footprinter-cli 1.0.0rc1__py3-none-any.whl

footprinter/cli/data.py

@@ -0,0 +1,879 @@
+"""fp data — export, template, and import commands for entity CSV data.
+
+Export current data:
+    ``fp data export clients``
+    ``fp data export files --status active --limit 100``
+
+Generate import-compatible templates:
+    ``fp data template clients``
+    ``fp data template files --file template.csv``
+
+Import metadata corrections for data-source entities:
+    ``fp data import files corrections.csv``
+    ``fp data import files corrections.csv --commit``
+"""
+
+import csv
+import sys
+from dataclasses import dataclass, field
+
+from footprinter.cli._common import FORMATTER, add_json_flag, console, open_db, output_json
+from footprinter.cli.upsert import CSV_COLUMNS
+
+# ---------------------------------------------------------------------------
+# Export column specs — derived from CSV_COLUMNS (import column spec)
+# ---------------------------------------------------------------------------
+
+#: Export columns per entity: required + optional from CSV_COLUMNS,
+#: minus client_id for projects (internal DB ID, not user-facing).
+EXPORT_COLUMNS: dict[str, list[str]] = {
+    "client": CSV_COLUMNS["client"][0] + CSV_COLUMNS["client"][1],
+    "project": [c for c in CSV_COLUMNS["project"][0] + CSV_COLUMNS["project"][1] if c != "client_id"],
+}
+
+# ---------------------------------------------------------------------------
+# Export SQL queries (clients/projects only — data-source uses registry)
+# ---------------------------------------------------------------------------
+
+
+def _export_query(entity_type: str, status_filter: str | None) -> tuple[str, list]:
+    """Return (sql, params) for an unbounded export query."""
+    params: list = []
+    if entity_type == "client":
+        sql = "SELECT name, client_type, slug, path_pattern, status FROM clients"
+        if status_filter:
+            sql += " WHERE status = ?"
+            params.append(status_filter)
+        else:
+            sql += " WHERE status != 'removed'"
+        sql += " ORDER BY name"
+    else:
+        sql = (
+            "SELECT p.project_name, p.root_path, "
+            "COALESCE(c.name, '') AS client, "
+            "p.project_type, p.description, p.github_url, p.status "
+            "FROM projects p LEFT JOIN clients c ON p.client_id = c.id"
+        )
+        if status_filter:
+            sql += " WHERE p.status = ?"
+            params.append(status_filter)
+        else:
+            sql += " WHERE p.status != 'removed'"
+        sql += " ORDER BY p.project_name"
+    return sql, params
+
+
+# ---------------------------------------------------------------------------
+# Template example rows (clients/projects only)
+# ---------------------------------------------------------------------------
+
+TEMPLATE_ROWS: dict[str, list[dict]] = {
+    "client": [
+        {
+            "name": "Acme Corp",
+            "client_type": "external",
+            "slug": "",
+            "path_pattern": "~/Work/clients/acme/",
+            "status": "active",
+        },
+        {"name": "Internal Tools", "client_type": "internal", "slug": "", "path_pattern": "", "status": "active"},
+        {"name": "Side Project", "client_type": "personal", "slug": "", "path_pattern": "", "status": "active"},
+    ],
+    "project": [
+        {
+            "project_name": "My Web App",
+            "root_path": "~/Work/projects/my-app",
+            "client": "Acme Corp",
+            "project_type": "python",
+            "description": "A web application",
+            "github_url": "",
+            "status": "active",
+        },
+        {
+            "project_name": "Documentation",
+            "root_path": "~/Work/docs",
+            "client": "",
+            "project_type": "docs",
+            "description": "Internal documentation",
+            "github_url": "",
+            "status": "active",
+        },
+        {
+            "project_name": "Mobile App",
+            "root_path": "~/Work/mobile",
+            "client": "Internal Tools",
+            "project_type": "typescript",
+            "description": "Mobile app",
+            "github_url": "",
+            "status": "active",
+        },
+    ],
+}
+
+VALID_VALUES_NOTES: dict[str, dict[str, str]] = {
+    "client": {
+        "client_type": "external, internal, personal",
+        "status": "active, hidden, removed",
+    },
+    "project": {
+        "status": "active, paused, completed, abandoned, removed",
+    },
+}
+
+
+# ---------------------------------------------------------------------------
+# Data-source entity registry
+# ---------------------------------------------------------------------------
+
+
+@dataclass
+class DataSourceSpec:
+    """Specification for a data-source entity's CSV operations."""
