства устарела"
+            ) from exc
+
+    async def _apply_action_to_state(
+        self,
+        *,
+        project: Project,
+        requester: User,
+        state: ProjectWorkspaceState,
+        action: WorkflowActionRequest,
+    ) -> ProjectWorkspaceState:
+        next_state = state.model_copy(deep=True)
+
+        if isinstance(action, SetSourceFileAction):
+            file_result = await self._validate_accessible_file(
+                requester=requester,
+                project=project,
+                file_id=action.file_id,
+            )
+            next_state.phase = WorkflowPhase.VERIFY
+            next_state.active_job = None
+            next_state.source_file_id = file_result.item_id
+            next_state.silence = SilenceState()
+            next_state.transcription = TranscriptionState()
+            next_state.captions = CaptionsState()
+            next_state.workspace_view = WorkspaceViewState(
+                used_file_ids=[file_result.item_id],
+                selected_file_id=file_result.item_id,
+            )
+            return next_state
+
+        if isinstance(action, ResetSourceFileAction):
+            return ProjectWorkspaceState(version=state.version)
+
+        if isinstance(action, StartMediaConvertAction):
+            self._require_phase(next_state, WorkflowPhase.VERIFY)
+            source_file = await self._require_source_file(
+                next_state,
+                project=project,
+                requester=requester,
+            )
+            task_service = self._task_service_factory()
+            response = await task_service.submit_media_convert(
+                requester=requester,
+                request=MediaConvertRequest(
+                    file_key=source_file.path,
+                    out_folder=action.out_folder,
+                    output_format=action.output_format,
+                    project_id=project.id,
+                ),
+            )
+            next_state.active_job = ActiveJobState(
+                job_id=response.job_id,
+                job_type="MEDIA_CONVERT",
+            )
+            return next_state
+
+        if isinstance(action, ConfirmVerifyAction):
+            self._require_phase(next_state, WorkflowPhase.VERIFY)
+            await self._require_source_file(
+                next_state,
+                project=project,
+                requester=requester,
+            )
+            next_state.phase = WorkflowPhase.SILENCE
+            next_state.active_job = None
+            next_state.silence.status = SilenceWorkflowStatus.CONFIGURED
+            return next_state
+
+        if isinstance(action, SetSilenceSettingsAction):
+            self._require_phase(next_state, WorkflowPhase.SILENCE)
+            if next_state.silence.status in {
+                SilenceWorkflowStatus.DETECTING,
+                SilenceWorkflowStatus.APPLYING,
+            }:
+                raise ProjectWorkflowValidationError("Нельзя менять настройки во время обработки")
+            next_state.silence.settings = action.settings
+            next_state.silence.status = SilenceWorkflowStatus.CONFIGURED
+            return next_state
+
+        if isinstance(action, StartSilenceDetectAction):
+            self._require_phase(next_state, WorkflowPhase.SILENCE)
+            source_file = await self._require_source_file(
+                next_state,
+                project=project,
+                requester=requester,
+            )
+            task_service = self._task_service_factory()
+            response = await task_service.submit_silence_detect(
+                requester=requester,
+                request=SilenceDetectRequest(
+                    file_key=source_file.path,
+                    project_id=project.id,
+                    min_silence_duration_ms=next_state.silence.settings.min_silence_duration_ms,
+                    silence_threshold_db=next_state.silence.settings.silence_threshold_db,
+                    padding_ms=next_state.silence.settings.padding_ms,
+                ),
+            )
+            next_state.active_job = ActiveJobState(
+                job_id=response.job_id,
+                job_type="SILENCE_DETECT",
+            )
+            next_state.silence.status = SilenceWorkflowStatus.DETECTING
+            next_state.silence.detect_job_id = response.job_id
+            next_state.silence.detected_segments = []
+            next_state.silence.reviewed_cuts = []
+            next_state.silence.duration_ms = None
+            next_state.silence.applied_output_file_id = None
+            return next_state
+
+        if isinstance(action, SetSilenceCutsAction):
+            self._require_phase(next_state, WorkflowPhase.SILENCE)
+            next_state.silence.reviewed_cuts = [
+                CutRegionState.model_validate(cut) for cut in action.cuts
+            ]
+            next_state.silence.status = SilenceWorkflowStatus.REVIEWING
+            return next_state
+
+        if isinstance(action, SkipSilenceApplyAction):
+            self._require_phase(next_state, WorkflowPhase.SILENCE)
+            next_state.phase = WorkflowPhase.TRANSCRIPTION
+            next_state.active_job = None
+            next_state.silence.status = SilenceWorkflowStatus.SKIPPED
+            return next_state
+
+        if isinstance(action, StartSilenceApplyAction):
+            self._require_phase(next_state, WorkflowPhase.SILENCE)
+            source_file = await self._require_source_file(
+                next_state,
+                project=project,
+                requester=requester,
+            )
+            if action.cuts is not None:
+                next_state.silence.reviewed_cuts = [
+                    CutRegionState.model_validate(cut) for cut in action.cuts
+                ]
+            if not next_state.silence.reviewed_cuts:
+                raise ProjectWorkflowValidationError("Нет выбранных фрагментов для применения")
+            task_service = self._task_service_factory()
+            response = await task_service.submit_silence_apply(
+                requester=requester,
+                request=SilenceApplyRequest(
+                    file_key=source_file.path,
+                    out_folder=action.out_folder,
+                    project_id=project.id,
+                    output_name=action.output_name,
+                    cuts=[cut.model_dump(mode="json") for cut in next_state.silence.reviewed_cuts],
+                ),
+            )
+            next_state.active_job = ActiveJobState(
+                job_id=response.job_id,
+                job_type="SILENCE_APPLY",
+            )
+            next_state.silence.status = SilenceWorkflowStatus.APPLYING
+            return next_state
+
+        if isinstance(action, ReopenSilenceReviewAction):
+            if not next_state.silence.detected_segments:
+                raise ProjectWorkflowValidationError("Нет результатов детекции тишины")
+            next_state.phase = WorkflowPhase.SILENCE
+            next_state.active_job = None
+            next_state.silence.status = SilenceWorkflowStatus.REVIEWING
+            return next_state
+
+        if isinstance(action, StartTranscriptionAction):
+            if next_state.phase not in {WorkflowPhase.SILENCE, WorkflowPhase.TRANSCRIPTION}:
+                raise ProjectWorkflowValidationError("Транскрибация пока недоступна")
+            transcription_file = await self._resolve_transcription_input_file(
+                next_state,
+                project=project,
+                requester=requester,
+            )
+            request_payload = action.request or TranscriptionGenerateRequest(
+                engine=action.engine,
+                language=action.language,
+                model=action.model,
+                file_key=transcription_file.path,
+                project_id=project.id,
+            )
+            task_service = self._task_service_factory()
+            response = await task_service.submit_transcription_generate(
+                requester=requester,
+                request=TranscriptionGenerateRequest(
+                    file_key=transcription_file.path,
+                    project_id=project.id,
+                    engine=request_payload.engine,
+                    language=request_payload.language,
+                    model=request_payload.model,
+                ),
+            )
+            next_state.phase = WorkflowPhase.TRANSCRIPTION
+            next_state.active_job = ActiveJobState(
+                job_id=response.job_id,
+                job_type="TRANSCRIPTION_GENERATE",
+            )
+            next_state.transcription.status = TranscriptionWorkflowStatus.PROCESSING
+            next_state.transcription.request = action.request or next_state.transcription.request
+            next_state.transcription.job_id = response.job_id
+            next_state.transcription.artifact_id = None
+            next_state.transcription.transcription_id = None
+            next_state.transcription.reviewed = False
+            next_state.captions = CaptionsState()
+            return next_state
+
+        if isinstance(action, ReopenTranscriptionConfigAction):
+            if next_state.phase not in {
+                WorkflowPhase.TRANSCRIPTION,
+                WorkflowPhase.CAPTIONS,
+                WorkflowPhase.DONE,
+            }:
+                raise ProjectWorkflowValidationError("Нельзя вернуться к настройкам транскрибации")
+            next_state.phase = WorkflowPhase.TRANSCRIPTION
+            next_state.active_job = None
+            next_state.transcription = TranscriptionState(
+                request=next_state.transcription.request,
+            )
+            next_state.captions = CaptionsState()
+            return next_state
+
+        if isinstance(action, MarkTranscriptionReviewedAction):
+            self._require_phase(next_state, WorkflowPhase.TRANSCRIPTION)
+            if next_state.transcription.transcription_id is None:
+                raise ProjectWorkflowValidationError("Сначала завершите транскрибацию")
+            next_state.transcription.reviewed = True
+            next_state.transcription.status = TranscriptionWorkflowStatus.COMPLETED
+            next_state.phase = WorkflowPhase.CAPTIONS
+            if next_state.captions.status == CaptionsWorkflowStatus.IDLE:
+                next_state.captions.status = CaptionsWorkflowStatus.CONFIGURED
+            return next_state
+
+        if isinstance(action, SelectCaptionPresetAction):
+            if next_state.phase not in {WorkflowPhase.CAPTIONS, WorkflowPhase.DONE}:
+                raise ProjectWorkflowValidationError("Сначала завершите транскрибацию")
+            preset = await self._validate_caption_preset(
+                requester=requester,
+                preset_id=action.preset_id,
+            )
+            next_state.phase = WorkflowPhase.CAPTIONS
+            next_state.active_job = None
+            next_state.captions.preset_id = action.preset_id
+            next_state.captions.style_config = (
+                action.style_config
+                if action.style_config is not None
+                else (preset.style_config if preset is not None else None)
+            )
+            next_state.captions.render_job_id = None
+            next_state.captions.output_file_id = None
+            next_state.captions.status = CaptionsWorkflowStatus.CONFIGURED
+            return next_state
+
+        if isinstance(action, StartCaptionRenderAction):
+            if next_state.phase not in {WorkflowPhase.CAPTIONS, WorkflowPhase.DONE}:
+                raise ProjectWorkflowValidationError("Рендер субтитров пока недоступен")
+            if next_state.transcription.transcription_id is None:
+                raise ProjectWorkflowValidationError("Сначала завершите транскрибацию")
+            video_file = await self._resolve_caption_video_file(
+                next_state,
+                project=project,
+                requester=requester,
+            )
+            transcription = await self._transcription_repo.get_by_id(
+                next_state.transcription.transcription_id
+            )
+            if transcription is None:
+                raise ProjectWorkflowValidationError("Транскрипция не найдена")
+
+            task_service = self._task_service_factory()
+            response = await task_service.submit_captions_generate(
+                requester=requester,
+                request=CaptionsGenerateRequest(
+                    video_s3_path=video_file.path,
+                    folder=action.folder,
+                    transcription_id=transcription.id,
+                    project_id=project.id,
+                    preset_id=next_state.captions.preset_id,
+                    style_config=next_state.captions.style_config,
+                ),
+            )
+            next_state.phase = WorkflowPhase.CAPTIONS
+            next_state.active_job = ActiveJobState(
+                job_id=response.job_id,
+                job_type="CAPTIONS_GENERATE",
+            )
+            next_state.captions.render_job_id = response.job_id
+            next_state.captions.output_file_id = None
+            next_state.captions.status = CaptionsWorkflowStatus.PROCESSING
+            return next_state
+
+        if isinstance(action, ReopenCaptionConfigAction):
+            if next_state.phase not in {WorkflowPhase.CAPTIONS, WorkflowPhase.DONE}:
+                raise ProjectWorkflowValidationError("Нельзя вернуться к настройкам рендера")
+            next_state.phase = WorkflowPhase.CAPTIONS
