unitysvc-services 0.1.24__py3-none-any.whl

unitysvc_services/publisher.py

@@ -0,0 +1,1628 @@
+"""Data publisher module for posting service data to UnitySVC backend."""
+
+import asyncio
+import base64
+import json
+import os
+from pathlib import Path
+from typing import Any
+
+import httpx
+import typer
+from rich.console import Console
+from rich.table import Table
+
+import unitysvc_services
+
+from .api import UnitySvcAPI
+from .models.base import ListingStatusEnum, ProviderStatusEnum, SellerStatusEnum
+from .utils import convert_convenience_fields_to_documents, find_files_by_schema, load_data_file, render_template_file
+from .validator import DataValidator
+
+
+class ServiceDataPublisher(UnitySvcAPI):
+    """Publishes service data to UnitySVC backend endpoints.
+
+    Inherits base HTTP client with curl fallback from UnitySvcAPI.
+    Extends with async operations for concurrent publishing.
+    """
+
+    def __init__(self) -> None:
+        # Initialize base class (provides self.client as AsyncClient with curl fallback)
+        super().__init__()
+
+        # Semaphore to limit concurrent requests and prevent connection pool exhaustion
+        self.max_concurrent_requests = 15
+
+    def load_file_content(self, file_path: Path, base_path: Path) -> str:
+        """Load content from a file (text or binary)."""
+        full_path = base_path / file_path if not file_path.is_absolute() else file_path
+
+        if not full_path.exists():
+            raise FileNotFoundError(f"File not found: {full_path}")
+
+        # Try to read as text first
+        try:
+            with open(full_path, encoding="utf-8") as f:
+                return f.read()
+        except UnicodeDecodeError:
+            # If it fails, read as binary and encode as base64
+            with open(full_path, "rb") as f:
+                return base64.b64encode(f.read()).decode("ascii")
+
+    def resolve_file_references(
+        self,
+        data: dict[str, Any],
+        base_path: Path,
+        listing: dict[str, Any] | None = None,
+        offering: dict[str, Any] | None = None,
+        provider: dict[str, Any] | None = None,
+        seller: dict[str, Any] | None = None,
+        listing_filename: str | None = None,
+        interface: dict[str, Any] | None = None,
+    ) -> dict[str, Any]:
+        """Recursively resolve file references and include content in data.
+
+        For Jinja2 template files (.j2), renders the template with provided context
+        and strips the .j2 extension from file_path.
+
+        Args:
+            data: Data dictionary potentially containing file_path references
+            base_path: Base path for resolving relative file paths
+            listing: Listing data for template rendering (optional)
+            offering: Offering data for template rendering (optional)
+            provider: Provider data for template rendering (optional)
+            seller: Seller data for template rendering (optional)
+            listing_filename: Listing filename for constructing output filenames (optional)
+            interface: AccessInterface data for template rendering (optional, for interface documents)
+
+        Returns:
+            Data with file references resolved and content loaded
+        """
+        result: dict[str, Any] = {}
+
+        # Check if this dict looks like an AccessInterface (has base_url or interface_type)
+        # If so, use it as the interface context for nested documents
+        current_interface = interface
+        if "base_url" in data or "interface_type" in data:
+            current_interface = data
+
+        for key, value in data.items():
+            if isinstance(value, dict):
+                # Recursively process nested dictionaries
+                result[key] = self.resolve_file_references(
+                    value,
+                    base_path,
+                    listing=listing,
+                    offering=offering,
+                    provider=provider,
+                    seller=seller,
+                    listing_filename=listing_filename,
+                    interface=current_interface,
+                )
+            elif isinstance(value, list):
+                # Process lists
+                result[key] = [
+                    (
+                        self.resolve_file_references(
+                            item,
+                            base_path,
+                            listing=listing,
+                            offering=offering,
+                            provider=provider,
+                            seller=seller,
+                            listing_filename=listing_filename,
+                            interface=current_interface,
+                        )
+                        if isinstance(item, dict)
+                        else item
+                    )
+                    for item in value
+                ]
+            elif key == "file_path" and isinstance(value, str):
+                # This is a file reference - load the content and render if template
+                full_path = base_path / value if not Path(value).is_absolute() else Path(value)
+
+                if not full_path.exists():
+                    raise FileNotFoundError(f"File not found: {full_path}")
+
+                # Render template if applicable
+                try:
+                    content, actual_filename = render_template_file(
+                        full_path,
+                        listing=listing,
+                        offering=offering,
+                        provider=provider,
+                        seller=seller,
+                        interface=current_interface,
+                    )
+                    result["file_content"] = content
+
+                    # Update file_path to remove .j2 extension if it was a template
+                    if full_path.name.endswith(".j2"):
+                        # Strip .j2 from the path
+                        new_path = str(value)[:-3]  # Remove last 3 characters (.j2)
+                        result[key] = new_path
+                    else:
+                        result[key] = value
+
+                except Exception as e:
+                    raise ValueError(f"Failed to load/render file content from '{value}': {e}")
+            else:
+                result[key] = value
+
+        # After processing all fields, check if this is a code_examples document
+        # If so, try to load corresponding .out file and add to meta.output
+        if result.get("category") == "code_examples" and result.get("file_content") and listing_filename:
+            # Get the actual filename (after .j2 stripping if applicable)
+            # If file_path was updated (e.g., from "test.py.j2" to "test.py"), use that
+            # Otherwise, extract basename from original file_path
+            output_base_filename: str | None = None
+
+            # Check if file_path was modified (original might have had .j2)
+            file_path_value = result.get("file_path", "")
+            if file_path_value:
+                output_base_filename = Path(file_path_value).name
+
+            if output_base_filename:
+                # Construct output filename: {listing_stem}_{output_base_filename}.out
+                # e.g., "svclisting_test.py.out" for svclisting.json and test.py
+                listing_stem = Path(listing_filename).stem
+                output_filename = f"{listing_stem}_{output_base_filename}.out"
+
+                # Try to find the .out file in base_path (listing directory)
+                output_path = base_path / output_filename
+
+                if output_path.exists():
+                    try:
+                        with open(output_path, encoding="utf-8") as f:
+                            output_content = f.read()
+
+                        # Add output to meta field
+                        if "meta" not in result or result["meta"] is None:
+                            result["meta"] = {}
+                        result["meta"]["output"] = output_content
+                    except Exception:
+                        # Don't fail if output file can't be read, just skip it
+                        pass
+
+        return result
+
+    async def post(  # type: ignore[override]
+        self, endpoint: str, data: dict[str, Any], check_status: bool = True, dryrun: bool = False
+    ) -> tuple[dict[str, Any], int]:
+        """Make a POST request to the backend API with automatic curl fallback.