+
+    table: str
+    export_columns: list[str]
+    writable_columns: list[str]
+    order_by: str
+    has_status: bool
+    template_rows: list[dict] = field(default_factory=list)
+    valid_values: dict[str, str] = field(default_factory=dict)
+
+
+DATA_SOURCE_SPECS: dict[str, DataSourceSpec] = {
+    "files": DataSourceSpec(
+        table="files",
+        export_columns=[
+            "id",
+            "name",
+            "path",
+            "source",
+            "status",
+            "content_type",
+            "size_bytes",
+            "modified_at",
+            "project_id",
+            "client_id",
+            "mcp_view",
+            "mcp_read",
+        ],
+        writable_columns=["status", "project_id", "client_id", "mcp_view", "mcp_read"],
+        order_by="id",
+        has_status=True,
+        template_rows=[
+            {
+                "id": "1",
+                "name": "readme.md",
+                "path": "/Users/me/Work/readme.md",
+                "source": "local",
+                "status": "active",
+                "content_type": "markdown",
+                "size_bytes": "1024",
+                "modified_at": "2026-01-15T10:00:00Z",
+                "project_id": "1",
+                "client_id": "1",
+                "mcp_view": "visible",
+                "mcp_read": "allow",
+            },
+            {
+                "id": "2",
+                "name": "notes.txt",
+                "path": "/Users/me/Work/notes.txt",
+                "source": "local",
+                "status": "hidden",
+                "content_type": "text",
+                "size_bytes": "512",
+                "modified_at": "2026-02-01T10:00:00Z",
+                "project_id": "",
+                "client_id": "",
+                "mcp_view": "inherit",
+                "mcp_read": "inherit",
+            },
+        ],
+        valid_values={
+            "status": "active, hidden, removed",
+            "mcp_view": "hidden, opaque, visible, inherit",
+            "mcp_read": "allow, deny, inherit",
+        },
+    ),
+    "folders": DataSourceSpec(
+        table="folders",
+        export_columns=[
+            "id",
+            "path",
+            "relative_path",
+            "name",
+            "source",
+            "status",
+            "project_id",
+            "client_id",
+            "mcp_view",
+            "mcp_read",
+        ],
+        writable_columns=["status", "project_id", "client_id", "mcp_view", "mcp_read"],
+        order_by="id",
+        has_status=True,
+        template_rows=[
+            {
+                "id": "1",
+                "path": "/Users/me/Work",
+                "relative_path": "Work",
+                "name": "Work",
+                "source": "local",
+                "status": "active",
+                "project_id": "1",
+                "client_id": "",
+                "mcp_view": "visible",
+                "mcp_read": "allow",
+            },
+            {
+                "id": "2",
+                "path": "/Users/me/Personal",
+                "relative_path": "Personal",
+                "name": "Personal",
+                "source": "local",
+                "status": "active",
+                "project_id": "",
+                "client_id": "",
+                "mcp_view": "inherit",
+                "mcp_read": "inherit",
+            },
+        ],
+        valid_values={
+            "status": "active, hidden, removed",
+            "mcp_view": "hidden, opaque, visible, inherit",
+            "mcp_read": "allow, deny, inherit",
+        },
+    ),
+    "emails": DataSourceSpec(
+        table="emails",
+        export_columns=[
+            "id",
+            "message_id",
+            "account",
+            "subject",
+            "from_address",
+            "received_at",
+            "status",
+            "project_id",
+            "client_id",
+            "mcp_view",
+            "mcp_read",
+        ],
+        writable_columns=["status", "project_id", "client_id", "mcp_view", "mcp_read"],
+        order_by="id",
+        has_status=True,
+        template_rows=[
+            {
+                "id": "1",