+            next_state.active_job = None
+            next_state.captions.render_job_id = None
+            next_state.captions.output_file_id = None
+            next_state.captions.status = CaptionsWorkflowStatus.CONFIGURED
+            return next_state
+
+        if isinstance(action, SetWorkspaceViewAction):
+            await self._validate_workspace_view_items(
+                requester=requester,
+                project=project,
+                used_file_ids=action.workspace_view.used_file_ids,
+                selected_file_id=action.workspace_view.selected_file_id,
+            )
+            next_state.workspace_view = action.workspace_view
+            return next_state
+
+        raise ProjectWorkflowValidationError("Неподдерживаемое действие")
+
+    def _apply_job_event_to_state(
+        self,
+        state: ProjectWorkspaceState,
+        job: Job,
+    ) -> ProjectWorkspaceState:
+        next_state = state.model_copy(deep=True)
+        output_data = job.output_data or {}
+
+        if next_state.active_job is not None and next_state.active_job.job_id == job.id:
+            next_state.active_job = None
+
+        if job.status in {"FAILED", "CANCELLED"}:
+            if job.job_type == "MEDIA_CONVERT":
+                next_state.phase = WorkflowPhase.VERIFY
+            elif job.job_type == "SILENCE_DETECT":
+                next_state.phase = WorkflowPhase.SILENCE
+                next_state.silence.status = SilenceWorkflowStatus.CONFIGURED
+                next_state.silence.detect_job_id = None
+            elif job.job_type == "SILENCE_APPLY":
+                next_state.phase = WorkflowPhase.SILENCE
+                next_state.silence.status = (
+                    SilenceWorkflowStatus.REVIEWING
+                    if next_state.silence.detected_segments
+                    else SilenceWorkflowStatus.CONFIGURED
+                )
+            elif job.job_type == "TRANSCRIPTION_GENERATE":
+                next_state.phase = WorkflowPhase.TRANSCRIPTION
+                next_state.transcription.status = TranscriptionWorkflowStatus.IDLE
+                next_state.transcription.job_id = None
+            elif job.job_type == "CAPTIONS_GENERATE":
+                next_state.phase = WorkflowPhase.CAPTIONS
+                next_state.captions.status = CaptionsWorkflowStatus.CONFIGURED
+                next_state.captions.render_job_id = None
+            return next_state
+
+        if job.status != "DONE":
+            return next_state
+
+        if job.job_type == "MEDIA_CONVERT":
+            converted_file_id = self._parse_uuid(output_data.get("file_id"))
+            if converted_file_id is None:
+                return next_state
+            next_state.source_file_id = converted_file_id
+            next_state.phase = WorkflowPhase.VERIFY
+            self._append_used_file(next_state, converted_file_id)
+            next_state.workspace_view.selected_file_id = converted_file_id
+            return next_state
+
+        if job.job_type == "SILENCE_DETECT":
+            silent_segments = output_data.get("silent_segments")
+            if not isinstance(silent_segments, list):
+                return next_state
+            cut_regions = [CutRegionState.model_validate(item) for item in silent_segments]
+            next_state.phase = WorkflowPhase.SILENCE
+            next_state.silence.detect_job_id = job.id
+            next_state.silence.detected_segments = cut_regions
+            next_state.silence.reviewed_cuts = cut_regions
+            next_state.silence.duration_ms = self._parse_int(output_data.get("duration_ms"))
+            next_state.silence.status = SilenceWorkflowStatus.REVIEWING
+            return next_state
+
+        if job.job_type == "SILENCE_APPLY":
+            output_file_id = self._parse_uuid(output_data.get("file_id"))
+            if output_file_id is None:
+                return next_state
+            next_state.phase = WorkflowPhase.TRANSCRIPTION
+            next_state.silence.applied_output_file_id = output_file_id
+            next_state.silence.status = SilenceWorkflowStatus.COMPLETED
+            self._append_used_file(next_state, output_file_id)
+            next_state.workspace_view.selected_file_id = output_file_id
+            return next_state
+
+        if job.job_type == "TRANSCRIPTION_GENERATE":
+            artifact_id = self._parse_uuid(output_data.get("artifact_id"))
+            transcription_id = self._parse_uuid(output_data.get("transcription_id"))
+            if artifact_id is None or transcription_id is None:
+                return next_state
+            next_state.phase = WorkflowPhase.TRANSCRIPTION
+            next_state.transcription.status = TranscriptionWorkflowStatus.REVIEWING
+            next_state.transcription.job_id = job.id
+            next_state.transcription.artifact_id = artifact_id
+            next_state.transcription.transcription_id = transcription_id
+            next_state.transcription.reviewed = False
+            return next_state
+
+        if job.job_type == "CAPTIONS_GENERATE":
+            output_file_id = self._parse_uuid(output_data.get("file_id"))
+            if output_file_id is None:
+                return next_state
+            next_state.phase = WorkflowPhase.DONE
+            next_state.captions.status = CaptionsWorkflowStatus.COMPLETED
+            next_state.captions.render_job_id = job.id
+            next_state.captions.output_file_id = output_file_id
+            self._append_used_file(next_state, output_file_id)
+            next_state.workspace_view.selected_file_id = output_file_id
+            return next_state
+
+        return next_state
+
+    def _to_read_model(self, workspace: ProjectWorkspace) -> ProjectWorkspaceRead:
+        state = ProjectWorkspaceState.model_validate(workspace.state)
+        return ProjectWorkspaceRead(
+            project_id=workspace.project_id,
+            revision=workspace.revision,
+            version=state.version,
+            phase=state.phase,
+            current_screen=self._derive_current_screen(state),
+            active_job=state.active_job,
+            source_file_id=state.source_file_id,
+            workspace_view=state.workspace_view,
+            silence=state.silence,
+            transcription=state.transcription,
+            captions=state.captions,
+        )
+
+    def _derive_current_screen(self, state: ProjectWorkspaceState) -> WorkspaceScreenEnum:
+        if state.phase == WorkflowPhase.INGEST:
+            return "upload"
+        if state.phase == WorkflowPhase.VERIFY:
+            return "verify"
+        if state.phase == WorkflowPhase.SILENCE:
+            if state.silence.status == SilenceWorkflowStatus.DETECTING:
+                return "processing"
+            if state.silence.status == SilenceWorkflowStatus.REVIEWING:
+                return "fragments"
+            if state.silence.status == SilenceWorkflowStatus.APPLYING:
+                return "silence-apply-processing"
+            return "silence-settings"
+        if state.phase == WorkflowPhase.TRANSCRIPTION:
+            if state.transcription.status == TranscriptionWorkflowStatus.PROCESSING:
+                return "transcription-processing"
+            if state.transcription.status in {
+                TranscriptionWorkflowStatus.REVIEWING,
+                TranscriptionWorkflowStatus.COMPLETED,
+            }:
+                return "subtitle-revision"
+            return "transcription-settings"
+        if state.phase == WorkflowPhase.CAPTIONS:
+            if state.captions.status == CaptionsWorkflowStatus.PROCESSING:
+                return "caption-processing"
+            if state.captions.status == CaptionsWorkflowStatus.COMPLETED:
+                return "caption-result"
+            return "caption-settings"
+        return "caption-result"
+
+    async def _validate_accessible_file(
+        self,
+        *,
+        requester: User,
+        project: Project,
+        file_id: UUID,
+    ) -> _WorkspaceItemValidationResult:
+        file = await self._file_repo.get_by_id(file_id)
+        if file is None:
+            raise ProjectWorkflowValidationError("Файл не найден")
+        if file.project_id not in {None, project.id}:
+            raise ProjectWorkflowValidationError("Файл не относится к текущему проекту")
+        if not requester.is_staff and file.owner_id not in {None, requester.id}:
+            raise ProjectWorkflowValidationError("Файл недоступен")
+        return _WorkspaceItemValidationResult(item_id=file.id, file=file)
+
+    async def _require_source_file(
+        self,
+        state: ProjectWorkspaceState,
+        *,
+        project: Project,
+        requester: User,
+    ) -> File:
+        if state.source_file_id is None:
+            raise ProjectWorkflowValidationError("Сначала выберите исходный файл")
+        validation = await self._validate_accessible_file(
+            requester=requester,
+            project=project,
+            file_id=state.source_file_id,
+        )
+        if validation.file is None:
+            raise ProjectWorkflowValidationError("Исходный файл не найден")
+        return validation.file
+
+    async def _resolve_transcription_input_file(
+        self,
+        state: ProjectWorkspaceState,
+        *,
+        project: Project,
+        requester: User,
+    ) -> File:
+        if state.silence.applied_output_file_id is not None:
+            validation = await self._validate_accessible_file(
+                requester=requester,
+                project=project,
+                file_id=state.silence.applied_output_file_id,
+            )
+            if validation.file is not None:
+                return validation.file
+        return await self._require_source_file(
+            state,
+            project=project,
+            requester=requester,
+        )
+
+    async def _resolve_caption_video_file(
+        self,
+        state: ProjectWorkspaceState,
+        *,
+        project: Project,
+        requester: User,
+    ) -> File:
+        return await self._resolve_transcription_input_file(
+            state,
+            project=project,
+            requester=requester,
+        )
+
+    async def _validate_caption_preset(
+        self,
+        *,
+        requester: User,
+        preset_id: UUID | None,
+    ) -> CaptionPreset | None:
+        if preset_id is None:
+            return None
+        preset = await self._caption_preset_repo.get_by_id(preset_id)
+        if preset is None:
+            raise ProjectWorkflowValidationError("Пресет субтитров не найден")
+        if not requester.is_staff and not preset.is_system and preset.user_id != requester.id:
+            raise ProjectWorkflowValidationError("Пресет субтитров недоступен")
+        return preset
+
+    async def _validate_workspace_view_items(
+        self,
+        *,
+        requester: User,
+        project: Project,
+        used_file_ids: list[UUID],
+        selected_file_id: UUID | None,
+    ) -> None:
+        if selected_file_id is not None and selected_file_id not in used_file_ids:
+            raise ProjectWorkflowValidationError(
+                "Выбранный файл должен входить в список используемых файлов"
+            )
+        seen: set[UUID] = set()
+        for item_id in used_file_ids:
+            if item_id in seen:
+                continue
+            seen.add(item_id)
+            await self._validate_workspace_item(
+                requester=requester,
+                project=project,
+                item_id=item_id,
+            )
+
+    async def _validate_workspace_item(
+        self,
+        *,
+        requester: User,
+        project: Project,
+        item_id: UUID,
+    ) -> None:
+        file = await self._file_repo.get_by_id(item_id)
+        if file is not None:
+            if file.project_id not in {None, project.id}:
+                raise ProjectWorkflowValidationError("Файл не относится к текущему проекту")
+            if not requester.is_staff and file.owner_id not in {None, requester.id}:
+                raise ProjectWorkflowValidationError("Файл недоступен")
+            return
+
+        artifact = await self._artifact_repo.get_by_id(item_id)
+        if artifact is None or artifact.project_id not in {None, project.id}:
+            raise ProjectWorkflowValidationError("Элемент рабочего пространства не найден")
+
+    def _build_task_service(self):
+        from cpv3.modules.tasks.service import TaskService
+
+        return TaskService(self._session)
+
+    def _require_phase(
+        self,
+        state: ProjectWorkspaceState,
+        required_phase: WorkflowPhase,
+    ) -> None:
+        if state.phase != required_phase:
+            raise ProjectWorkflowValidationError(
+                f"Ожидалась фаза {required_phase}, текущая фаза {state.phase}"
+            )
+
+    def _append_used_file(
+        self,
+        state: ProjectWorkspaceState,
+        file_id: UUID,
+    ) -> None:
+        if file_id not in state.workspace_view.used_file_ids:
+            state.workspace_view.used_file_ids.append(file_id)
+
+    def _parse_int(self, value: object) -> int | None:
+        if value is None:
+            return None
+        try:
+            return int(value)
+        except (TypeError, ValueError):
+            return None
+
+    def _parse_uuid(self, value: object) -> UUID | None:
+        if value is None:
+            return None
+        try:
+            return UUID(str(value))
+        except (TypeError, ValueError):
+            return None