+
+        Override of base class post() that returns both JSON and status code.
+        Uses base class client with automatic curl fallback.
+
+        Args:
+            endpoint: API endpoint path (e.g., "/publish/seller")
+            data: JSON data to post
+            check_status: Whether to raise on non-2xx status codes (default: True)
+            dryrun: If True, adds dryrun=true as query parameter
+
+        Returns:
+            Tuple of (JSON response, HTTP status code)
+
+        Raises:
+            RuntimeError: If both httpx and curl fail
+        """
+        # Build query parameters
+        params = {"dryrun": "true"} if dryrun else None
+
+        # Use base class client (self.client from UnitySvcQuery) with automatic curl fallback
+        # If we already know curl is needed, use it directly
+        if self.use_curl_fallback:
+            # Use base class curl fallback method
+            response_json = await super().post(endpoint, json_data=data, params=params)
+            # Curl POST doesn't return status code separately, assume 2xx if no exception
+            status_code = 200
+        else:
+            try:
+                response = await self.client.post(f"{self.base_url}{endpoint}", json=data, params=params)
+                status_code = response.status_code
+
+                if check_status:
+                    response.raise_for_status()
+
+                response_json = response.json()
+            except (httpx.ConnectError, OSError):
+                # Connection failed - switch to curl fallback and retry
+                self.use_curl_fallback = True
+                response_json = await super().post(endpoint, json_data=data, params=params)
+                status_code = 200  # Assume success if curl didn't raise
+
+        return (response_json, status_code)
+
+    async def _post_with_retry(
+        self,
+        endpoint: str,
+        data: dict[str, Any],
+        entity_type: str,
+        entity_name: str,
+        context_info: str = "",
+        max_retries: int = 3,
+        dryrun: bool = False,
+    ) -> dict[str, Any]:
+        """
+        Generic retry wrapper for posting data to backend API with task polling.
+
+        The backend now returns HTTP 202 with a task_id. This method:
+        1. Submits the publish request
+        2. Gets the task_id from the response
+        3. Polls /tasks/{task_id} until completion
+        4. Returns the final result
+
+        Args:
+            endpoint: API endpoint path (e.g., "/publish/listing")
+            data: JSON data to post
+            entity_type: Type of entity being published (for error messages)
+            entity_name: Name of the entity being published (for error messages)
+            context_info: Additional context for error messages (e.g., provider, service info)
+            max_retries: Maximum number of retry attempts
+            dryrun: If True, runs in dry run mode (no actual changes)
+
+        Returns:
+            Response JSON from successful API call
+
+        Raises:
+            ValueError: On client errors (4xx) or after exhausting retries
+        """
+        last_exception = None
+        for attempt in range(max_retries):
+            try:
+                # Use the public post() method with automatic curl fallback
+                response_json, status_code = await self.post(endpoint, data, check_status=False, dryrun=dryrun)
+
+                # Handle task-based response (HTTP 202)
+                if status_code == 202:
+                    # Backend returns task_id - poll for completion
+                    task_id = response_json.get("task_id")
+
+                    if not task_id:
+                        context_msg = f" ({context_info})" if context_info else ""
+                        raise ValueError(f"No task_id in response for {entity_type} '{entity_name}'{context_msg}")
+
+                    # Poll task status until completion using check_task utility
+                    try:
+                        result = await self.check_task(task_id)
+                        return result
+                    except ValueError as e:
+                        # Add context to task errors
+                        context_msg = f" ({context_info})" if context_info else ""
+                        raise ValueError(f"Task failed for {entity_type} '{entity_name}'{context_msg}: {e}")
+
+                # Check for errors
+                if status_code >= 400:
+                    # Don't retry on 4xx errors (client errors) - they won't succeed on retry
+                    if 400 <= status_code < 500:
+                        error_detail = response_json.get("detail", str(response_json))
+                        context_msg = f" ({context_info})" if context_info else ""
+                        raise ValueError(
+                            f"Failed to publish {entity_type} '{entity_name}'{context_msg}: {error_detail}"
+                        )
+
+                    # 5xx errors - retry with exponential backoff
+                    if attempt < max_retries - 1:
+                        wait_time = 2**attempt  # Exponential backoff: 1s, 2s, 4s
+                        await asyncio.sleep(wait_time)
+                        continue
+                    else:
+                        # Last attempt failed
+                        error_detail = response_json.get("detail", str(response_json))
+                        context_msg = f" ({context_info})" if context_info else ""
+                        raise ValueError(
+                            f"Failed to publish {entity_type} after {max_retries} attempts: "
+                            f"'{entity_name}'{context_msg}: {error_detail}"
+                        )
+
+                # Success response (2xx)
+                return response_json
+
+            except (httpx.NetworkError, httpx.TimeoutException, RuntimeError) as e:
+                # Network/connection errors - the post() method should have tried curl fallback
+                # If we're here, both httpx and curl failed
+                last_exception = e
+                if attempt < max_retries - 1:
+                    wait_time = 2**attempt  # Exponential backoff: 1s, 2s, 4s
+                    await asyncio.sleep(wait_time)
+                    continue
+                else:
+                    raise ValueError(
+                        f"Network error after {max_retries} attempts for {entity_type} '{entity_name}': {str(e)}"
+                    )
+
+        # Should never reach here, but just in case
+        if last_exception:
+            raise last_exception
+        raise ValueError("Unexpected error in retry logic")
+
+    async def post_service_listing_async(
+        self, listing_file: Path, max_retries: int = 3, dryrun: bool = False
+    ) -> dict[str, Any]:
+        """Async version of post_service_listing for concurrent publishing with retry logic."""
+        # Load the listing data file
+        data, _ = load_data_file(listing_file)
+
+        # If name is not provided, use filename (without extension)
+        if "name" not in data or not data.get("name"):
+            data["name"] = listing_file.stem
+
+        # Extract provider_name from directory structure (needed before loading provider data)
+        parts = listing_file.parts
+        try:
+            services_idx = parts.index("services")
+            provider_name = parts[services_idx - 1]
+            data["provider_name"] = provider_name
+
+            # Find provider directory to load provider data
+            provider_dir = Path(*parts[:services_idx])
+        except (ValueError, IndexError):
+            raise ValueError(
+                f"Cannot extract provider_name from path: {listing_file}. "
+                f"Expected path to contain .../{{provider_name}}/services/..."