+                "message_id": "msg-001@example.com",
+                "account": "work",
+                "subject": "Project Update",
+                "from_address": "sender@example.com",
+                "received_at": "2026-02-01T09:00:00Z",
+                "status": "active",
+                "project_id": "1",
+                "client_id": "1",
+                "mcp_view": "visible",
+                "mcp_read": "allow",
+            },
+            {
+                "id": "2",
+                "message_id": "msg-002@example.com",
+                "account": "personal",
+                "subject": "Newsletter",
+                "from_address": "news@example.com",
+                "received_at": "2026-02-02T09:00:00Z",
+                "status": "active",
+                "project_id": "",
+                "client_id": "",
+                "mcp_view": "inherit",
+                "mcp_read": "inherit",
+            },
+        ],
+        valid_values={
+            "status": "active, hidden, removed",
+            "mcp_view": "hidden, opaque, visible, inherit",
+            "mcp_read": "allow, deny, inherit",
+        },
+    ),
+    "chats": DataSourceSpec(
+        table="chats",
+        export_columns=[
+            "id",
+            "external_id",
+            "account",
+            "title",
+            "message_count",
+            "status",
+            "created_at",
+            "updated_at",
+            "project_id",
+            "client_id",
+            "mcp_view",
+            "mcp_read",
+        ],
+        writable_columns=["status", "project_id", "client_id", "mcp_view", "mcp_read"],
+        order_by="id",
+        has_status=True,
+        template_rows=[
+            {
+                "id": "1",
+                "external_id": "conv-001",
+                "account": "personal",
+                "title": "Architecture Chat",
+                "message_count": "5",
+                "status": "active",
+                "created_at": "2026-01-10T08:00:00Z",
+                "updated_at": "2026-01-10T09:00:00Z",
+                "project_id": "1",
+                "client_id": "1",
+                "mcp_view": "visible",
+                "mcp_read": "allow",
+            },
+            {
+                "id": "2",
+                "external_id": "conv-002",
+                "account": "personal",
+                "title": "Random Chat",
+                "message_count": "3",
+                "status": "active",
+                "created_at": "2026-01-11T08:00:00Z",
+                "updated_at": "2026-01-11T09:00:00Z",
+                "project_id": "",
+                "client_id": "",
+                "mcp_view": "inherit",
+                "mcp_read": "inherit",
+            },
+        ],
+        valid_values={
+            "status": "active, hidden, removed, merged",
+            "mcp_view": "hidden, opaque, visible, inherit",
+            "mcp_read": "allow, deny, inherit",
+        },
+    ),
+    "messages": DataSourceSpec(
+        table="messages",
+        export_columns=[
+            "id",
+            "chat_id",
+            "message_id",
+            "role",
+            "created_at",
+            "mcp_view",
+            "mcp_read",
+        ],
+        writable_columns=["mcp_view", "mcp_read"],
+        order_by="id",
+        has_status=False,
+        template_rows=[
+            {
+                "id": "1",
+                "chat_id": "1",
+                "message_id": "msg-1",
+                "role": "user",
+                "created_at": "2026-01-10T08:01:00Z",
+                "mcp_view": "visible",
+                "mcp_read": "allow",
+            },
+            {
+                "id": "2",
+                "chat_id": "1",
+                "message_id": "msg-2",
+                "role": "assistant",
+                "created_at": "2026-01-10T08:02:00Z",
+                "mcp_view": "visible",
+                "mcp_read": "allow",
+            },
+        ],
+        valid_values={
+            "mcp_view": "hidden, opaque, visible, inherit",
+            "mcp_read": "allow, deny, inherit",
+        },
+    ),
+    "visits": DataSourceSpec(
+        table="visits",
+        export_columns=[
+            "id",