cpv3/modules/projects/schemas.py

@@ -19,8 +19,6 @@ class ProjectRead(Schema):
     folder: str | None
     status: ProjectStatusEnum
 
-    workspace_state: dict | None
-
     is_active: bool
     created_at: datetime
     updated_at: datetime
@@ -38,4 +36,3 @@ class ProjectUpdate(Schema):
     language: str | None = None
     folder: str | None = None
     status: ProjectStatusEnum | None = None
-    workspace_state: dict | None = None

cpv3/modules/projects/service.py

@@ -14,6 +14,7 @@ class ProjectService:
     """Service for project business logic and orchestration."""
 
     def __init__(self, session: AsyncSession) -> None:
+        self._session = session
         self._repo = ProjectRepository(session)
 
     async def list_projects(
@@ -32,9 +33,13 @@ class ProjectService:
 
     async def create_project(self, *, requester: User, data: ProjectCreate) -> Project:
         folder = f"/{requester.username}/{data.name}"
-        return await self._repo.create(
+        project = await self._repo.create(
             requester=requester, data=data, folder=folder, status="DRAFT",
         )
+        from cpv3.modules.project_workspaces.service import ProjectWorkspaceService
+
+        await ProjectWorkspaceService(self._session).create_for_project(project)
+        return project
 
     async def update_project(self, project: Project, data: ProjectUpdate) -> Project:
         return await self._repo.update(project, data)

cpv3/modules/tasks/service.py

@@ -36,6 +36,7 @@ from cpv3.modules.jobs.schemas import (
 )
 from cpv3.modules.media.repository import ArtifactRepository
 from cpv3.modules.media.schemas import ArtifactMediaFileCreate
+from cpv3.modules.project_workspaces.service import ProjectWorkspaceService
 from cpv3.modules.tasks.schemas import (
     CaptionsGenerateRequest,
     FrameExtractRequest,
@@ -140,6 +141,7 @@ DRAMATIQ_BROKER_REF_SEPARATOR = ":"
 class JobCancelledError(RuntimeError):
     """Raised when a job was cancelled before completion."""
 
+
 # ---------------------------------------------------------------------------
 # Dramatiq broker setup
 # ---------------------------------------------------------------------------
@@ -808,10 +810,12 @@ def transcription_generate_actor(
                 )
             )
         elif engine == "salutespeech":
-            audio_stream = next(
-                (s for s in probe.streams if s.codec_type == "audio"), None
+            audio_stream = next((s for s in probe.streams if s.codec_type == "audio"), None)
+            sr = (
+                int(audio_stream.sample_rate)
+                if audio_stream and audio_stream.sample_rate
+                else 16000
             )
-            sr = int(audio_stream.sample_rate) if audio_stream and audio_stream.sample_rate else 16000
             document = _run_async(
                 transcribe_with_salute_speech(
                     storage,
@@ -1125,6 +1129,9 @@ class TaskService:
         self._event_repo = JobEventRepository(session)
         self._webhook_repo = WebhookRepository(session)
 
+    def _get_project_workspace_service(self):
+        return ProjectWorkspaceService(self._session)
+
     async def _update_job_broker_reference(self, job: Job, broker_reference: str) -> Job:
         """Persist the transport-specific broker reference after enqueueing."""
         job.broker_id = broker_reference
@@ -1310,28 +1317,33 @@ class TaskService:
         # Save artifacts BEFORE sending notifications so data exists when frontend refetches
         if job.job_type == JOB_TYPE_TRANSCRIPTION_GENERATE and event.status == JOB_STATUS_DONE:
             try:
-                await self._save_transcription_artifacts(job)
+                job = await self._save_transcription_artifacts(job)
             except Exception:
                 logger.exception("Failed to save transcription artifacts for job %s", job_id)
 
         if job.job_type == JOB_TYPE_MEDIA_CONVERT and event.status == JOB_STATUS_DONE:
             try:
-                await self._save_convert_artifacts(job)
+                job = await self._save_convert_artifacts(job)
             except Exception:
                 logger.exception("Failed to save convert artifacts for job %s", job_id)
 
         if job.job_type == JOB_TYPE_SILENCE_APPLY and event.status == JOB_STATUS_DONE:
             try:
-                await self._save_silence_apply_artifacts(job)
+                job = await self._save_silence_apply_artifacts(job)
             except Exception:
                 logger.exception("Failed to save silence apply artifacts for job %s", job_id)
 
         if job.job_type == JOB_TYPE_CAPTIONS_GENERATE and event.status == JOB_STATUS_DONE:
             try:
-                await self._save_captions_artifacts(job)
+                job = await self._save_captions_artifacts(job)
             except Exception:
                 logger.exception("Failed to save captions artifacts for job %s", job_id)
 
+        try:
+            await self._sync_project_workspace_after_webhook(job)
+        except Exception:
+            logger.exception("Failed to project workspace state for job %s", job_id)
+
         # Push real-time notification via WebSocket (after artifacts are persisted)
         if job.user_id is not None:
             try:
@@ -1344,6 +1356,11 @@ class TaskService:
 
         return job
 
+    async def _sync_project_workspace_after_webhook(self, job: Job) -> None:
+        if job.project_id is None:
+            return
+        await self._get_project_workspace_service().handle_job_update(job=job)
+
     async def cancel_job(self, job: Job) -> Job:
         """Cancel a job, clean queued transport state and ignore late webhooks."""
         if job.status in (JOB_STATUS_DONE, JOB_STATUS_FAILED, JOB_STATUS_CANCELLED):
@@ -1386,9 +1403,14 @@ class TaskService:
         except Exception:
             logger.exception("Failed to create cancellation notification for job %s", job.id)
 
+        try:
+            await self._sync_project_workspace_after_webhook(job)
+        except Exception:
+            logger.exception("Failed to project workspace state for cancelled job %s", job.id)
+
         return job
 
-    async def _save_transcription_artifacts(self, job: Job) -> None:
+    async def _save_transcription_artifacts(self, job: Job) -> Job:
         """Create Transcription, ArtifactMediaFile and File records."""
         input_data = job.input_data or {}
         output_data = job.output_data or {}
@@ -1407,7 +1429,7 @@ class TaskService:
         user = await user_repo.get_by_id(job.user_id)  # type: ignore[arg-type]
         if user is None:
             logger.warning("User %s not found, skipping artifact save", job.user_id)
-            return
+            return job
 