+            )
+
+        # If service_name is not in listing data, find it from service files in the same directory
+        if "service_name" not in data or not data["service_name"]:
+            # Find all service files in the same directory
+            service_files = find_files_by_schema(listing_file.parent, "service_v1")
+
+            if len(service_files) == 0:
+                raise ValueError(
+                    f"Cannot find any service_v1 files in {listing_file.parent}. "
+                    f"Listing files must be in the same directory as a service definition."
+                )
+            elif len(service_files) > 1:
+                service_names = [svc_data.get("name", "unknown") for _, _, svc_data in service_files]
+                raise ValueError(
+                    f"Multiple services found in {listing_file.parent}: {', '.join(service_names)}. "
+                    f"Please add 'service_name' field to {listing_file.name} to specify which "
+                    f"service this listing belongs to."
+                )
+            else:
+                # Exactly one service found - use it
+                _service_file, _format, service_data = service_files[0]
+                data["service_name"] = service_data.get("name")
+                data["service_version"] = service_data.get("version")
+        else:
+            # service_name is provided in listing data, find the matching service to get version
+            service_name = data["service_name"]
+            service_files = find_files_by_schema(
+                listing_file.parent, "service_v1", field_filter=(("name", service_name),)
+            )
+
+            if not service_files:
+                raise ValueError(
+                    f"Service '{service_name}' specified in {listing_file.name} not found in {listing_file.parent}."
+                )
+
+            # Get version from the found service
+            _service_file, _format, service_data = service_files[0]
+            data["service_version"] = service_data.get("version")
+
+        # Load provider data for template rendering
+        provider_files = find_files_by_schema(provider_dir, "provider_v1")
+        if provider_files:
+            _provider_file, _format, provider_data = provider_files[0]
+        else:
+            provider_data = {}
+
+        # Find seller_name from seller definition in the data directory
+        # Navigate up to find the data directory and look for seller file
+        data_dir = listing_file.parent
+        while data_dir.name != "data" and data_dir.parent != data_dir:
+            data_dir = data_dir.parent
+
+        if data_dir.name != "data":
+            raise ValueError(
+                f"Cannot find 'data' directory in path: {listing_file}. "
+                f"Expected path structure includes a 'data' directory."
+            )
+
+        # Look for seller file in the data directory by checking schema field
+        seller_files = find_files_by_schema(data_dir, "seller_v1")
+
+        if not seller_files:
+            raise ValueError(
+                f"Cannot find seller_v1 file in {data_dir}. A seller definition is required in the data directory."
+            )
+
+        # Should only be one seller file in the data directory
+        _seller_file, _format, seller_data = seller_files[0]
+
+        # Check seller status - skip if draft
+        seller_status = seller_data.get("status", SellerStatusEnum.active)
+        if seller_status == SellerStatusEnum.draft:
+            return {
+                "skipped": True,
+                "reason": f"Seller status is '{seller_status}' - not publishing listing to backend",
+                "name": data.get("name", "unknown"),
+            }
+
+        seller_name = seller_data.get("name")
+        if not seller_name:
+            raise ValueError("Seller data missing 'name' field")
+
+        data["seller_name"] = seller_name
+
+        # Map listing_status to status if present
+        if "listing_status" in data:
+            data["status"] = data.pop("listing_status")
+
+        # Check listing status - skip if draft
+        listing_status = data.get("status", ListingStatusEnum.ready)
+        if listing_status == ListingStatusEnum.draft:
+            return {
+                "skipped": True,
+                "reason": f"Listing status is '{listing_status}' - not publishing to backend (still in draft)",
+                "name": data.get("name", "unknown"),
+            }
+
+        # NOW resolve file references with all context (listing, offering, provider, seller)
+        base_path = listing_file.parent
+        data_with_content = self.resolve_file_references(
+            data,
+            base_path,
+            listing=data,
+            offering=service_data,
+            provider=provider_data,
+            seller=seller_data,
+            listing_filename=listing_file.name,
+        )
+
+        # Post to the endpoint using retry helper
+        context_info = (
+            f"service: {data_with_content.get('service_name')}, "
+            f"provider: {data_with_content.get('provider_name')}, "
+            f"seller: {data_with_content.get('seller_name')}"
+        )
+        result = await self._post_with_retry(
+            endpoint="/publish/listing",
+            data=data_with_content,
+            entity_type="listing",
+            entity_name=data.get("name", "unknown"),
+            context_info=context_info,
+            max_retries=max_retries,
+            dryrun=dryrun,
+        )
+
+        # Add local metadata to result for display purposes
+        result["service_name"] = data_with_content.get("service_name")
+        result["provider_name"] = data_with_content.get("provider_name")
+        result["seller_name"] = data_with_content.get("seller_name")
+
+        return result
+
+    async def post_service_offering_async(
+        self, data_file: Path, max_retries: int = 3, dryrun: bool = False
+    ) -> dict[str, Any]:
+        """Async version of post_service_offering for concurrent publishing with retry logic."""
+        # Load the data file
+        data, _ = load_data_file(data_file)
+
+        # Convert convenience fields first
+        base_path = data_file.parent
+        data = convert_convenience_fields_to_documents(
+            data, base_path, logo_field="logo", terms_field="terms_of_service"
+        )
+
+        # Extract provider_name from directory structure
+        # Find the 'services' directory and use its parent as provider_name
+        parts = data_file.parts
+        try:
+            services_idx = parts.index("services")
+            provider_name = parts[services_idx - 1]
+            data["provider_name"] = provider_name
+
+            # Find provider directory to check status and load data
+            provider_dir = Path(*parts[:services_idx])
+        except (ValueError, IndexError):
+            raise ValueError(
+                f"Cannot extract provider_name from path: {data_file}. "
+                f"Expected path to contain .../{{provider_name}}/services/..."
+            )
+
+        # Load provider data for status check and template rendering
+        provider_files = find_files_by_schema(provider_dir, "provider_v1")
+        if provider_files:
+            # Should only be one provider file in the directory
+            _provider_file, _format, provider_data = provider_files[0]
+            provider_status = provider_data.get("status", ProviderStatusEnum.active)
+            if provider_status == ProviderStatusEnum.draft:
+                return {
+                    "skipped": True,
+                    "reason": f"Provider status is '{provider_status}' - not publishing offering to backend",
+                    "name": data.get("name", "unknown"),
+                }
+        else:
+            provider_data = {}
+
+        # NOW resolve file references with all context (offering, provider)
+        data_with_content = self.resolve_file_references(
+            data,
+            base_path,
+            listing=None,
+            offering=data,
+            provider=provider_data,
+            seller=None,
+        )
+
+        # Post to the endpoint using retry helper
+        context_info = f"provider: {data_with_content.get('provider_name')}"
+        result = await self._post_with_retry(
+            endpoint="/publish/offering",
+            data=data_with_content,
+            entity_type="offering",
+            entity_name=data.get("name", "unknown"),
+            context_info=context_info,
+            max_retries=max_retries,
+            dryrun=dryrun,
+        )
+
+        # Add local metadata to result for display purposes
+        result["provider_name"] = data_with_content.get("provider_name")
+
+        return result
+
+    async def post_provider_async(self, data_file: Path, max_retries: int = 3, dryrun: bool = False) -> dict[str, Any]:
+        """Async version of post_provider for concurrent publishing with retry logic."""