+            "url",
+            "title",
+            "visit_time",
+            "browser",
+            "status",
+            "project_id",
+            "client_id",
+            "mcp_view",
+            "mcp_read",
+        ],
+        writable_columns=["status", "project_id", "client_id", "mcp_view", "mcp_read"],
+        order_by="id",
+        has_status=True,
+        template_rows=[
+            {
+                "id": "1",
+                "url": "https://example.com",
+                "title": "Example",
+                "visit_time": "2026-03-01T12:00:00Z",
+                "browser": "safari",
+                "status": "active",
+                "project_id": "1",
+                "client_id": "1",
+                "mcp_view": "visible",
+                "mcp_read": "allow",
+            },
+            {
+                "id": "2",
+                "url": "https://news.com",
+                "title": "News",
+                "visit_time": "2026-03-02T12:00:00Z",
+                "browser": "chrome",
+                "status": "active",
+                "project_id": "",
+                "client_id": "",
+                "mcp_view": "inherit",
+                "mcp_read": "inherit",
+            },
+        ],
+        valid_values={
+            "status": "active, hidden, removed",
+            "mcp_view": "hidden, opaque, visible, inherit",
+            "mcp_read": "allow, deny, inherit",
+        },
+    ),
+}
+
+#: All entity nouns accepted by export/template (clients, projects + data-source)
+ALL_EXPORT_NOUNS = ["clients", "projects"] + list(DATA_SOURCE_SPECS.keys())
+
+#: Entity nouns that support import (data-source only)
+IMPORT_NOUNS = list(DATA_SOURCE_SPECS.keys())
+
+
+# ---------------------------------------------------------------------------
+# Handlers
+# ---------------------------------------------------------------------------
+
+
+def _write_csv(columns: list[str], rows: list[dict], file_path: str | None) -> None:
+    """Write CSV to file or stdout."""
+    if file_path:
+        with open(file_path, "w", newline="") as f:
+            writer = csv.DictWriter(f, fieldnames=columns, extrasaction="ignore")
+            writer.writeheader()
+            for row in rows:
+                writer.writerow({k: v if (v := row.get(k)) is not None else "" for k in columns})
+    else:
+        writer = csv.DictWriter(sys.stdout, fieldnames=columns, extrasaction="ignore")
+        writer.writeheader()
+        for row in rows:
+            writer.writerow({k: v if (v := row.get(k)) is not None else "" for k in columns})
+
+
+def _handle_export(args) -> None:
+    """Handle ``fp data export <noun>``."""
+    noun = args.noun
+
+    # Data-source entities go through the registry
+    if noun in DATA_SOURCE_SPECS:
+        _handle_export_data_source(args)
+        return
+
+    # Existing client/project path
+    entity_type = "client" if noun == "clients" else "project"
+    columns = EXPORT_COLUMNS[entity_type]
+
+    status_filter = getattr(args, "status", None)
+    if status_filter:
+        valid = VALID_VALUES_NOTES.get(entity_type, {}).get("status", "")
+        valid_set = {v.strip() for v in valid.split(",")} if valid else set()
+        if valid_set and status_filter not in valid_set:
+            print(
+                f"Unknown status '{status_filter}'. Valid values: {', '.join(sorted(valid_set))}",
+                file=sys.stderr,
+            )
+            sys.exit(1)
+
+    sql, params = _export_query(entity_type, status_filter)
+
+    # Apply limit/offset if provided (OFFSET requires LIMIT in SQLite)
+    limit = getattr(args, "limit", None)
+    offset = getattr(args, "offset", None)
+    if limit is not None:
+        sql += " LIMIT ?"
+        params.append(limit)
+    elif offset is not None:
+        sql += " LIMIT -1"
+    if offset is not None:
+        sql += " OFFSET ?"
+        params.append(offset)
+
+    with open_db() as conn:
+        cur = conn.execute(sql, params)
+        rows = [dict(r) for r in cur.fetchall()]
+
+    _write_csv(columns, rows, getattr(args, "file", None))
+
+
+def _handle_export_data_source(args) -> None:
+    """Export a data-source entity via the registry."""
