Perspectives undoredo

backend/src/common/crud_enum.py

@@ -26,9 +26,11 @@
 from modules.ml.tag_recommendation.tag_recommendation_crud import (
     crud_tag_recommendation_link,
 )
-from modules.perspectives.aspect_crud import crud_aspect
-from modules.perspectives.cluster_crud import crud_cluster
-from modules.perspectives.document_aspect_crud import crud_document_aspect
+from modules.perspectives.aspect.aspect_crud import crud_aspect
+from modules.perspectives.cluster.cluster_crud import crud_cluster
+from modules.perspectives.document_aspect.document_aspect_crud import (
+    crud_document_aspect,
+)
 from modules.timeline_analysis.timeline_analysis_crud import (
     crud_timeline_analysis,
 )

backend/src/core/doc/source_document_data_crud.py

@@ -1,8 +1,13 @@
+from sqlalchemy.orm import Session
+
+from common.doc_type import DocType
 from core.doc.source_document_data_dto import (
     SourceDocumentDataCreate,
     SourceDocumentDataUpdate,
 )
 from core.doc.source_document_data_orm import SourceDocumentDataORM
+from core.doc.source_document_orm import SourceDocumentORM
+from core.tag.tag_orm import TagORM
 from repos.db.crud_base import CRUDBase
 
 
@@ -13,7 +18,31 @@ class CRUDSourceDocumentData(
         SourceDocumentDataUpdate,
     ]
 ):
-    pass
+    def read_by_doctype_and_tag(
+        self,
+        db: Session,
+        *,
+        project_id: int,
+        tag_id: int | None,
+        doctype: DocType,
+    ) -> list[SourceDocumentDataORM]:
+        query = (
+            db.query(SourceDocumentDataORM)
+            .join(SourceDocumentORM, SourceDocumentORM.id == SourceDocumentDataORM.id)
+            .filter(
+                SourceDocumentORM.project_id == project_id,
+                SourceDocumentORM.doctype == doctype,
+            )
+        )
+
+        if tag_id is not None:
+            query = query.join(
+                SourceDocumentORM.tags,
+            ).filter(
+                SourceDocumentORM.tags.any(TagORM.id == tag_id),
+            )
+
+        return query.all()
 
 
 crud_sdoc_data = CRUDSourceDocumentData(SourceDocumentDataORM)

backend/src/core/doc/source_document_orm.py

@@ -30,10 +30,14 @@
     from modules.ml.tag_recommendation.tag_recommendation_orm import (
         TagRecommendationLinkORM,
     )
-    from modules.perspectives.aspect_orm import AspectORM
-    from modules.perspectives.cluster_orm import ClusterORM
-    from modules.perspectives.document_aspect_orm import DocumentAspectORM
-    from modules.perspectives.document_cluster_orm import DocumentClusterORM
+    from modules.perspectives.aspect.aspect_orm import AspectORM
+    from modules.perspectives.cluster.cluster_orm import ClusterORM
+    from modules.perspectives.document_aspect.document_aspect_orm import (
+        DocumentAspectORM,
+    )
+    from modules.perspectives.document_cluster.document_cluster_orm import (
+        DocumentClusterORM,
+    )
     from modules.word_frequency.word_frequency_orm import WordFrequencyORM
 
 

backend/src/core/project/project_orm.py

@@ -19,7 +19,7 @@
     from core.tag.tag_orm import TagORM
     from core.user.user_orm import UserORM
     from modules.classifier.classifier_orm import ClassifierORM
-    from modules.perspectives.aspect_orm import AspectORM
+    from modules.perspectives.aspect.aspect_orm import AspectORM
     from modules.whiteboard.whiteboard_orm import WhiteboardORM
 
 

backend/src/migrations/alembic/versions/02320d77b8dd_add_history_table.py

@@ -0,0 +1,43 @@
+"""add history table
+
+Revision ID: 02320d77b8dd
+Revises: 7e88b47a2f6c
+Create Date: 2025-12-11 15:04:12.210086
+
+"""
+
+from typing import Sequence
+
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = "02320d77b8dd"
+down_revision: str | None = "7e88b47a2f6c"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table(
+        "perspectivehistory",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("history_number", sa.Integer(), nullable=False),
+        sa.Column("is_undone", sa.Boolean(), nullable=False),
+        sa.Column("undo_data", sa.JSON(), nullable=False),
+        sa.Column("redo_data", sa.JSON(), nullable=False),
+        sa.Column("aspect_id", sa.Integer(), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["aspect_id"],
+            ["aspect.id"],
+        ),
+        sa.PrimaryKeyConstraint("id"),
+    )
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_table("perspectivehistory")
+    # ### end Alembic commands ###

backend/src/migrations/alembic/versions/4e79b051e348_add_perspectives_action_to_history.py