+        # Load the data file
+        data, _ = load_data_file(data_file)
+
+        # Check provider status - skip if draft
+        provider_status = data.get("status", ProviderStatusEnum.active)
+        if provider_status == ProviderStatusEnum.draft:
+            # Return success without publishing - provider is incomplete
+            return {
+                "skipped": True,
+                "reason": f"Provider status is '{provider_status}' - not publishing to backend",
+                "name": data.get("name", "unknown"),
+            }
+
+        # Convert convenience fields (logo, terms_of_service) to documents
+        base_path = data_file.parent
+        data = convert_convenience_fields_to_documents(
+            data, base_path, logo_field="logo", terms_field="terms_of_service"
+        )
+
+        # Resolve file references and include content with provider context
+        data_with_content = self.resolve_file_references(
+            data,
+            base_path,
+            listing=None,
+            offering=None,
+            provider=data,
+            seller=None,
+        )
+
+        # Post to the endpoint using retry helper
+        return await self._post_with_retry(
+            endpoint="/publish/provider",
+            data=data_with_content,
+            entity_type="provider",
+            entity_name=data.get("name", "unknown"),
+            max_retries=max_retries,
+            dryrun=dryrun,
+        )
+
+    async def post_seller_async(self, data_file: Path, max_retries: int = 3, dryrun: bool = False) -> dict[str, Any]:
+        """Async version of post_seller for concurrent publishing with retry logic."""
+        # Load the data file
+        data, _ = load_data_file(data_file)
+
+        # Check seller status - skip if draft
+        seller_status = data.get("status", SellerStatusEnum.active)
+        if seller_status == SellerStatusEnum.draft:
+            # Return success without publishing - seller is incomplete
+            return {
+                "skipped": True,
+                "reason": f"Seller status is '{seller_status}' - not publishing to backend",
+                "name": data.get("name", "unknown"),
+            }
+
+        # Convert convenience fields (logo only for sellers, no terms_of_service)
+        base_path = data_file.parent
+        data = convert_convenience_fields_to_documents(data, base_path, logo_field="logo", terms_field=None)
+
+        # Resolve file references and include content with seller context
+        data_with_content = self.resolve_file_references(
+            data,
+            base_path,
+            listing=None,
+            offering=None,
+            provider=None,
+            seller=data,
+        )
+
+        # Post to the endpoint using retry helper
+        return await self._post_with_retry(
+            endpoint="/publish/seller",
+            data=data_with_content,
+            entity_type="seller",
+            entity_name=data.get("name", "unknown"),
+            max_retries=max_retries,
+            dryrun=dryrun,
+        )
+
+    def find_offering_files(self, data_dir: Path) -> list[Path]:
+        """Find all service offering files in a directory tree."""
+        files = find_files_by_schema(data_dir, "service_v1")
+        return sorted([f[0] for f in files])
+
+    def find_listing_files(self, data_dir: Path) -> list[Path]:
+        """Find all service listing files in a directory tree."""
+        files = find_files_by_schema(data_dir, "listing_v1")
+        return sorted([f[0] for f in files])
+
+    def find_provider_files(self, data_dir: Path) -> list[Path]:
+        """Find all provider files in a directory tree."""
+        files = find_files_by_schema(data_dir, "provider_v1")
+        return sorted([f[0] for f in files])
+
+    def find_seller_files(self, data_dir: Path) -> list[Path]:
+        """Find all seller files in a directory tree."""
+        files = find_files_by_schema(data_dir, "seller_v1")
+        return sorted([f[0] for f in files])
+
+    @staticmethod
+    def _get_status_display(status: str) -> tuple[str, str]:
+        """Get color and symbol for status display."""
+        status_map = {
+            "created": ("[green]+[/green]", "green"),
+            "updated": ("[blue]~[/blue]", "blue"),
+            "unchanged": ("[dim]=[/dim]", "dim"),
+            "create": ("[yellow]?[/yellow]", "yellow"),  # Dryrun: would be created
+            "update": ("[cyan]?[/cyan]", "cyan"),  # Dryrun: would be updated
+        }
+        return status_map.get(status, ("[green]✓[/green]", "green"))
+
+    async def _publish_offering_task(
+        self, offering_file: Path, console: Console, semaphore: asyncio.Semaphore, dryrun: bool = False
+    ) -> tuple[Path, dict[str, Any] | Exception]:
+        """
+        Async task to publish a single offering with concurrency control.
+
+        Returns tuple of (offering_file, result_or_exception).
+        """
+        async with semaphore:  # Limit concurrent requests
+            try:
+                # Load offering data to get the name
+                data, _ = load_data_file(offering_file)
+                offering_name = data.get("name", offering_file.stem)
+
+                # Publish the offering
+                result = await self.post_service_offering_async(offering_file, dryrun=dryrun)
+
+                # Print complete statement after publication
+                if result.get("skipped"):
+                    reason = result.get("reason", "unknown")
+                    console.print(f"  [yellow]⊘[/yellow] Skipped offering: [cyan]{offering_name}[/cyan] - {reason}")
+                else:
+                    provider_name = result.get("provider_name")
+                    status = result.get("status", "created")
+                    symbol, color = self._get_status_display(status)
+                    console.print(
+                        f"  {symbol} [{color}]{status.capitalize()}[/{color}] offering: [cyan]{offering_name}[/cyan] "
+                        f"(provider: {provider_name})"
+                    )
+
+                return (offering_file, result)
+            except Exception as e:
+                data, _ = load_data_file(offering_file)
+                offering_name = data.get("name", offering_file.stem)
+                console.print(f"  [red]✗[/red] Failed to publish offering: [cyan]{offering_name}[/cyan] - {str(e)}")
+                return (offering_file, e)
+
+    async def publish_all_offerings(self, data_dir: Path, dryrun: bool = False) -> dict[str, Any]:
+        """
+        Publish all service offerings found in a directory tree concurrently.
+
+        Validates data consistency before publishing.