+    noun = args.noun
+    spec = DATA_SOURCE_SPECS[noun]
+    columns = spec.export_columns
+    status_filter = getattr(args, "status", None)
+
+    col_list = ", ".join(columns)
+    sql = f"SELECT {col_list} FROM {spec.table}"  # noqa: S608
+    params: list = []
+
+    # Validate status filter against known values
+    if status_filter and spec.has_status:
+        valid = spec.valid_values.get("status", "")
+        valid_set = {v.strip() for v in valid.split(",")} if valid else set()
+        if valid_set and status_filter not in valid_set:
+            print(
+                f"Unknown status '{status_filter}'. Valid values: {', '.join(sorted(valid_set))}",
+                file=sys.stderr,
+            )
+            sys.exit(1)
+
+    # Default: exclude removed rows (for entities with status)
+    if spec.has_status:
+        if status_filter:
+            sql += " WHERE status = ?"
+            params.append(status_filter)
+        else:
+            sql += " WHERE status != 'removed'"
+    elif status_filter:
+        print(
+            f"Entity '{noun}' does not have a status column.",
+            file=sys.stderr,
+        )
+        sys.exit(1)
+
+    sql += f" ORDER BY {spec.order_by}"
+
+    # Apply limit/offset (OFFSET requires LIMIT in SQLite)
+    limit = getattr(args, "limit", None)
+    offset = getattr(args, "offset", None)
+    if limit is not None:
+        sql += " LIMIT ?"
+        params.append(limit)
+    elif offset is not None:
+        sql += " LIMIT -1"
+    if offset is not None:
+        sql += " OFFSET ?"
+        params.append(offset)
+
+    with open_db() as conn:
+        cur = conn.execute(sql, params)
+        rows = [dict(r) for r in cur.fetchall()]
+
+    _write_csv(columns, rows, getattr(args, "file", None))
+
+
+def _handle_template(args) -> None:
+    """Handle ``fp data template <noun>``."""
+    noun = args.noun
+
+    # Data-source entities go through the registry
+    if noun in DATA_SOURCE_SPECS:
+        spec = DATA_SOURCE_SPECS[noun]
+        _write_csv(spec.export_columns, spec.template_rows, getattr(args, "file", None))
+        notes = spec.valid_values
+        if notes:
+            print("\nValid values:", file=sys.stderr)
+            for fld, values in notes.items():
+                print(f"  {fld}: {values}", file=sys.stderr)
+        return
+
+    # Existing client/project path
+    entity_type = "client" if noun == "clients" else "project"
+    columns = EXPORT_COLUMNS[entity_type]
+    rows = TEMPLATE_ROWS[entity_type]
+
+    _write_csv(columns, rows, getattr(args, "file", None))
+
+    # Print valid value notes to stderr
+    notes = VALID_VALUES_NOTES.get(entity_type, {})
+    if notes:
+        print("\nValid values:", file=sys.stderr)
+        for fld, values in notes.items():
+            print(f"  {fld}: {values}", file=sys.stderr)
+
+
+def _handle_import(args) -> None:
+    """Handle ``fp data import <noun> <file>``."""
