feat: added file management CRUD routes and pgvector database schema

alembic/script.py.mako

@@ -9,6 +9,7 @@ from typing import Sequence, Union
 
 from alembic import op
 import sqlalchemy as sa
+import pgvector.sqlalchemy
 ${imports if imports else ""}
 
 # revision identifiers, used by Alembic.

alembic/versions/ecca7b879961_genesis_migration.py

@@ -0,0 +1,75 @@
+"""genesis migration
+
+Revision ID: ecca7b879961
+Revises: 
+Create Date: 2026-06-20 16:46:17.090119
+
+"""
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+import pgvector.sqlalchemy
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision: str = 'ecca7b879961'
+down_revision: Union[str, Sequence[str], None] = None
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    """Upgrade schema."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table('files',
+    sa.Column('id', sa.UUID(), server_default=sa.text('gen_random_uuid()'), nullable=False),
+    sa.Column('file_path', sa.Text(), nullable=False),
+    sa.Column('file_hash', sa.String(length=64), nullable=False),
+    sa.Column('mime_type', sa.String(length=50), nullable=False),
+    sa.Column('last_modified', sa.DateTime(timezone=True), nullable=False),
+    sa.Column('indexed_at', sa.DateTime(timezone=True), server_default=sa.text('now()'), nullable=False),
+    sa.Column('tags', postgresql.ARRAY(sa.Text()), server_default='{}', nullable=False),
+    sa.Column('context', sa.Text(), nullable=True),
+    sa.PrimaryKeyConstraint('id'),
+    sa.UniqueConstraint('file_path')
+    )
+    op.create_index('files_path_hash_idx', 'files', ['file_path', 'file_hash'], unique=False)
+    op.create_index('files_tags_idx', 'files', ['tags'], unique=False, postgresql_using='gin')
+    op.create_table('file_content',
+    sa.Column('id', sa.UUID(), server_default=sa.text('gen_random_uuid()'), nullable=False),
+    sa.Column('file_id', sa.UUID(), nullable=False),
+    sa.Column('chunk_index', sa.Integer(), nullable=False),
+    sa.Column('content_text', sa.Text(), nullable=False),
+    sa.Column('embedding', pgvector.sqlalchemy.vector.VECTOR(dim=384), nullable=False),
+    sa.Column('keyword_tokens', postgresql.TSVECTOR(), sa.Computed("to_tsvector('english', content_text)", persisted=True), nullable=True),
+    sa.ForeignKeyConstraint(['file_id'], ['files.id'], ondelete='CASCADE'),
+    sa.PrimaryKeyConstraint('id')
+    )
+    op.create_index('content_embedding_idx', 'file_content', ['embedding'], unique=False, postgresql_using='hnsw', postgresql_with={'m': 16, 'ef_construction': 64}, postgresql_ops={'embedding': 'vector_cosine_ops'})
+    op.create_index('content_fts_idx', 'file_content', ['keyword_tokens'], unique=False, postgresql_using='gin')
+    op.create_table('file_relationships',
+    sa.Column('id', sa.UUID(), server_default=sa.text('gen_random_uuid()'), nullable=False),
+    sa.Column('source_file_id', sa.UUID(), nullable=False),
+    sa.Column('target_file_id', sa.UUID(), nullable=False),
+    sa.Column('similarity_score', sa.Float(), nullable=False),
+    sa.Column('relation_type', sa.String(length=50), server_default='semantic_similarity', nullable=False),
+    sa.ForeignKeyConstraint(['source_file_id'], ['files.id'], ondelete='CASCADE'),
+    sa.ForeignKeyConstraint(['target_file_id'], ['files.id'], ondelete='CASCADE'),
+    sa.PrimaryKeyConstraint('id'),
+    sa.UniqueConstraint('source_file_id', 'target_file_id', name='uq_file_relationship')
+    )
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    """Downgrade schema."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_table('file_relationships')
+    op.drop_index('content_fts_idx', table_name='file_content', postgresql_using='gin')
+    op.drop_index('content_embedding_idx', table_name='file_content', postgresql_using='hnsw', postgresql_with={'m': 16, 'ef_construction': 64}, postgresql_ops={'embedding': 'vector_cosine_ops'})
+    op.drop_table('file_content')
+    op.drop_index('files_tags_idx', table_name='files', postgresql_using='gin')
+    op.drop_index('files_path_hash_idx', table_name='files')
+    op.drop_table('files')
+    # ### end Alembic commands ###

app/main.py

@@ -1,14 +1,15 @@
 from fastapi import FastAPI
 from app.routes import search
 from app.routes import upload
-
+from app.routes import files 
 
 app = FastAPI(
     title="WhereTF Backend"
 )
 
 app.include_router(search.router)
 app.include_router(upload.router)
+app.include_router(files.router)
 
 @app.get("/health", tags=["System"])
 def health_check():

app/models/file.py

@@ -19,8 +19,10 @@ class File(Base):
     indexed_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now())
 
     tags: Mapped[list[str]] = mapped_column(ARRAY(Text), server_default='{}')
+
+    # NEWLY ADDED : The context column for storing user-defined descriptions
+    context: Mapped[str | None] = mapped_column(Text, nullable=True)
 
-    # Link to the contents table
     contents = relationship("FileContent", back_populates="file", cascade="all, delete-orphan")
 
     # Indexes from the image

app/routes/files.py

@@ -0,0 +1,69 @@
+# code for file-idexing,addition and updation of tags and deleting a specific file
+
+import uuid
+from fastapi import APIRouter, Depends, HTTPException
+from sqlalchemy.orm import Session
+from sqlalchemy import select
+from pydantic import BaseModel
+from typing import List, Optional
+
+from app.database import get_db
+from app.models import File
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+# --- Pydantic Schema for the PATCH request ---
+class FileUpdatePayload(BaseModel):
+    tags: Optional[List[str]] = None
+    context: Optional[str] = None
+
+# 1. THE DASHBOARD: Get all files
+@router.get("/")
+def get_all_files(db: Session = Depends(get_db)):
+    """Returns a list of all indexed files for the frontend UI."""
+    files = db.execute(select(File)).scalars().all()
+
+    return [
+        {
+            "id": str(f.id),
+            "file_path": f.file_path,
+            "mime_type": f.mime_type,
+            "tags": f.tags,
+            "context": f.context,
+            "last_modified": f.last_modified
+        }
+        for f in files
+    ]
+
+# 2. THE METADATA MANAGER: Update tags and context
+@router.patch("/{file_id}")
+def update_file_metadata(file_id: uuid.UUID, payload: FileUpdatePayload, db: Session = Depends(get_db)):
+    """Updates the user-defined tags and context for a specific file."""
+    db_file = db.scalar(select(File).where(File.id == file_id))
+
+    if not db_file:
+        raise HTTPException(status_code=404, detail="File not found")
+
+    if payload.tags is not None:
+        db_file.tags = payload.tags
+    if payload.context is not None:
+        db_file.context = payload.context
+
+    db.commit()
+    db.refresh(db_file)
+
+    return {"status": "success", "message": "Metadata updated", "file_id": str(db_file.id)}
+
+# 3. THE ERASER: Delete a file and its vectors
+@router.delete("/{file_id}")
+def delete_file(file_id: uuid.UUID, db: Session = Depends(get_db)):
+    """Deletes a file and all its associated AI chunks automatically."""
+    db_file = db.scalar(select(File).where(File.id == file_id))
+
+    if not db_file:
+        raise HTTPException(status_code=404, detail="File not found")
+
+    db.delete(db_file)
+    db.commit()
+
+    return {"status": "success", "message": f"File and vectors wiped successfully."}