+        Returns a summary of successes and failures.
+
+        Args:
+            data_dir: Directory to search for offering files
+            dryrun: If True, runs in dry run mode (no actual changes)
+        """
+        # Validate all service directories first
+        schema_dir = Path(unitysvc_services.__file__).parent / "schema"
+        validator = DataValidator(data_dir, schema_dir)
+        validation_errors = validator.validate_all_service_directories(data_dir)
+        if validation_errors:
+            return {
+                "total": 0,
+                "success": 0,
+                "failed": 0,
+                "errors": [{"file": "validation", "error": error} for error in validation_errors],
+            }
+
+        offering_files = self.find_offering_files(data_dir)
+        results: dict[str, Any] = {
+            "total": len(offering_files),
+            "success": 0,
+            "failed": 0,
+            "created": 0,
+            "updated": 0,
+            "unchanged": 0,
+            "errors": [],
+        }
+
+        if not offering_files:
+            return results
+
+        console = Console()
+
+        # Run all offering publications concurrently with rate limiting
+        # Create semaphore to limit concurrent requests
+        semaphore = asyncio.Semaphore(self.max_concurrent_requests)
+        tasks = [
+            self._publish_offering_task(offering_file, console, semaphore, dryrun=dryrun)
+            for offering_file in offering_files
+        ]
+        task_results = await asyncio.gather(*tasks)
+
+        # Process results
+        for offering_file, result in task_results:
+            if isinstance(result, Exception):
+                results["failed"] += 1
+                results["errors"].append({"file": str(offering_file), "error": str(result)})
+            else:
+                results["success"] += 1
+                # Track status counts (handle both normal and dryrun statuses)
+                status = result.get("status", "created")
+                if status in ("created", "create"):  # "create" is dryrun mode
+                    results["created"] += 1
+                elif status in ("updated", "update"):  # "update" is dryrun mode
+                    results["updated"] += 1
+                elif status == "unchanged":
+                    results["unchanged"] += 1
+
+        return results
+
+    async def _publish_listing_task(
+        self, listing_file: Path, console: Console, semaphore: asyncio.Semaphore, dryrun: bool = False
+    ) -> tuple[Path, dict[str, Any] | Exception]:
+        """
+        Async task to publish a single listing with concurrency control.
+
+        Returns tuple of (listing_file, result_or_exception).
+        """
+        async with semaphore:  # Limit concurrent requests
+            try:
+                # Load listing data to get the name
+                data, _ = load_data_file(listing_file)
+                listing_name = data.get("name", listing_file.stem)
+
+                # Publish the listing
+                result = await self.post_service_listing_async(listing_file, dryrun=dryrun)
+
+                # Print complete statement after publication
+                if result.get("skipped"):
+                    reason = result.get("reason", "unknown")
+                    console.print(f"  [yellow]⊘[/yellow] Skipped listing: [cyan]{listing_name}[/cyan] - {reason}")
+                else:
+                    service_name = result.get("service_name")
+                    provider_name = result.get("provider_name")
+                    status = result.get("status", "created")
+                    symbol, color = self._get_status_display(status)
+                    console.print(
+                        f"  {symbol} [{color}]{status.capitalize()}[/{color}] listing: [cyan]{listing_name}[/cyan] "
+                        f"(service: {service_name}, provider: {provider_name})"
+                    )
+
+                return (listing_file, result)
+            except Exception as e:
+                data, _ = load_data_file(listing_file)
+                listing_name = data.get("name", listing_file.stem)
+                console.print(f"  [red]✗[/red] Failed to publish listing: [cyan]{listing_file}[/cyan] - {str(e)}")
+                return (listing_file, e)
+
+    async def publish_all_listings(self, data_dir: Path, dryrun: bool = False) -> dict[str, Any]:
+        """
+        Publish all service listings found in a directory tree concurrently.
+
+        Validates data consistency before publishing.
+        Returns a summary of successes and failures.
+        """
+        # Validate all service directories first
+        schema_dir = Path(unitysvc_services.__file__).parent / "schema"
+        validator = DataValidator(data_dir, schema_dir)
+        validation_errors = validator.validate_all_service_directories(data_dir)
+        if validation_errors:
+            return {
+                "total": 0,
+                "success": 0,
+                "failed": 0,
+                "errors": [{"file": "validation", "error": error} for error in validation_errors],
+            }
+
+        listing_files = self.find_listing_files(data_dir)
+        results: dict[str, Any] = {
+            "total": len(listing_files),
+            "success": 0,
+            "failed": 0,
+            "created": 0,
+            "updated": 0,
+            "unchanged": 0,
+            "errors": [],
+        }
+
+        if not listing_files:
+            return results
+
+        console = Console()
+
+        # Run all listing publications concurrently with rate limiting
+        # Create semaphore to limit concurrent requests
+        semaphore = asyncio.Semaphore(self.max_concurrent_requests)
+        tasks = [
+            self._publish_listing_task(listing_file, console, semaphore, dryrun=dryrun)
+            for listing_file in listing_files
+        ]
+        task_results = await asyncio.gather(*tasks)
+
+        # Process results
+        for listing_file, result in task_results:
+            if isinstance(result, Exception):
+                results["failed"] += 1
+                results["errors"].append({"file": str(listing_file), "error": str(result)})
+            else:
+                results["success"] += 1
+                # Track status counts (handle both normal and dryrun statuses)
+                status = result.get("status", "created")
+                if status in ("created", "create"):  # "create" is dryrun mode
+                    results["created"] += 1
+                elif status in ("updated", "update"):  # "update" is dryrun mode
+                    results["updated"] += 1
+                elif status == "unchanged":
+                    results["unchanged"] += 1
+
+        return results
+
+    async def _publish_provider_task(
+        self, provider_file: Path, console: Console, semaphore: asyncio.Semaphore, dryrun: bool = False
+    ) -> tuple[Path, dict[str, Any] | Exception]:
+        """
+        Async task to publish a single provider with concurrency control.
+
+        Returns tuple of (provider_file, result_or_exception).