         # Find or create source File record
         file_repo = FileRepository(self._session)
@@ -1472,7 +1494,7 @@ class TaskService:
         # Create Transcription record
         transcription_repo = TranscriptionRepository(self._session)
         engine_db = ENGINE_MAP.get(engine_raw, "LOCAL_WHISPER")
-        await transcription_repo.create(
+        transcription = await transcription_repo.create(
             data=TranscriptionCreate(
                 project_id=project_id,
                 source_file_id=source_file.id,
@@ -1483,9 +1505,16 @@ class TaskService:
             ),
         )
 
-        logger.info("Saved transcription artifacts for job %s", job.id)
+        updated_output = dict(output_data)
+        updated_output["artifact_id"] = str(artifact.id)
+        updated_output["transcription_id"] = str(transcription.id)
+        updated_output["source_file_id"] = str(source_file.id)
+        job = await self._job_repo.update(job, JobUpdate(output_data=updated_output))
 
-    async def _save_convert_artifacts(self, job: Job) -> None:
+        logger.info("Saved transcription artifacts for job %s", job.id)
+        return job
+
+    async def _save_convert_artifacts(self, job: Job) -> Job:
         """Create File and ArtifactMediaFile records for converted MP4."""
         input_data = job.input_data or {}
         output_data = job.output_data or {}
@@ -1503,7 +1532,7 @@ class TaskService:
         user = await user_repo.get_by_id(job.user_id)  # type: ignore[arg-type]
         if user is None:
             logger.warning("User %s not found, skipping convert artifact save", job.user_id)
-            return
+            return job
 
         # Derive output filename from source file
         file_repo = FileRepository(self._session)
@@ -1542,11 +1571,12 @@ class TaskService:
 
         updated_output = dict(output_data)
         updated_output["file_id"] = str(converted_file.id)
-        await self._job_repo.update(job, JobUpdate(output_data=updated_output))
+        job = await self._job_repo.update(job, JobUpdate(output_data=updated_output))
 
         logger.info("Saved convert artifacts for job %s", job.id)
+        return job
 
-    async def _save_silence_apply_artifacts(self, job: Job) -> None:
+    async def _save_silence_apply_artifacts(self, job: Job) -> Job:
         """Create File and ArtifactMediaFile records for silence-applied video."""
         input_data = job.input_data or {}
         output_data = job.output_data or {}
@@ -1562,10 +1592,8 @@ class TaskService:
         user_repo = UserRepository(self._session)
         user = await user_repo.get_by_id(job.user_id)  # type: ignore[arg-type]
         if user is None:
-            logger.warning(
-                "User %s not found, skipping silence apply artifact save", job.user_id
-            )
-            return
+            logger.warning("User %s not found, skipping silence apply artifact save", job.user_id)
+            return job
 
         file_repo = FileRepository(self._session)
         source_file = await file_repo.get_by_path(file_key)
@@ -1601,15 +1629,16 @@ class TaskService:
 
         updated_output = dict(output_data)
         updated_output["file_id"] = str(processed_file.id)
-        await self._job_repo.update(job, JobUpdate(output_data=updated_output))
+        job = await self._job_repo.update(job, JobUpdate(output_data=updated_output))
 
         logger.info(
             "Saved silence apply artifacts for job %s (file_id=%s)",
             job.id,
             processed_file.id,
         )
+        return job
 
-    async def _save_captions_artifacts(self, job: Job) -> None:
+    async def _save_captions_artifacts(self, job: Job) -> Job:
         """Create File and ArtifactMediaFile records for captioned video."""
         input_data = job.input_data or {}
         output_data = job.output_data or {}
@@ -1626,7 +1655,7 @@ class TaskService:
         user = await user_repo.get_by_id(job.user_id)  # type: ignore[arg-type]
         if user is None:
             logger.warning("User %s not found, skipping captions artifact save", job.user_id)
-            return
+            return job
 
         # Get file size from S3
         storage = _get_storage_service()
@@ -1670,11 +1699,10 @@ class TaskService:
         # Update job output_data with file_id so frontend can reference it
         updated_output = dict(output_data)
         updated_output["file_id"] = str(captioned_file.id)
-        job = await self._job_repo.update(
-            job, JobUpdate(output_data=updated_output)
-        )
+        job = await self._job_repo.update(job, JobUpdate(output_data=updated_output))
 
         logger.info("Saved captions artifacts for job %s (file_id=%s)", job.id, captioned_file.id)
+        return job
 
     async def submit_media_probe(
         self, *, requester: User, request: MediaProbeRequest

pyproject.toml

@@ -26,6 +26,7 @@ dependencies = [
   "redis>=5.0.0",
   "psycopg2-binary>=2.9.9",
   "tiktoken>=0.3.3",
+  "greenlet>=3.3.0",
 ]
 
 [dependency-groups]

tests/conftest.py

@@ -6,11 +6,14 @@ from __future__ import annotations
 
 import uuid
 from datetime import timedelta
+from pathlib import Path
+from tempfile import NamedTemporaryFile
 from typing import AsyncGenerator
 from unittest.mock import AsyncMock, MagicMock
 
 import pytest
 from httpx import ASGITransport, AsyncClient
+from sqlalchemy import create_engine
 from sqlalchemy.ext.asyncio import AsyncSession, async_sessionmaker, create_async_engine
 
 from cpv3.db.base import Base
@@ -22,20 +25,23 @@ from cpv3.main import app
 from cpv3.modules.users.models import User
 
 
-# Use in-memory SQLite for tests (or configure a test database)
-TEST_DATABASE_URL = "sqlite+aiosqlite:///:memory:"
-
-
 @pytest.fixture
 async def test_engine():
     """Create a test database engine with tables."""
-    engine = create_async_engine(TEST_DATABASE_URL, echo=False)
-    async with engine.begin() as conn:
-        await conn.run_sync(Base.metadata.create_all)
+    with NamedTemporaryFile(suffix=".sqlite3", delete=False) as tmp_db:
+        db_path = Path(tmp_db.name)
+
+    sync_engine = create_engine(f"sqlite:///{db_path}", echo=False)
+    Base.metadata.create_all(bind=sync_engine)
+
+    engine = create_async_engine(f"sqlite+aiosqlite:///{db_path}", echo=False)
+    try:
         yield engine
-    async with engine.begin() as conn:
-        await conn.run_sync(Base.metadata.drop_all)
+    finally:
         await engine.dispose()
+        Base.metadata.drop_all(bind=sync_engine)
+        sync_engine.dispose()
+        db_path.unlink(missing_ok=True)
 
 
 @pytest.fixture

tests/integration/test_project_workspaces_endpoints.py

@@ -0,0 +1,212 @@
+from __future__ import annotations
+
+import uuid
+
+import pytest
+from httpx import AsyncClient
+from sqlalchemy.ext.asyncio import AsyncSession
+
+from cpv3.modules.files.models import File
+from cpv3.modules.projects.models import Project
+from cpv3.modules.tasks.schemas import TaskSubmitResponse
+from cpv3.modules.tasks.service import TaskService
+from cpv3.modules.users.models import User
+
+pytest.importorskip("greenlet")
+
+
+@pytest.fixture
+async def workflow_project(test_db_session: AsyncSession, test_user: User) -> Project:
+    project = Project(
+        id=uuid.uuid4(),
+        owner_id=test_user.id,
+        name="Workflow Project",
+        description="Typed workflow test project",
+        language="ru",
+        status="DRAFT",
+        is_active=True,
+    )
+    test_db_session.add(project)
+    await test_db_session.commit()
+    await test_db_session.refresh(project)
+    return project
+
+
+@pytest.fixture
+async def source_file(
+    test_db_session: AsyncSession,
+    test_user: User,
+    workflow_project: Project,
+) -> File:
+    file = File(
+        id=uuid.uuid4(),
+        owner_id=test_user.id,
+        project_id=workflow_project.id,
+        original_filename="source.mp4",
+        path="users/test/source.mp4",
+        storage_backend="S3",
+        mime_type="video/mp4",
+        size_bytes=1024,
+        file_format="mp4",
+        is_uploaded=True,
+        is_deleted=False,
+    )
+    test_db_session.add(file)
+    await test_db_session.commit()
+    await test_db_session.refresh(file)
+    return file
+
+
+class TestProjectWorkspaceEndpoints:
+    async def test_get_workspace_returns_default_state(
+        self,
+        auth_client: AsyncClient,
+        workflow_project: Project,
+    ) -> None:
+        response = await auth_client.get(f"/api/projects/{workflow_project.id}/workspace")
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["project_id"] == str(workflow_project.id)
+        assert data["revision"] == 0
+        assert data["version"] == 1
+        assert data["phase"] == "INGEST"
+        assert data["current_screen"] == "upload"
+        assert data["source_file_id"] is None
+        assert data["active_job"] is None
+        assert data["workspace_view"] == {
+            "used_file_ids": [],
+            "selected_file_id": None,
+        }
+
+    async def test_get_workspace_forbidden_for_other_users_project(
+        self,
+        auth_client: AsyncClient,
+        test_db_session: AsyncSession,
+        other_user: User,
+    ) -> None:
+        foreign_project = Project(
+            id=uuid.uuid4(),
+            owner_id=other_user.id,
+            name="Other Project",
+            description=None,
+            language="ru",
+            status="DRAFT",
+            is_active=True,
+        )
+        test_db_session.add(foreign_project)
+        await test_db_session.commit()
+
+        response = await auth_client.get(f"/api/projects/{foreign_project.id}/workspace")
+
+        assert response.status_code == 403
+
+    async def test_set_source_file_action_updates_workspace(
+        self,
+        auth_client: AsyncClient,
+        workflow_project: Project,
+        source_file: File,
+    ) -> None:
+        response = await auth_client.post(
+            f"/api/projects/{workflow_project.id}/workflow/actions",
+            json={
+                "type": "SET_SOURCE_FILE",
+                "revision": 0,
+                "file_id": str(source_file.id),
+            },
+        )
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["revision"] == 1
+        assert data["phase"] == "VERIFY"
+        assert data["current_screen"] == "verify"
+        assert data["source_file_id"] == str(source_file.id)
+        assert data["workspace_view"] == {
+            "used_file_ids": [str(source_file.id)],
+            "selected_file_id": str(source_file.id),
+        }
+
+    async def test_action_returns_conflict_on_stale_revision(
+        self,
+        auth_client: AsyncClient,
+        workflow_project: Project,
+        source_file: File,
+    ) -> None:
+        first_response = await auth_client.post(
+            f"/api/projects/{workflow_project.id}/workflow/actions",
+            json={
+                "type": "SET_SOURCE_FILE",
+                "revision": 0,
+                "file_id": str(source_file.id),
+            },
+        )
+        assert first_response.status_code == 200
+
+        response = await auth_client.post(
+            f"/api/projects/{workflow_project.id}/workflow/actions",
+            json={
+                "type": "RESET_SOURCE_FILE",
+                "revision": 0,
+            },
+        )
+
+        assert response.status_code == 409
+
+    async def test_start_media_convert_action_sets_active_job(
+        self,
+        auth_client: AsyncClient,
+        workflow_project: Project,
+        source_file: File,
+        monkeypatch: pytest.MonkeyPatch,
+    ) -> None:
+        async def fake_submit_media_convert(
+            self,
+            *,
+            requester: User,
+            request,
+        ) -> TaskSubmitResponse:
+            assert requester.id == workflow_project.owner_id
+            assert request.file_key == source_file.path
+            assert request.project_id == workflow_project.id
+            return TaskSubmitResponse(
+                job_id=uuid.UUID("00000000-0000-4000-a000-000000000123"),
+                webhook_url=("http://test/api/tasks/webhook/00000000-0000-4000-a000-000000000123/"),
+                status="PENDING",
+            )
+
+        monkeypatch.setattr(
+            TaskService,
+            "submit_media_convert",
+            fake_submit_media_convert,
+        )
+
+        set_source_response = await auth_client.post(
+            f"/api/projects/{workflow_project.id}/workflow/actions",
+            json={
+                "type": "SET_SOURCE_FILE",
+                "revision": 0,
+                "file_id": str(source_file.id),
+            },
+        )
+        assert set_source_response.status_code == 200
+
+        response = await auth_client.post(
+            f"/api/projects/{workflow_project.id}/workflow/actions",
+            json={
+                "type": "START_MEDIA_CONVERT",
+                "revision": 1,
+                "output_format": "mp4",
+                "out_folder": "output_files",
+            },
+        )
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["revision"] == 2
+        assert data["phase"] == "VERIFY"
+        assert data["current_screen"] == "verify"
+        assert data["active_job"] == {
+            "job_id": "00000000-0000-4000-a000-000000000123",
+            "job_type": "MEDIA_CONVERT",
+        }