+    from pathlib import Path
+
+    from rich.table import Table
+
+    noun = args.noun
+    spec = DATA_SOURCE_SPECS[noun]
+    csv_path = Path(args.file)
+    has_dry_run = getattr(args, "dry_run", False)
+    has_commit = getattr(args, "commit", False)
+    if has_dry_run and has_commit:
+        console.print("[red]Cannot use --dry-run and --commit together.[/red]")
+        sys.exit(1)
+    dry_run = not has_commit
+
+    # Read and validate CSV
+    if not csv_path.exists():
+        console.print(f"[red]File not found: {csv_path}[/red]")
+        sys.exit(1)
+
+    with open(csv_path, encoding="utf-8", newline="") as f:
+        reader = csv.DictReader(f)
+        if reader.fieldnames is None:
+            console.print("[red]Empty or invalid CSV file.[/red]")
+            sys.exit(1)
+        if "id" not in reader.fieldnames:
+            console.print("[red]CSV must contain an 'id' column.[/red]")
+            sys.exit(1)
+        rows = list(reader)
+
+    if not rows:
+        if getattr(args, "json", False):
+            output_json({"total": 0, "updated": 0, "skipped": 0, "errors": 0})
+        else:
+            console.print("[dim]No rows in CSV — nothing to do.[/dim]")
+        return
+
+    # Determine which writable columns are present in the CSV
+    csv_writable = [c for c in spec.writable_columns if c in reader.fieldnames]
+
+    if not csv_writable:
+        console.print(
+            f"[red]No writable columns found in CSV. "
+            f"Writable columns for {noun}: {', '.join(spec.writable_columns)}[/red]"
+        )
+        sys.exit(1)
+
+    # Process rows
+    updated = 0
+    skipped = 0
+    errors = 0
+    error_details: list[dict] = []
+
+    with open_db() as conn:
+        # Begin ingest tracking before data writes (matches upsert.py pattern)
+        ingest_svc = None
+        ingest_id = None
+        if not dry_run:
+            from footprinter.services.ingest_service import IngestService
+
+            ingest_svc = IngestService(conn)
+            ingest_id = ingest_svc.begin(
+                f"import_{noun}",
+                mode="bulk",
+                trigger="cli:data:import",
+            )
+
+        for i, row in enumerate(rows, 1):
+            row_id = row.get("id", "").strip()
+            if not row_id:
+                errors += 1
+                error_details.append({"row": i, "error": "Missing id value"})
+                continue
+
+            try:
+                row_id_int = int(row_id)
+            except ValueError:
+                errors += 1
+                error_details.append({"row": i, "error": f"Invalid id: {row_id!r}"})
+                continue
+
+            # Check record exists
+            existing = conn.execute(
+                f"SELECT id FROM {spec.table} WHERE id = ?",  # noqa: S608
+                (row_id_int,),
+            ).fetchone()
+            if existing is None:
+                errors += 1
+                error_details.append({"row": i, "error": f"ID {row_id_int} not found"})
+                continue
+
+            # Build SET clause from writable columns with non-empty values
+            set_parts: list[str] = []
+            set_params: list = []
+            for col in csv_writable:
+                val = row.get(col, "")
+                if val == "":
+                    continue  # Empty = skip (preserve existing)
+                # Sentinel: "0" for project_id/client_id clears to NULL
+                if col in ("project_id", "client_id") and val == "0":
+                    set_parts.append(f"{col} = ?")
+                    set_params.append(None)
+                else:
+                    set_parts.append(f"{col} = ?")
+                    set_params.append(val)
+
+            if not set_parts:
+                skipped += 1
+                continue
+
+            if dry_run:
+                updated += 1
+            else:
+                update_sql = (
+                    f"UPDATE {spec.table} SET {', '.join(set_parts)} "  # noqa: S608
+                    f"WHERE id = ?"
+                )
+                set_params.append(row_id_int)
+                conn.execute(update_sql, set_params)
+                updated += 1
+
+        # Complete ingest tracking and commit everything together
+        if ingest_svc is not None and ingest_id is not None:
+            ingest_svc.complete(
+                ingest_id,
+                result={
+                    "items_processed": updated + skipped + errors,
+                    "items_updated": updated,
+                    "items_skipped": skipped,
+                    "errors": errors,
+                },
+                metadata={"error_details": error_details} if error_details else None,
+            )
+
+    summary: dict = {
+        "total": updated + skipped + errors,
+        "updated": updated,
+        "skipped": skipped,
+        "errors": errors,
+    }
+    if error_details:
+        summary["error_details"] = error_details
+
+    if getattr(args, "json", False):
+        output_json(summary)
+    elif dry_run:
+        table = Table(title=f"Dry run — import {noun}")
+        table.add_column("Metric", style="cyan")
+        table.add_column("Count", justify="right")
+        table.add_row("Would update", str(updated))
+        table.add_row("Skipped (no changes)", str(skipped))
+        table.add_row("Errors", str(errors))
+        table.add_row("Total rows", str(updated + skipped + errors))
+        console.print(table)
+        if error_details:
+            for ed in error_details:
+                console.print(f"  [red]Row {ed['row']}: {ed['error']}[/red]")
+        console.print("[dim]Pass --commit to apply these changes.[/dim]")
+    else:
+        table = Table(title=f"Import {noun}")
+        table.add_column("Metric", style="cyan")
+        table.add_column("Count", justify="right")
+        table.add_row("Updated", str(updated))
+        table.add_row("Skipped (no changes)", str(skipped))
+        table.add_row("Errors", str(errors))
+        table.add_row("Total rows", str(updated + skipped + errors))
+        console.print(table)
+        if error_details:
+            for ed in error_details:
+                console.print(f"  [red]Row {ed['row']}: {ed['error']}[/red]")
+
+
+# ---------------------------------------------------------------------------
+# Registration
+# ---------------------------------------------------------------------------
+
+
+def register(subparsers) -> None:
+    """Register ``fp data`` with ``export``, ``template``, and ``import`` subcommands."""