+        """
+        async with semaphore:  # Limit concurrent requests
+            try:
+                # Load provider data to get the name
+                data, _ = load_data_file(provider_file)
+                provider_name = data.get("name", provider_file.stem)
+
+                # Publish the provider
+                result = await self.post_provider_async(provider_file, dryrun=dryrun)
+
+                # Print complete statement after publication
+                if result.get("skipped"):
+                    reason = result.get("reason", "unknown")
+                    console.print(f"  [yellow]⊘[/yellow] Skipped provider: [cyan]{provider_name}[/cyan] - {reason}")
+                else:
+                    status = result.get("status", "created")
+                    symbol, color = self._get_status_display(status)
+                    console.print(
+                        f"  {symbol} [{color}]{status.capitalize()}[/{color}] provider: [cyan]{provider_name}[/cyan]"
+                    )
+
+                return (provider_file, result)
+            except Exception as e:
+                data, _ = load_data_file(provider_file)
+                provider_name = data.get("name", provider_file.stem)
+                console.print(f"  [red]✗[/red] Failed to publish provider: [cyan]{provider_name}[/cyan] - {str(e)}")
+                return (provider_file, e)
+
+    async def publish_all_providers(self, data_dir: Path, dryrun: bool = False) -> dict[str, Any]:
+        """
+        Publish all providers found in a directory tree concurrently.
+
+        Returns a summary of successes and failures.
+        """
+        provider_files = self.find_provider_files(data_dir)
+        results: dict[str, Any] = {
+            "total": len(provider_files),
+            "success": 0,
+            "failed": 0,
+            "created": 0,
+            "updated": 0,
+            "unchanged": 0,
+            "errors": [],
+        }
+
+        if not provider_files:
+            return results
+
+        console = Console()
+
+        # Run all provider publications concurrently with rate limiting
+        # Create semaphore to limit concurrent requests
+        semaphore = asyncio.Semaphore(self.max_concurrent_requests)
+        tasks = [
+            self._publish_provider_task(provider_file, console, semaphore, dryrun=dryrun)
+            for provider_file in provider_files
+        ]
+        task_results = await asyncio.gather(*tasks)
+
+        # Process results
+        for provider_file, result in task_results:
+            if isinstance(result, Exception):
+                results["failed"] += 1
+                results["errors"].append({"file": str(provider_file), "error": str(result)})
+            else:
+                results["success"] += 1
+                # Track status counts (handle both normal and dryrun statuses)
+                status = result.get("status", "created")
+                if status in ("created", "create"):  # "create" is dryrun mode
+                    results["created"] += 1
+                elif status in ("updated", "update"):  # "update" is dryrun mode
+                    results["updated"] += 1
+                elif status == "unchanged":
+                    results["unchanged"] += 1
+
+        return results
+
+    async def _publish_seller_task(
+        self, seller_file: Path, console: Console, semaphore: asyncio.Semaphore, dryrun: bool = False
+    ) -> tuple[Path, dict[str, Any] | Exception]:
+        """
+        Async task to publish a single seller with concurrency control.
+
+        Returns tuple of (seller_file, result_or_exception).
+        """
+        async with semaphore:  # Limit concurrent requests
+            try:
+                # Load seller data to get the name
+                data, _ = load_data_file(seller_file)
+                seller_name = data.get("name", seller_file.stem)
+
+                # Publish the seller
+                result = await self.post_seller_async(seller_file, dryrun=dryrun)
+
+                # Print complete statement after publication
+                if result.get("skipped"):
+                    reason = result.get("reason", "unknown")
+                    console.print(f"  [yellow]⊘[/yellow] Skipped seller: [cyan]{seller_name}[/cyan] - {reason}")
+                else:
+                    status = result.get("status", "created")
+                    symbol, color = self._get_status_display(status)
+                    console.print(
+                        f"  {symbol} [{color}]{status.capitalize()}[/{color}] seller: [cyan]{seller_name}[/cyan]"
+                    )
+
+                return (seller_file, result)
+            except Exception as e:
+                data, _ = load_data_file(seller_file)
+                seller_name = data.get("name", seller_file.stem)
+                console.print(f"  [red]✗[/red] Failed to publish seller: [cyan]{seller_name}[/cyan] - {str(e)}")
+                return (seller_file, e)
+
+    async def publish_all_sellers(self, data_dir: Path, dryrun: bool = False) -> dict[str, Any]:
+        """
+        Publish all sellers found in a directory tree concurrently.
+
+        Returns a summary of successes and failures.
+        """
+        seller_files = self.find_seller_files(data_dir)
+        results: dict[str, Any] = {
+            "total": len(seller_files),
+            "success": 0,
+            "failed": 0,
+            "created": 0,
+            "updated": 0,
+            "unchanged": 0,
+            "errors": [],
+        }
+
+        if not seller_files:
+            return results
+
+        console = Console()
+
+        # Run all seller publications concurrently with rate limiting
+        # Create semaphore to limit concurrent requests
+        semaphore = asyncio.Semaphore(self.max_concurrent_requests)
+        tasks = [
+            self._publish_seller_task(seller_file, console, semaphore, dryrun=dryrun) for seller_file in seller_files
+        ]
+        task_results = await asyncio.gather(*tasks)
+
+        # Process results
+        for seller_file, result in task_results:
+            if isinstance(result, Exception):
+                results["failed"] += 1
+                results["errors"].append({"file": str(seller_file), "error": str(result)})
+            else:
+                results["success"] += 1
+                # Track status counts (handle both normal and dryrun statuses)
+                status = result.get("status", "created")
+                if status in ("created", "create"):  # "create" is dryrun mode
+                    results["created"] += 1
+                elif status in ("updated", "update"):  # "update" is dryrun mode
+                    results["updated"] += 1
+                elif status == "unchanged":
+                    results["unchanged"] += 1
+
+        return results
+
+    async def publish_all_models(self, data_dir: Path, dryrun: bool = False) -> dict[str, Any]:
+        """
+        Publish all data types in the correct order.
+
+        Publishing order:
+        1. Sellers - Must exist before listings
+        2. Providers - Must exist before offerings
+        3. Service Offerings - Must exist before listings
+        4. Service Listings - Depends on sellers, providers, and offerings
+
+        Returns a dict with results for each data type and overall summary.