tests/unit/test_project_workspace_service.py

@@ -0,0 +1,210 @@
+from __future__ import annotations
+
+import uuid
+from types import SimpleNamespace
+
+import pytest
+
+from cpv3.modules.project_workspaces.schemas import (
+    ProjectWorkspaceState,
+    build_workspace_state_from_legacy,
+)
+from cpv3.modules.project_workspaces.service import ProjectWorkspaceService
+
+
+def test_build_workspace_state_from_legacy_maps_known_fields() -> None:
+    source_file_id = uuid.uuid4()
+    active_job_id = uuid.uuid4()
+    silence_job_id = uuid.uuid4()
+    artifact_id = uuid.uuid4()
+
+    workspace = build_workspace_state_from_legacy(
+        {
+            "wizard": {
+                "current_step": "subtitle-revision",
+                "primary_file_id": str(source_file_id),
+                "active_job_id": str(active_job_id),
+                "active_job_type": "TRANSCRIPTION_GENERATE",
+                "silence_job_id": str(silence_job_id),
+                "transcription_artifact_id": str(artifact_id),
+                "silence_settings": {
+                    "min_silence_duration_ms": 350,
+                    "silence_threshold_db": 21,
+                    "padding_ms": 180,
+                },
+                "unknown_field": "ignored",
+            },
+            "used_files": [
+                {"id": str(source_file_id), "path": "users/test/source.mp4"},
+                {"id": "not-a-uuid", "path": "broken"},
+            ],
+            "unknown_root": {"ignored": True},
+        }
+    )
+
+    assert workspace.phase == "TRANSCRIPTION"
+    assert workspace.source_file_id == source_file_id
+    assert workspace.active_job is not None
+    assert workspace.active_job.job_id == active_job_id
+    assert workspace.active_job.job_type == "TRANSCRIPTION_GENERATE"
+    assert workspace.workspace_view.used_file_ids == [source_file_id]
+    assert workspace.workspace_view.selected_file_id == source_file_id
+    assert workspace.silence.detect_job_id == silence_job_id
+    assert workspace.silence.settings.min_silence_duration_ms == 350
+    assert workspace.silence.settings.silence_threshold_db == 21
+    assert workspace.silence.settings.padding_ms == 180
+    assert workspace.transcription.artifact_id == artifact_id
+    assert workspace.captions.output_file_id is None
+
+
+@pytest.mark.asyncio
+@pytest.mark.parametrize(
+    ("job_type", "output_data", "initial_state", "expected"),
+    [
+        (
+            "MEDIA_CONVERT",
+            {"file_id": "00000000-0000-4000-a000-000000000101"},
+            {
+                "phase": "VERIFY",
+                "source_file_id": "00000000-0000-4000-a000-000000000001",
+                "active_job": {
+                    "job_id": "00000000-0000-4000-a000-000000000010",
+                    "job_type": "MEDIA_CONVERT",
+                },
+            },
+            {
+                "phase": "VERIFY",
+                "source_file_id": "00000000-0000-4000-a000-000000000101",
+                "active_job": None,
+                "current_screen": "verify",
+            },
+        ),
+        (
+            "SILENCE_DETECT",
+            {
+                "silent_segments": [{"start_ms": 100, "end_ms": 220}],
+                "duration_ms": 1000,
+            },
+            {
+                "phase": "SILENCE",
+                "source_file_id": "00000000-0000-4000-a000-000000000001",
+                "active_job": {
+                    "job_id": "00000000-0000-4000-a000-000000000011",
+                    "job_type": "SILENCE_DETECT",
+                },
+                "silence": {"status": "DETECTING"},
+            },
+            {
+                "phase": "SILENCE",
+                "active_job": None,
+                "silence_status": "REVIEWING",
+                "current_screen": "fragments",
+            },
+        ),
+        (
+            "SILENCE_APPLY",
+            {"file_id": "00000000-0000-4000-a000-000000000102"},
+            {
+                "phase": "SILENCE",
+                "source_file_id": "00000000-0000-4000-a000-000000000001",
+                "active_job": {
+                    "job_id": "00000000-0000-4000-a000-000000000012",
+                    "job_type": "SILENCE_APPLY",
+                },
+                "silence": {"status": "APPLYING"},
+            },
+            {
+                "phase": "TRANSCRIPTION",
+                "active_job": None,
+                "silence_status": "COMPLETED",
+                "current_screen": "transcription-settings",
+            },
+        ),
+        (
+            "TRANSCRIPTION_GENERATE",
+            {
+                "artifact_id": "00000000-0000-4000-a000-000000000103",
+                "transcription_id": "00000000-0000-4000-a000-000000000104",
+            },
+            {
+                "phase": "TRANSCRIPTION",
+                "source_file_id": "00000000-0000-4000-a000-000000000001",
+                "active_job": {
+                    "job_id": "00000000-0000-4000-a000-000000000013",
+                    "job_type": "TRANSCRIPTION_GENERATE",
+                },
+                "transcription": {"status": "PROCESSING"},
+            },
+            {
+                "phase": "TRANSCRIPTION",
+                "active_job": None,
+                "transcription_status": "REVIEWING",
+                "current_screen": "subtitle-revision",
+            },
+        ),
+        (
+            "CAPTIONS_GENERATE",
+            {"file_id": "00000000-0000-4000-a000-000000000105"},
+            {
+                "phase": "CAPTIONS",
+                "source_file_id": "00000000-0000-4000-a000-000000000001",
+                "active_job": {
+                    "job_id": "00000000-0000-4000-a000-000000000014",
+                    "job_type": "CAPTIONS_GENERATE",
+                },
+                "captions": {"status": "PROCESSING"},
+            },
+            {
+                "phase": "DONE",
+                "active_job": None,
+                "captions_status": "COMPLETED",
+                "current_screen": "caption-result",
+            },
+        ),
+    ],
+)
+async def test_apply_job_event_advances_workspace_for_done_jobs(
+    job_type: str,
+    output_data: dict[str, object],
+    initial_state: dict[str, object],
+    expected: dict[str, object],
+) -> None:
+    service = ProjectWorkspaceService(session=SimpleNamespace())
+
+    state = ProjectWorkspaceState.model_validate(
+        {
+            "version": 1,
+            "phase": "INGEST",
+            "active_job": None,
+            "source_file_id": None,
+            "workspace_view": {"used_file_ids": [], "selected_file_id": None},
+            "silence": {},
+            "transcription": {},
+            "captions": {},
+            **initial_state,
+        }
+    )
+
+    job = SimpleNamespace(
+        id=uuid.UUID(str(state.active_job.job_id)) if state.active_job else uuid.uuid4(),
+        project_id=uuid.uuid4(),
+        job_type=job_type,
+        status="DONE",
+        output_data=output_data,
+    )
+
+    next_state = service._apply_job_event_to_state(state, job)
+    current_screen = service._derive_current_screen(next_state)
+
+    assert next_state.phase == expected["phase"]
+    assert next_state.active_job == expected["active_job"]
+    assert current_screen == expected["current_screen"]
+
+    if "source_file_id" in expected:
+        assert str(next_state.source_file_id) == expected["source_file_id"]
+    if "silence_status" in expected:
+        assert next_state.silence.status == expected["silence_status"]
+    if "transcription_status" in expected:
+        assert next_state.transcription.status == expected["transcription_status"]
+    if "captions_status" in expected:
+        assert next_state.captions.status == expected["captions_status"]