+    data_parser = subparsers.add_parser(
+        "data",
+        help="Export data, generate templates, or import metadata corrections",
+        formatter_class=FORMATTER,
+    )
+    data_parser.set_defaults(func=lambda args: data_parser.print_help())
+    data_sub = data_parser.add_subparsers(dest="data_action", metavar="ACTION")
+
+    # -- fp data export ---------------------------------------------------
+    export_parser = data_sub.add_parser(
+        "export",
+        help="Export entity data as CSV",
+        formatter_class=FORMATTER,
+    )
+    export_parser.add_argument(
+        "noun",
+        choices=ALL_EXPORT_NOUNS,
+        help="Entity type to export",
+    )
+    export_parser.add_argument(
+        "--file",
+        default=None,
+        help="Write output to file instead of stdout",
+    )
+    export_parser.add_argument(
+        "--status",
+        default=None,
+        help="Filter by status (e.g., active)",
+    )
+    export_parser.add_argument(
+        "--limit",
+        type=int,
+        default=None,
+        help="Maximum number of rows to export",
+    )
+    export_parser.add_argument(
+        "--offset",
+        type=int,
+        default=None,
+        help="Number of rows to skip before exporting",
+    )
+    export_parser.set_defaults(func=_handle_export)
+
+    # -- fp data template -------------------------------------------------
+    template_parser = data_sub.add_parser(
+        "template",
+        help="Generate an import-compatible CSV template",
+        formatter_class=FORMATTER,
+    )
+    template_parser.add_argument(
+        "noun",
+        choices=ALL_EXPORT_NOUNS,
+        help="Entity type for template",
+    )
+    template_parser.add_argument(
+        "--file",
+        default=None,
+        help="Write template to file instead of stdout",
+    )
+    template_parser.set_defaults(func=_handle_template)
+
+    # -- fp data import ---------------------------------------------------
+    import_parser = data_sub.add_parser(
+        "import",
+        help="Import metadata corrections from CSV",
+        description=(
+            "Import metadata corrections for data-source entities.\n\n"
+            "Reads a CSV file with an 'id' column and updates writable metadata\n"
+            "columns. Pipeline-managed fields (path, external_id, etc.) are\n"
+            "read-only and ignored during import.\n\n"
+            "Default mode is dry-run (preview only). Pass --commit to apply."
+        ),
+        formatter_class=FORMATTER,
+    )
+    import_parser.add_argument(
+        "noun",
+        choices=IMPORT_NOUNS,
+        help="Entity type to import",
+    )
+    import_parser.add_argument(
+        "file",
+        help="Path to CSV file",
+    )
+    import_parser.add_argument(
+        "--dry-run",
+        action="store_true",
+        default=False,
+        help="Preview changes without writing (default behavior)",
+    )
+    import_parser.add_argument(
+        "--commit",
+        action="store_true",
+        default=False,
+        help="Apply changes to the database",
+    )
+    add_json_flag(import_parser)
+    import_parser.set_defaults(func=_handle_import)