+        """
+        all_results: dict[str, Any] = {
+            "sellers": {},
+            "providers": {},
+            "offerings": {},
+            "listings": {},
+            "total_success": 0,
+            "total_failed": 0,
+            "total_found": 0,
+            "total_created": 0,
+            "total_updated": 0,
+            "total_unchanged": 0,
+        }
+
+        # Publish in order: sellers -> providers -> offerings -> listings
+        publish_order = [
+            ("sellers", self.publish_all_sellers),
+            ("providers", self.publish_all_providers),
+            ("offerings", self.publish_all_offerings),
+            ("listings", self.publish_all_listings),
+        ]
+
+        for data_type, publish_method in publish_order:
+            try:
+                results = await publish_method(data_dir, dryrun=dryrun)
+                all_results[data_type] = results
+                all_results["total_success"] += results["success"]
+                all_results["total_failed"] += results["failed"]
+                all_results["total_found"] += results["total"]
+                all_results["total_created"] += results.get("created", 0)
+                all_results["total_updated"] += results.get("updated", 0)
+                all_results["total_unchanged"] += results.get("unchanged", 0)
+            except Exception as e:
+                # If a publish method fails catastrophically, record the error
+                all_results[data_type] = {
+                    "total": 0,
+                    "success": 0,
+                    "failed": 1,
+                    "errors": [{"file": "N/A", "error": str(e)}],
+                }
+                all_results["total_failed"] += 1
+
+        return all_results
+
+
+# CLI commands for publishing
+app = typer.Typer(help="Publish data to backend")
+console = Console()
+
+
+@app.callback(invoke_without_command=True)
+def publish_callback(
+    ctx: typer.Context,
+    data_path: Path | None = typer.Option(
+        None,
+        "--data-path",
+        "-d",
+        help="Path to data directory (default: current directory)",
+    ),
+    dryrun: bool = typer.Option(
+        False,
+        "--dryrun",
+        help="Run in dry run mode (no actual changes)",
+    ),
+):
+    """
+    Publish data to backend.
+
+    When called without a subcommand, publishes all data types in order:
+    sellers → providers → offerings → listings.
+
+    Use subcommands to publish specific data types:
+    - providers: Publish only providers
+    - sellers: Publish only sellers
+    - offerings: Publish only service offerings
+    - listings: Publish only service listings
+
+    Required environment variables:
+    - UNITYSVC_BASE_URL: Backend API URL
+    - UNITYSVC_API_KEY: API key for authentication
+    """
+    # If a subcommand was invoked, skip this callback logic
+    if ctx.invoked_subcommand is not None:
+        return
+
+    # No subcommand - publish all
+    # Set data path
+    if data_path is None:
+        data_path = Path.cwd()
+
+    if not data_path.is_absolute():
+        data_path = Path.cwd() / data_path
+
+    if not data_path.exists():
+        console.print(f"[red]✗[/red] Path not found: {data_path}", style="bold red")
+        raise typer.Exit(code=1)
+
+    console.print(f"[bold blue]Publishing all data from:[/bold blue] {data_path}")
+    console.print(f"[bold blue]Backend URL:[/bold blue] {os.getenv('UNITYSVC_BASE_URL', 'N/A')}\n")
+
+    async def _publish_all_async():
+        async with ServiceDataPublisher() as publisher:
+            return await publisher.publish_all_models(data_path, dryrun=dryrun)
+
+    try:
+        all_results = asyncio.run(_publish_all_async())
+
+        # Create summary table
+        console.print("\n[bold cyan]Publishing Summary[/bold cyan]")
+
+        table = Table(show_header=True, header_style="bold cyan", border_style="cyan")
+        table.add_column("Type", style="cyan", no_wrap=True)
+        table.add_column("Found", justify="right")
+        table.add_column("Success", justify="right", style="green")
+        table.add_column("Failed", justify="right", style="red")
+        table.add_column("Created", justify="right", style="green")
+        table.add_column("Updated", justify="right", style="blue")
+        table.add_column("Unchanged", justify="right", style="dim")
+
+        data_type_display_names = {
+            "sellers": "Sellers",
+            "providers": "Providers",
+            "offerings": "Offerings",
+            "listings": "Listings",
+        }
+
+        # Add rows for each data type
+        for data_type in ["sellers", "providers", "offerings", "listings"]:
+            display_name = data_type_display_names[data_type]
+            results = all_results[data_type]
+
+            table.add_row(
+                display_name,
+                str(results["total"]),
+                str(results["success"]),
+                str(results["failed"]) if results["failed"] > 0 else "",
+                str(results.get("created", 0)) if results.get("created", 0) > 0 else "",
+                str(results.get("updated", 0)) if results.get("updated", 0) > 0 else "",
+                str(results.get("unchanged", 0)) if results.get("unchanged", 0) > 0 else "",
+            )
+
+        # Add separator and total row
+        table.add_section()
+        table.add_row(
+            "[bold]Total[/bold]",
+            f"[bold]{all_results['total_found']}[/bold]",
+            f"[bold green]{all_results['total_success']}[/bold green]",
+            f"[bold red]{all_results['total_failed']}[/bold red]" if all_results["total_failed"] > 0 else "",
+            f"[bold green]{all_results['total_created']}[/bold green]" if all_results["total_created"] > 0 else "",
+            f"[bold blue]{all_results['total_updated']}[/bold blue]" if all_results["total_updated"] > 0 else "",
+            f"[bold]{all_results['total_unchanged']}[/bold]" if all_results["total_unchanged"] > 0 else "",
+        )
+
+        console.print(table)
+
+        # Display errors if any
+        has_errors = False
+        for data_type in ["sellers", "providers", "offerings", "listings"]:
+            display_name = data_type_display_names[data_type]
+            results = all_results[data_type]
+
+            if results.get("errors"):
+                if not has_errors:
+                    console.print("\n[bold red]Errors:[/bold red]")
+                    has_errors = True
+
+                console.print(f"\n  [bold red]{display_name}:[/bold red]")
+                for error in results["errors"]:
+                    # Check if this is a skipped item
+                    if isinstance(error, dict) and error.get("error", "").startswith("skipped"):
+                        continue
+                    console.print(f"    [red]✗[/red] {error.get('file', 'unknown')}")
+                    console.print(f"      {error.get('error', 'unknown error')}")
+
+        if all_results["total_failed"] > 0:
+            console.print(
+                f"\n[yellow]⚠[/yellow]  Completed with {all_results['total_failed']} failure(s)",
+                style="bold yellow",
+            )
+            raise typer.Exit(code=1)
+        else:
+            if dryrun:
+                console.print(
+                    "\n[green]✓[/green] Dry run completed successfully - no changes made!",
+                    style="bold green",
+                )
+            else:
+                console.print(
+                    "\n[green]✓[/green] All data published successfully!",
+                    style="bold green",
+                )
+
+    except typer.Exit:
+        raise
+    except Exception as e:
+        console.print(f"[red]✗[/red] Failed to publish all data: {e}", style="bold red")
+        raise typer.Exit(code=1)
+
+
+@app.command("providers")
+def publish_providers(
+    data_path: Path | None = typer.Option(
+        None,
+        "--data-path",
+        "-d",
+        help="Path to provider file or directory (default: current directory)",
+    ),
+    dryrun: bool = typer.Option(
+        False,
+        "--dryrun",
+        help="Run in dry run mode (no actual changes)",
+    ),
+):
+    """Publish provider(s) from a file or directory."""