tests/unit/test_project_workspaces_service.py

@@ -0,0 +1,331 @@
+from __future__ import annotations
+
+import uuid
+from types import SimpleNamespace
+from unittest.mock import AsyncMock
+
+import pytest
+
+from cpv3.modules.project_workspaces.schemas import (
+    ActiveJobState,
+    ProjectWorkspaceState,
+    SetSourceFileAction,
+    SilenceSettingsState,
+    SilenceState,
+    StartSilenceDetectAction,
+    StartTranscriptionAction,
+    TranscriptionRequestState,
+)
+from cpv3.modules.project_workspaces.service import (
+    ProjectWorkspaceService,
+    WorkspaceRevisionConflictError,
+)
+
+
+@pytest.mark.asyncio
+async def test_get_workspace_returns_default_state_when_workspace_missing() -> None:
+    project = SimpleNamespace(id=uuid.uuid4(), workspace_state=None)
+
+    service = ProjectWorkspaceService(session=AsyncMock())
+    service._repo = SimpleNamespace(
+        get_by_project_id=AsyncMock(return_value=None),
+        create=AsyncMock(
+            return_value=SimpleNamespace(
+                project_id=project.id,
+                revision=0,
+                state=ProjectWorkspaceState().model_dump(mode="json"),
+            )
+        ),
+    )
+
+    workspace = await service.get_workspace(project=project)
+
+    assert workspace.revision == 0
+    assert workspace.phase == "INGEST"
+    assert workspace.current_screen == "upload"
+    assert workspace.active_job is None
+    assert workspace.source_file_id is None
+    assert workspace.workspace_view.used_file_ids == []
+    assert workspace.workspace_view.selected_file_id is None
+
+
+@pytest.mark.asyncio
+async def test_apply_action_set_source_file_moves_workspace_to_verify() -> None:
+    project = SimpleNamespace(id=uuid.uuid4(), workspace_state=None)
+    requester = SimpleNamespace(id=uuid.uuid4(), is_staff=False)
+    file_id = uuid.uuid4()
+    workspace_row = SimpleNamespace(
+        project_id=project.id,
+        revision=0,
+        state=ProjectWorkspaceState().model_dump(mode="json"),
+    )
+    saved_state: dict[str, object] = {}
+
+    async def update_state(*, project_id, expected_revision, state):
+        saved_state.update(state)
+        return SimpleNamespace(
+            project_id=project_id,
+            revision=expected_revision + 1,
+            state=state,
+        )
+
+    service = ProjectWorkspaceService(session=AsyncMock())
+    service._repo = SimpleNamespace(
+        get_by_project_id=AsyncMock(return_value=workspace_row),
+        create=AsyncMock(),
+        update_state=AsyncMock(side_effect=update_state),
+    )
+    service._file_repo = SimpleNamespace(
+        get_by_id=AsyncMock(
+            return_value=SimpleNamespace(
+                id=file_id,
+                owner_id=requester.id,
+                project_id=project.id,
+                path="users/test/source.mp4",
+            )
+        )
+    )
+
+    workspace = await service.apply_action(
+        requester=requester,
+        project=project,
+        action=SetSourceFileAction(
+            type="SET_SOURCE_FILE",
+            revision=0,
+            file_id=file_id,
+        ),
+    )
+
+    assert workspace.revision == 1
+    assert workspace.phase == "VERIFY"
+    assert workspace.current_screen == "verify"
+    assert workspace.source_file_id == file_id
+    assert saved_state["source_file_id"] == str(file_id)
+    assert saved_state["workspace_view"] == {
+        "used_file_ids": [str(file_id)],
+        "selected_file_id": str(file_id),
+    }
+
+
+@pytest.mark.asyncio
+async def test_apply_action_rejects_stale_revision() -> None:
+    project = SimpleNamespace(id=uuid.uuid4(), workspace_state=None)
+    workspace_row = SimpleNamespace(
+        project_id=project.id,
+        revision=2,
+        state=ProjectWorkspaceState().model_dump(mode="json"),
+    )
+
+    service = ProjectWorkspaceService(session=AsyncMock())
+    service._repo = SimpleNamespace(
+        get_by_project_id=AsyncMock(return_value=workspace_row),
+        create=AsyncMock(),
+        update_state=AsyncMock(),
+    )
+
+    with pytest.raises(WorkspaceRevisionConflictError):
+        await service.apply_action(
+            requester=SimpleNamespace(id=uuid.uuid4(), is_staff=False),
+            project=project,
+            action=SetSourceFileAction(
+                type="SET_SOURCE_FILE",
+                revision=1,
+                file_id=uuid.uuid4(),
+            ),
+        )
+
+
+@pytest.mark.asyncio
+async def test_start_silence_detect_submits_task_and_tracks_active_job() -> None:
+    project = SimpleNamespace(id=uuid.uuid4(), workspace_state=None)
+    requester = SimpleNamespace(id=uuid.uuid4(), is_staff=False)
+    source_file_id = uuid.uuid4()
+    workspace_state = ProjectWorkspaceState(
+        phase="SILENCE",
+        source_file_id=source_file_id,
+        silence=SilenceState(
+            status="CONFIGURED",
+            settings=SilenceSettingsState(
+                min_silence_duration_ms=250,
+                silence_threshold_db=18,
+                padding_ms=125,
+            ),
+        ),
+    )
+    workspace_row = SimpleNamespace(
+        project_id=project.id,
+        revision=0,
+        state=workspace_state.model_dump(mode="json"),
+    )
+    submitted_response = SimpleNamespace(job_id=uuid.uuid4(), status="PENDING")
+
+    async def update_state(*, project_id, expected_revision, state):
+        return SimpleNamespace(
+            project_id=project_id,
+            revision=expected_revision + 1,
+            state=state,
+        )
+
+    task_service = SimpleNamespace(
+        submit_silence_detect=AsyncMock(return_value=submitted_response),
+    )
+
+    service = ProjectWorkspaceService(session=AsyncMock())
+    service._repo = SimpleNamespace(
+        get_by_project_id=AsyncMock(return_value=workspace_row),
+        create=AsyncMock(),
+        update_state=AsyncMock(side_effect=update_state),
+    )
+    service._file_repo = SimpleNamespace(
+        get_by_id=AsyncMock(
+            return_value=SimpleNamespace(
+                id=source_file_id,
+                owner_id=requester.id,
+                project_id=project.id,
+                path="projects/test/video.mp4",
+            )
+        )
+    )
+    service._task_service_factory = lambda: task_service
+
+    workspace = await service.apply_action(
+        requester=requester,
+        project=project,
+        action=StartSilenceDetectAction(
+            type="START_SILENCE_DETECT",
+            revision=0,
+        ),
+    )
+
+    task_service.submit_silence_detect.assert_awaited_once()
+    assert workspace.current_screen == "processing"
+    assert workspace.active_job == ActiveJobState(
+        job_id=submitted_response.job_id,
+        job_type="SILENCE_DETECT",
+    )
+    assert workspace.silence.detect_job_id == submitted_response.job_id
+
+
+@pytest.mark.asyncio
+async def test_start_transcription_persists_request_and_processing_job() -> None:
+    project = SimpleNamespace(id=uuid.uuid4(), workspace_state=None)
+    requester = SimpleNamespace(id=uuid.uuid4(), is_staff=False)
+    source_file_id = uuid.uuid4()
+    workspace_state = ProjectWorkspaceState(
+        phase="TRANSCRIPTION",
+        source_file_id=source_file_id,
+    )
+    workspace_row = SimpleNamespace(
+        project_id=project.id,
+        revision=3,
+        state=workspace_state.model_dump(mode="json"),
+    )
+    submitted_response = SimpleNamespace(job_id=uuid.uuid4(), status="PENDING")
+
+    async def update_state(*, project_id, expected_revision, state):
+        return SimpleNamespace(
+            project_id=project_id,
+            revision=expected_revision + 1,
+            state=state,
+        )
+
+    task_service = SimpleNamespace(
+        submit_transcription_generate=AsyncMock(return_value=submitted_response),
+    )
+
+    service = ProjectWorkspaceService(session=AsyncMock())
+    service._repo = SimpleNamespace(
+        get_by_project_id=AsyncMock(return_value=workspace_row),
+        create=AsyncMock(),
+        update_state=AsyncMock(side_effect=update_state),
+    )
+    service._file_repo = SimpleNamespace(
+        get_by_id=AsyncMock(
+            return_value=SimpleNamespace(
+                id=source_file_id,
+                owner_id=requester.id,
+                project_id=project.id,
+                path="projects/test/video.mp4",
+            )
+        )
+    )
+    service._task_service_factory = lambda: task_service
+
+    request = TranscriptionRequestState(engine="whisper", language="ru", model="base")
+    workspace = await service.apply_action(
+        requester=requester,
+        project=project,
+        action=StartTranscriptionAction(
+            type="START_TRANSCRIPTION",
+            revision=3,
+            request=request,
+        ),
+    )
+
+    assert workspace.current_screen == "transcription-processing"
+    assert workspace.transcription.request == request
+    assert workspace.active_job == ActiveJobState(
+        job_id=submitted_response.job_id,
+        job_type="TRANSCRIPTION_GENERATE",
+    )
+
+
+@pytest.mark.asyncio
+async def test_apply_job_update_moves_transcription_job_to_review() -> None:
+    project = SimpleNamespace(id=uuid.uuid4(), workspace_state=None)
+    job_id = uuid.uuid4()
+    transcription_id = uuid.uuid4()
+    artifact_id = uuid.uuid4()
+    workspace_state = ProjectWorkspaceState(
+        phase="TRANSCRIPTION",
+        active_job=ActiveJobState(job_id=job_id, job_type="TRANSCRIPTION_GENERATE"),
+        transcription={
+            "status": "PROCESSING",
+            "job_id": job_id,
+            "artifact_id": None,
+            "transcription_id": None,
+            "reviewed": False,
+        },
+    )
+    workspace_row = SimpleNamespace(
+        project_id=project.id,
+        revision=4,
+        state=workspace_state.model_dump(mode="json"),
+    )
+
+    async def update_state(*, project_id, expected_revision, state):
+        return SimpleNamespace(
+            project_id=project_id,
+            revision=expected_revision + 1,
+            state=state,
+        )
+
+    service = ProjectWorkspaceService(session=AsyncMock())
+    service._repo = SimpleNamespace(
+        get_by_project_id=AsyncMock(return_value=workspace_row),
+        create=AsyncMock(),
+        update_state=AsyncMock(side_effect=update_state),
+    )
+
+    workspace = await service.apply_job_update(
+        project=project,
+        job=SimpleNamespace(
+            id=job_id,
+            project_id=project.id,
+            job_type="TRANSCRIPTION_GENERATE",
+            status="DONE",
+            output_data={
+                "transcription_id": str(transcription_id),
+                "artifact_id": str(artifact_id),
+            },
+        ),
+    )
+
+    assert workspace is not None
+    assert workspace.revision == 5
+    assert workspace.phase == "TRANSCRIPTION"
+    assert workspace.current_screen == "subtitle-revision"
+    assert workspace.active_job is None
+    assert workspace.transcription.transcription_id == transcription_id
+    assert workspace.transcription.artifact_id == artifact_id
+    assert workspace.transcription.reviewed is False

tests/unit/test_task_service.py

@@ -95,6 +95,7 @@ async def test_cancel_job_marks_job_cancelled_and_keeps_record() -> None:
     service._cancel_dramatiq_message = AsyncMock()
     service._cancel_caption_render = AsyncMock()
     service._create_cancellation_notification = AsyncMock()
+    service._sync_project_workspace_after_webhook = AsyncMock()
 
     result = await service.cancel_job(job)
 