@@ -0,0 +1,33 @@
+"""add perspectives_action to perspectivehistory
+
+Revision ID: 4e79b051e348
+Revises: 02320d77b8dd
+Create Date: 2025-12-17 13:02:39.967887
+
+"""
+
+from typing import Sequence
+
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = "4e79b051e348"
+down_revision: str | None = "02320d77b8dd"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column(
+        "perspectivehistory",
+        sa.Column("perspectives_action", sa.String(), nullable=False),
+    )
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column("perspectivehistory", "perspectives_action")
+    # ### end Alembic commands ###

backend/src/migrations/alembic/versions/6617866c6b24_remove_is_hierarchical.py

@@ -0,0 +1,33 @@
+"""remove is_hierarchical column from aspect table
+
+Revision ID: 6617866c6b24
+Revises: 4e79b051e348
+Create Date: 2025-12-17 13:57:21.059533
+
+"""
+
+from typing import Sequence
+
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = "6617866c6b24"
+down_revision: str | None = "4e79b051e348"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column("aspect", "is_hierarchical")
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column(
+        "aspect",
+        sa.Column("is_hierarchical", sa.BOOLEAN(), autoincrement=False, nullable=False),
+    )
+    # ### end Alembic commands ###

backend/src/migrations/alembic/versions/7e88b47a2f6c_remove_cluster_hierarchy.py

@@ -0,0 +1,52 @@
+"""remove cluster hierarchy
+
+Revision ID: 7e88b47a2f6c
+Revises: 781d4852a256
+Create Date: 2025-12-11 09:22:57.085470
+
+"""
+
+from typing import Sequence
+
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = "7e88b47a2f6c"
+down_revision: str | None = "781d4852a256"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_index("ix_cluster_parent_cluster_id", table_name="cluster")
+    op.drop_constraint("cluster_parent_cluster_id_fkey", "cluster", type_="foreignkey")
+    op.drop_column("cluster", "parent_cluster_id")
+    op.drop_column("cluster", "level")
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column(
+        "cluster", sa.Column("level", sa.INTEGER(), autoincrement=False, nullable=False)
+    )
+    op.add_column(
+        "cluster",
+        sa.Column(
+            "parent_cluster_id", sa.INTEGER(), autoincrement=False, nullable=True
+        ),
+    )
+    op.create_foreign_key(
+        "cluster_parent_cluster_id_fkey",
+        "cluster",
+        "cluster",
+        ["parent_cluster_id"],
+        ["id"],
+        ondelete="SET NULL",
+    )
+    op.create_index(
+        "ix_cluster_parent_cluster_id", "cluster", ["parent_cluster_id"], unique=False
+    )
+    # ### end Alembic commands ###

backend/src/modules/perspectives/aspect_crud.py → backend/src/modules/perspectives/aspect/aspect_crud.py

@@ -1,5 +1,5 @@
-from modules.perspectives.aspect_dto import AspectCreate, AspectUpdateIntern
-from modules.perspectives.aspect_orm import AspectORM
+from modules.perspectives.aspect.aspect_dto import AspectCreate, AspectUpdateIntern
+from modules.perspectives.aspect.aspect_orm import AspectORM
 from repos.db.crud_base import CRUDBase
 
 

backend/src/modules/perspectives/aspect_dto.py → backend/src/modules/perspectives/aspect/aspect_dto.py

@@ -44,7 +44,6 @@ class AspectBase(BaseModel):
     doc_modification_prompt: str | None = Field(
         default=None, description="Prompt for document modification"
     )
-    is_hierarchical: bool = Field(description="Whether the aspect is hierarchical")
     modality: DocType = Field(description="Modality of the documents of this aspect")
     pipeline_settings: PipelineSettings = Field(
         description="Pipeline settings for this aspect"

backend/src/modules/perspectives/aspect_embedding_crud.py → backend/src/modules/perspectives/aspect/aspect_embedding_crud.py