@@ -103,9 +104,8 @@ async def test_cancel_job_marks_job_cancelled_and_keeps_record() -> None:
     service._event_repo.create.assert_awaited_once()
     service._cancel_dramatiq_message.assert_awaited_once_with(job.broker_id)
     service._cancel_caption_render.assert_awaited_once_with(job)
-	service._create_cancellation_notification.assert_awaited_once_with(
-		cancelled_job
-	)
+    service._create_cancellation_notification.assert_awaited_once_with(cancelled_job)
+    service._sync_project_workspace_after_webhook.assert_awaited_once_with(cancelled_job)
 
 
 @pytest.mark.asyncio
@@ -114,6 +114,7 @@ async def test_record_webhook_event_updates_progress_for_conversion_job() -> Non
         id=uuid.uuid4(),
         status="RUNNING",
         job_type="MEDIA_CONVERT",
+        project_id=uuid.uuid4(),
         user_id=None,
     )
     updated_job = SimpleNamespace(**job.__dict__, project_pct=52.5)
@@ -126,6 +127,7 @@ async def test_record_webhook_event_updates_progress_for_conversion_job() -> Non
     service = TaskService(session=AsyncMock())
     service._job_repo = job_repo
     service._event_repo = event_repo
+    service._sync_project_workspace_after_webhook = AsyncMock()
 
     result = await service.record_webhook_event(
         job_id=job.id,
@@ -143,3 +145,85 @@ async def test_record_webhook_event_updates_progress_for_conversion_job() -> Non
     assert update_call.current_message == "Конвертация видео"
     assert event_call.event_type == "progress"
     assert event_call.payload["progress_pct"] == 52.5
+
+
+@pytest.mark.asyncio
+async def test_record_webhook_event_syncs_workspace_for_completed_supported_job() -> None:
+    job = SimpleNamespace(
+        id=uuid.uuid4(),
+        status="RUNNING",
+        job_type="SILENCE_DETECT",
+        user_id=None,
+        project_id=uuid.uuid4(),
+        output_data={"silent_segments": []},
+    )
+    updated_job = SimpleNamespace(**{**job.__dict__, "status": "DONE"})
+    job_repo = SimpleNamespace(
+        get_by_id=AsyncMock(return_value=job),
+        update=AsyncMock(return_value=updated_job),
+    )
+    event_repo = SimpleNamespace(create=AsyncMock())
+
+    service = TaskService(session=AsyncMock())
+    service._job_repo = job_repo
+    service._event_repo = event_repo
+    service._sync_project_workspace_after_webhook = AsyncMock()
+
+    result = await service.record_webhook_event(
+        job_id=job.id,
+        event=TaskWebhookEvent(
+            status="DONE",
+            current_message="Готово",
+            output_data={"silent_segments": []},
+        ),
+    )
+
+    assert result is updated_job
+    service._sync_project_workspace_after_webhook.assert_awaited_once_with(updated_job)
+
+
+@pytest.mark.asyncio
+async def test_record_webhook_event_projects_workspace_after_done_job() -> None:
+    job = SimpleNamespace(
+        id=uuid.uuid4(),
+        status="RUNNING",
+        job_type="MEDIA_CONVERT",
+        project_id=uuid.uuid4(),
+        user_id=None,
+        output_data={"file_path": "users/test/converted.mp4"},
+    )
+    updated_job = SimpleNamespace(
+        **{
+            **job.__dict__,
+            "status": "DONE",
+            "output_data": {
+                "file_path": "users/test/converted.mp4",
+                "file_id": "00000000-0000-4000-a000-000000000777",
+            },
+        }
+    )
+    job_repo = SimpleNamespace(
+        get_by_id=AsyncMock(return_value=job),
+        update=AsyncMock(return_value=updated_job),
+    )
+    event_repo = SimpleNamespace(create=AsyncMock())
+    workspace_service = SimpleNamespace(handle_job_update=AsyncMock())
+
+    service = TaskService(session=AsyncMock())
+    service._job_repo = job_repo
+    service._event_repo = event_repo
+    service._save_convert_artifacts = AsyncMock(return_value=updated_job)
+    service._get_project_workspace_service = lambda: workspace_service
+
+    result = await service.record_webhook_event(
+        job_id=job.id,
+        event=TaskWebhookEvent(
+            status="DONE",
+            current_message="Готово",
+            output_data={"file_path": "users/test/converted.mp4"},
+        ),
+    )
+
+    assert result is updated_job
+    service._save_convert_artifacts.assert_awaited_once_with(updated_job)
+    workspace_service.handle_job_update.assert_awaited_once_with(job=updated_job)

uv.lock

@@ -404,6 +404,7 @@ dependencies = [
     { name = "dramatiq", extra = ["redis"] },
     { name = "fastapi" },
     { name = "google-cloud-speech" },
+    { name = "greenlet" },
     { name = "httpx" },
     { name = "openai-whisper" },
     { name = "passlib", extra = ["bcrypt"] },
@@ -444,6 +445,7 @@ requires-dist = [
     { name = "dramatiq", extras = ["redis"], specifier = ">=1.17.0" },
     { name = "fastapi", specifier = ">=0.115.0" },
     { name = "google-cloud-speech", specifier = ">=2.34.0" },
+    { name = "greenlet", specifier = ">=3.3.0" },
     { name = "httpx", specifier = ">=0.27.0" },
     { name = "openai-whisper", specifier = ">=20250625" },
     { name = "passlib", extras = ["bcrypt"], specifier = ">=1.7.4" },
@@ -628,6 +630,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/1f/cb/48e964c452ca2b92175a9b2dca037a553036cb053ba69e284650ce755f13/greenlet-3.3.0-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:e29f3018580e8412d6aaf5641bb7745d38c85228dacf51a73bd4e26ddf2a6a8e", size = 274908, upload-time = "2025-12-04T14:23:26.435Z" },
     { url = "https://files.pythonhosted.org/packages/28/da/38d7bff4d0277b594ec557f479d65272a893f1f2a716cad91efeb8680953/greenlet-3.3.0-cp311-cp311-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a687205fb22794e838f947e2194c0566d3812966b41c78709554aa883183fb62", size = 577113, upload-time = "2025-12-04T14:50:05.493Z" },
     { url = "https://files.pythonhosted.org/packages/3c/f2/89c5eb0faddc3ff014f1c04467d67dee0d1d334ab81fadbf3744847f8a8a/greenlet-3.3.0-cp311-cp311-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:4243050a88ba61842186cb9e63c7dfa677ec146160b0efd73b855a3d9c7fcf32", size = 590338, upload-time = "2025-12-04T14:57:41.136Z" },
+    { url = "https://files.pythonhosted.org/packages/80/d7/db0a5085035d05134f8c089643da2b44cc9b80647c39e93129c5ef170d8f/greenlet-3.3.0-cp311-cp311-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:670d0f94cd302d81796e37299bcd04b95d62403883b24225c6b5271466612f45", size = 601098, upload-time = "2025-12-04T15:07:11.898Z" },
     { url = "https://files.pythonhosted.org/packages/dc/a6/e959a127b630a58e23529972dbc868c107f9d583b5a9f878fb858c46bc1a/greenlet-3.3.0-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6cb3a8ec3db4a3b0eb8a3c25436c2d49e3505821802074969db017b87bc6a948", size = 590206, upload-time = "2025-12-04T14:26:01.254Z" },
     { url = "https://files.pythonhosted.org/packages/48/60/29035719feb91798693023608447283b266b12efc576ed013dd9442364bb/greenlet-3.3.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:2de5a0b09eab81fc6a382791b995b1ccf2b172a9fec934747a7a23d2ff291794", size = 1550668, upload-time = "2025-12-04T15:04:22.439Z" },
     { url = "https://files.pythonhosted.org/packages/0a/5f/783a23754b691bfa86bd72c3033aa107490deac9b2ef190837b860996c9f/greenlet-3.3.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:4449a736606bd30f27f8e1ff4678ee193bc47f6ca810d705981cfffd6ce0d8c5", size = 1615483, upload-time = "2025-12-04T14:27:28.083Z" },
@@ -635,6 +638,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/f8/0a/a3871375c7b9727edaeeea994bfff7c63ff7804c9829c19309ba2e058807/greenlet-3.3.0-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:b01548f6e0b9e9784a2c99c5651e5dc89ffcbe870bc5fb2e5ef864e9cc6b5dcb", size = 276379, upload-time = "2025-12-04T14:23:30.498Z" },
     { url = "https://files.pythonhosted.org/packages/43/ab/7ebfe34dce8b87be0d11dae91acbf76f7b8246bf9d6b319c741f99fa59c6/greenlet-3.3.0-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:349345b770dc88f81506c6861d22a6ccd422207829d2c854ae2af8025af303e3", size = 597294, upload-time = "2025-12-04T14:50:06.847Z" },
     { url = "https://files.pythonhosted.org/packages/a4/39/f1c8da50024feecd0793dbd5e08f526809b8ab5609224a2da40aad3a7641/greenlet-3.3.0-cp312-cp312-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:e8e18ed6995e9e2c0b4ed264d2cf89260ab3ac7e13555b8032b25a74c6d18655", size = 607742, upload-time = "2025-12-04T14:57:42.349Z" },
+    { url = "https://files.pythonhosted.org/packages/77/cb/43692bcd5f7a0da6ec0ec6d58ee7cddb606d055ce94a62ac9b1aa481e969/greenlet-3.3.0-cp312-cp312-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:c024b1e5696626890038e34f76140ed1daf858e37496d33f2af57f06189e70d7", size = 622297, upload-time = "2025-12-04T15:07:13.552Z" },
     { url = "https://files.pythonhosted.org/packages/75/b0/6bde0b1011a60782108c01de5913c588cf51a839174538d266de15e4bf4d/greenlet-3.3.0-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:047ab3df20ede6a57c35c14bf5200fcf04039d50f908270d3f9a7a82064f543b", size = 609885, upload-time = "2025-12-04T14:26:02.368Z" },
     { url = "https://files.pythonhosted.org/packages/49/0e/49b46ac39f931f59f987b7cd9f34bfec8ef81d2a1e6e00682f55be5de9f4/greenlet-3.3.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2d9ad37fc657b1102ec880e637cccf20191581f75c64087a549e66c57e1ceb53", size = 1567424, upload-time = "2025-12-04T15:04:23.757Z" },
     { url = "https://files.pythonhosted.org/packages/05/f5/49a9ac2dff7f10091935def9165c90236d8f175afb27cbed38fb1d61ab6b/greenlet-3.3.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:83cd0e36932e0e7f36a64b732a6f60c2fc2df28c351bae79fbaf4f8092fe7614", size = 1636017, upload-time = "2025-12-04T14:27:29.688Z" },
@@ -642,6 +646,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/02/2f/28592176381b9ab2cafa12829ba7b472d177f3acc35d8fbcf3673d966fff/greenlet-3.3.0-cp313-cp313-macosx_11_0_universal2.whl", hash = "sha256:a1e41a81c7e2825822f4e068c48cb2196002362619e2d70b148f20a831c00739", size = 275140, upload-time = "2025-12-04T14:23:01.282Z" },
     { url = "https://files.pythonhosted.org/packages/2c/80/fbe937bf81e9fca98c981fe499e59a3f45df2a04da0baa5c2be0dca0d329/greenlet-3.3.0-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9f515a47d02da4d30caaa85b69474cec77b7929b2e936ff7fb853d42f4bf8808", size = 599219, upload-time = "2025-12-04T14:50:08.309Z" },
     { url = "https://files.pythonhosted.org/packages/c2/ff/7c985128f0514271b8268476af89aee6866df5eec04ac17dcfbc676213df/greenlet-3.3.0-cp313-cp313-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:7d2d9fd66bfadf230b385fdc90426fcd6eb64db54b40c495b72ac0feb5766c54", size = 610211, upload-time = "2025-12-04T14:57:43.968Z" },
+    { url = "https://files.pythonhosted.org/packages/79/07/c47a82d881319ec18a4510bb30463ed6891f2ad2c1901ed5ec23d3de351f/greenlet-3.3.0-cp313-cp313-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:30a6e28487a790417d036088b3bcb3f3ac7d8babaa7d0139edbaddebf3af9492", size = 624311, upload-time = "2025-12-04T15:07:14.697Z" },
     { url = "https://files.pythonhosted.org/packages/fd/8e/424b8c6e78bd9837d14ff7df01a9829fc883ba2ab4ea787d4f848435f23f/greenlet-3.3.0-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:087ea5e004437321508a8d6f20efc4cfec5e3c30118e1417ea96ed1d93950527", size = 612833, upload-time = "2025-12-04T14:26:03.669Z" },
     { url = "https://files.pythonhosted.org/packages/b5/ba/56699ff9b7c76ca12f1cdc27a886d0f81f2189c3455ff9f65246780f713d/greenlet-3.3.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ab97cf74045343f6c60a39913fa59710e4bd26a536ce7ab2397adf8b27e67c39", size = 1567256, upload-time = "2025-12-04T15:04:25.276Z" },
     { url = "https://files.pythonhosted.org/packages/1e/37/f31136132967982d698c71a281a8901daf1a8fbab935dce7c0cf15f942cc/greenlet-3.3.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:5375d2e23184629112ca1ea89a53389dddbffcf417dad40125713d88eb5f96e8", size = 1636483, upload-time = "2025-12-04T14:27:30.804Z" },
@@ -649,6 +654,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/d7/7c/f0a6d0ede2c7bf092d00bc83ad5bafb7e6ec9b4aab2fbdfa6f134dc73327/greenlet-3.3.0-cp314-cp314-macosx_11_0_universal2.whl", hash = "sha256:60c2ef0f578afb3c8d92ea07ad327f9a062547137afe91f38408f08aacab667f", size = 275671, upload-time = "2025-12-04T14:23:05.267Z" },
     { url = "https://files.pythonhosted.org/packages/44/06/dac639ae1a50f5969d82d2e3dd9767d30d6dbdbab0e1a54010c8fe90263c/greenlet-3.3.0-cp314-cp314-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0a5d554d0712ba1de0a6c94c640f7aeba3f85b3a6e1f2899c11c2c0428da9365", size = 646360, upload-time = "2025-12-04T14:50:10.026Z" },
     { url = "https://files.pythonhosted.org/packages/e0/94/0fb76fe6c5369fba9bf98529ada6f4c3a1adf19e406a47332245ef0eb357/greenlet-3.3.0-cp314-cp314-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3a898b1e9c5f7307ebbde4102908e6cbfcb9ea16284a3abe15cab996bee8b9b3", size = 658160, upload-time = "2025-12-04T14:57:45.41Z" },
+    { url = "https://files.pythonhosted.org/packages/93/79/d2c70cae6e823fac36c3bbc9077962105052b7ef81db2f01ec3b9bf17e2b/greenlet-3.3.0-cp314-cp314-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:dcd2bdbd444ff340e8d6bdf54d2f206ccddbb3ccfdcd3c25bf4afaa7b8f0cf45", size = 671388, upload-time = "2025-12-04T15:07:15.789Z" },
     { url = "https://files.pythonhosted.org/packages/b8/14/bab308fc2c1b5228c3224ec2bf928ce2e4d21d8046c161e44a2012b5203e/greenlet-3.3.0-cp314-cp314-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5773edda4dc00e173820722711d043799d3adb4f01731f40619e07ea2750b955", size = 660166, upload-time = "2025-12-04T14:26:05.099Z" },
     { url = "https://files.pythonhosted.org/packages/4b/d2/91465d39164eaa0085177f61983d80ffe746c5a1860f009811d498e7259c/greenlet-3.3.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:ac0549373982b36d5fd5d30beb8a7a33ee541ff98d2b502714a09f1169f31b55", size = 1615193, upload-time = "2025-12-04T15:04:27.041Z" },
     { url = "https://files.pythonhosted.org/packages/42/1b/83d110a37044b92423084d52d5d5a3b3a73cafb51b547e6d7366ff62eff1/greenlet-3.3.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:d198d2d977460358c3b3a4dc844f875d1adb33817f0613f663a656f463764ccc", size = 1683653, upload-time = "2025-12-04T14:27:32.366Z" },
@@ -656,6 +662,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/a0/66/bd6317bc5932accf351fc19f177ffba53712a202f9df10587da8df257c7e/greenlet-3.3.0-cp314-cp314t-macosx_11_0_universal2.whl", hash = "sha256:d6ed6f85fae6cdfdb9ce04c9bf7a08d666cfcfb914e7d006f44f840b46741931", size = 282638, upload-time = "2025-12-04T14:25:20.941Z" },
     { url = "https://files.pythonhosted.org/packages/30/cf/cc81cb030b40e738d6e69502ccbd0dd1bced0588e958f9e757945de24404/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d9125050fcf24554e69c4cacb086b87b3b55dc395a8b3ebe6487b045b2614388", size = 651145, upload-time = "2025-12-04T14:50:11.039Z" },
     { url = "https://files.pythonhosted.org/packages/9c/ea/1020037b5ecfe95ca7df8d8549959baceb8186031da83d5ecceff8b08cd2/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:87e63ccfa13c0a0f6234ed0add552af24cc67dd886731f2261e46e241608bee3", size = 654236, upload-time = "2025-12-04T14:57:47.007Z" },
+    { url = "https://files.pythonhosted.org/packages/69/cc/1e4bae2e45ca2fa55299f4e85854606a78ecc37fead20d69322f96000504/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:2662433acbca297c9153a4023fe2161c8dcfdcc91f10433171cf7e7d94ba2221", size = 662506, upload-time = "2025-12-04T15:07:16.906Z" },
     { url = "https://files.pythonhosted.org/packages/57/b9/f8025d71a6085c441a7eaff0fd928bbb275a6633773667023d19179fe815/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:3c6e9b9c1527a78520357de498b0e709fb9e2f49c3a513afd5a249007261911b", size = 653783, upload-time = "2025-12-04T14:26:06.225Z" },
     { url = "https://files.pythonhosted.org/packages/f6/c7/876a8c7a7485d5d6b5c6821201d542ef28be645aa024cfe1145b35c120c1/greenlet-3.3.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:286d093f95ec98fdd92fcb955003b8a3d054b4e2cab3e2707a5039e7b50520fd", size = 1614857, upload-time = "2025-12-04T15:04:28.484Z" },
     { url = "https://files.pythonhosted.org/packages/4f/dc/041be1dff9f23dac5f48a43323cd0789cb798342011c19a248d9c9335536/greenlet-3.3.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:6c10513330af5b8ae16f023e8ddbfb486ab355d04467c4679c5cfe4659975dd9", size = 1676034, upload-time = "2025-12-04T14:27:33.531Z" },