@@ -1,8 +1,8 @@
 from weaviate import WeaviateClient
 from weaviate.classes.query import Filter
 
-from modules.perspectives.aspect_collection import AspectCollection
-from modules.perspectives.aspect_embedding_dto import AspectObjectIdentifier
+from modules.perspectives.aspect.aspect_collection import AspectCollection
+from modules.perspectives.aspect.aspect_embedding_dto import AspectObjectIdentifier
 from repos.vector.embedding_crud_base import CRUDBase
 from systems.event_system.events import project_deleted, source_document_deleted
 

backend/src/modules/perspectives/aspect_orm.py → backend/src/modules/perspectives/aspect/aspect_orm.py

@@ -1,6 +1,6 @@
 from typing import TYPE_CHECKING
 
-from sqlalchemy import Boolean, ForeignKey, Integer, String, Text, text
+from sqlalchemy import ForeignKey, Integer, String, Text, text
 from sqlalchemy.dialects.postgresql import JSONB
 from sqlalchemy.orm import Mapped, mapped_column, relationship
 
@@ -10,8 +10,11 @@
 if TYPE_CHECKING:
     from core.doc.source_document_orm import SourceDocumentORM
     from core.project.project_orm import ProjectORM
-    from modules.perspectives.cluster_orm import ClusterORM
-    from modules.perspectives.document_aspect_orm import DocumentAspectORM
+    from modules.perspectives.cluster.cluster_orm import ClusterORM
+    from modules.perspectives.document_aspect.document_aspect_orm import (
+        DocumentAspectORM,
+    )
+    from modules.perspectives.history.history_orm import PerspectiveHistoryORM
 
 
 class AspectORM(ORMBase):
@@ -20,9 +23,6 @@ class AspectORM(ORMBase):
     name: Mapped[str] = mapped_column(String, nullable=False)
     doc_embedding_prompt: Mapped[str] = mapped_column(Text, nullable=False)
     doc_modification_prompt: Mapped[str | None] = mapped_column(Text, nullable=True)
-    is_hierarchical: Mapped[bool] = mapped_column(
-        Boolean, default=False, nullable=False
-    )
     most_recent_job_id: Mapped[str | None] = mapped_column(String, nullable=True)
     embedding_model: Mapped[str] = mapped_column(
         String, server_default="default", nullable=False
@@ -54,6 +54,10 @@ class AspectORM(ORMBase):
         "ClusterORM", back_populates="aspect", cascade="all, delete-orphan"
     )
 
+    histories: Mapped[list["PerspectiveHistoryORM"]] = relationship(
+        "PerspectiveHistoryORM", back_populates="aspect", cascade="all, delete-orphan"
+    )
+
     # many to many
     document_aspects: Mapped[list["DocumentAspectORM"]] = relationship(
         "DocumentAspectORM", back_populates="aspect", cascade="all, delete-orphan"

backend/src/modules/perspectives/cluster/cluster_crud.py

@@ -0,0 +1,37 @@
+from modules.perspectives.cluster.cluster_dto import (
+    ClusterCreateIntern,
+    ClusterUpdateIntern,
+)
+from modules.perspectives.cluster.cluster_orm import ClusterORM
+from modules.perspectives.document_cluster.document_cluster_orm import (
+    DocumentClusterORM,
+)
+from repos.db.crud_base import CRUDBase
+
+
+class CRUDCluster(CRUDBase[ClusterORM, ClusterCreateIntern, ClusterUpdateIntern]):
+    def read_outlier_cluster(self, db, *, aspect_id: int) -> ClusterORM | None:
+        return (
+            db.query(self.model)
+            .filter(
+                self.model.aspect_id == aspect_id,
+                self.model.is_outlier.is_(True),
+            )
+            .first()
+        )
+
+    def read_by_aspect_and_sdoc(
+        self, db, *, aspect_id: int, sdoc_id: int
+    ) -> list[ClusterORM]:
+        return (
+            db.query(self.model)
+            .join(DocumentClusterORM, DocumentClusterORM.cluster_id == self.model.id)
+            .filter(
+                self.model.aspect_id == aspect_id,
+                DocumentClusterORM.sdoc_id == sdoc_id,
+            )
+            .all()
+        )
+
+
+crud_cluster = CRUDCluster(ClusterORM)