+
+    # Set data path
+    if data_path is None:
+        data_path = Path.cwd()
+
+    if not data_path.is_absolute():
+        data_path = Path.cwd() / data_path
+
+    if not data_path.exists():
+        console.print(f"[red]✗[/red] Path not found: {data_path}", style="bold red")
+        raise typer.Exit(code=1)
+
+    # Handle single file
+    if data_path.is_file():
+        console.print(f"[blue]Publishing provider:[/blue] {data_path}")
+        console.print(f"[blue]Backend URL:[/blue] {os.getenv('UNITYSVC_BASE_URL', 'N/A')}\n")
+    else:
+        console.print(f"[blue]Scanning for providers in:[/blue] {data_path}")
+        console.print(f"[blue]Backend URL:[/blue] {os.getenv('UNITYSVC_BASE_URL', 'N/A')}\n")
+
+    async def _publish_providers_async():
+        async with ServiceDataPublisher() as publisher:
+            # Handle single file
+            if data_path.is_file():
+                return await publisher.post_provider_async(data_path, dryrun=dryrun), True
+            # Handle directory
+            else:
+                return await publisher.publish_all_providers(data_path, dryrun=dryrun), False
+
+    try:
+        result, is_single = asyncio.run(_publish_providers_async())
+
+        if is_single:
+            console.print("[green]✓[/green] Provider published successfully!")
+            console.print(f"[cyan]Response:[/cyan] {json.dumps(result, indent=2)}")
+        else:
+            # Display summary
+            console.print("\n[bold cyan]Publishing Summary[/bold cyan]")
+            table = Table(show_header=True, header_style="bold cyan", border_style="cyan")
+            table.add_column("Type", style="cyan")
+            table.add_column("Found", justify="right")
+            table.add_column("Success", justify="right")
+            table.add_column("Failed", justify="right")
+            table.add_column("Created", justify="right")
+            table.add_column("Updated", justify="right")
+            table.add_column("Unchanged", justify="right")
+
+            table.add_row(
+                "Providers",
+                str(result["total"]),
+                f"[green]{result['success']}[/green]",
+                f"[red]{result['failed']}[/red]" if result["failed"] > 0 else "",
+                f"[green]{result['created']}[/green]" if result["created"] > 0 else "",
+                f"[blue]{result['updated']}[/blue]" if result["updated"] > 0 else "",
+                f"[dim]{result['unchanged']}[/dim]" if result["unchanged"] > 0 else "",
+            )
+
+            console.print(table)
+
+            # Display errors if any
+            if result["errors"]:
+                console.print("\n[bold red]Errors:[/bold red]")
+                for error in result["errors"]:
+                    console.print(f"  [red]✗[/red] {error['file']}")
+                    console.print(f"    {error['error']}")
+
+            if result["failed"] > 0:
+                raise typer.Exit(code=1)
+            else:
+                if dryrun:
+                    console.print("\n[green]✓[/green] Dry run completed successfully - no changes made!")
+                else:
+                    console.print("\n[green]✓[/green] All providers published successfully!")
+
+    except typer.Exit:
+        raise
+    except Exception as e:
+        console.print(f"[red]✗[/red] Failed to publish providers: {e}", style="bold red")
+        raise typer.Exit(code=1)
+
+
+@app.command("sellers")
+def publish_sellers(
+    data_path: Path | None = typer.Option(
+        None,
+        "--data-path",
+        "-d",
+        help="Path to seller file or directory (default: current directory)",
+    ),
+    dryrun: bool = typer.Option(
+        False,
+        "--dryrun",
+        help="Run in dry run mode (no actual changes)",
+    ),
+):
+    """Publish seller(s) from a file or directory."""
+    # Set data path
+    if data_path is None:
+        data_path = Path.cwd()
+
+    if not data_path.is_absolute():
+        data_path = Path.cwd() / data_path
+
+    if not data_path.exists():
+        console.print(f"[red]✗[/red] Path not found: {data_path}", style="bold red")
+        raise typer.Exit(code=1)
+
+    # Handle single file
+    if data_path.is_file():
+        console.print(f"[blue]Publishing seller:[/blue] {data_path}")
+        console.print(f"[blue]Backend URL:[/blue] {os.getenv('UNITYSVC_BASE_URL', 'N/A')}\n")
+    else:
+        console.print(f"[blue]Scanning for sellers in:[/blue] {data_path}")
+        console.print(f"[blue]Backend URL:[/blue] {os.getenv('UNITYSVC_BASE_URL', 'N/A')}\n")
+
+    async def _publish_sellers_async():
+        async with ServiceDataPublisher() as publisher:
+            # Handle single file
+            if data_path.is_file():
+                return await publisher.post_seller_async(data_path, dryrun=dryrun), True
+            # Handle directory
+            else:
+                return await publisher.publish_all_sellers(data_path, dryrun=dryrun), False
+
+    try:
+        result, is_single = asyncio.run(_publish_sellers_async())
+
+        if is_single:
+            console.print("[green]✓[/green] Seller published successfully!")
+            console.print(f"[cyan]Response:[/cyan] {json.dumps(result, indent=2)}")
+        else:
+            console.print("\n[bold cyan]Publishing Summary[/bold cyan]")
+            table = Table(show_header=True, header_style="bold cyan", border_style="cyan")
+            table.add_column("Type", style="cyan")
+            table.add_column("Found", justify="right")
+            table.add_column("Success", justify="right")
+            table.add_column("Failed", justify="right")
+            table.add_column("Created", justify="right")
+            table.add_column("Updated", justify="right")
+            table.add_column("Unchanged", justify="right")
+
+            table.add_row(
+                "Sellers",
+                str(result["total"]),
+                f"[green]{result['success']}[/green]",
+                f"[red]{result['failed']}[/red]" if result["failed"] > 0 else "",
+                f"[green]{result['created']}[/green]" if result["created"] > 0 else "",
+                f"[blue]{result['updated']}[/blue]" if result["updated"] > 0 else "",
+                f"[dim]{result['unchanged']}[/dim]" if result["unchanged"] > 0 else "",
+            )
+
+            console.print(table)
+
+            if result["errors"]:
+                console.print("\n[bold red]Errors:[/bold red]")
+                for error in result["errors"]:
+                    console.print(f"  [red]✗[/red] {error['file']}")
+                    console.print(f"    {error['error']}")
+                raise typer.Exit(code=1)
+            else:
+                if dryrun:
+                    console.print("\n[green]✓[/green] Dry run completed successfully - no changes made!")
+                else:
+                    console.print("\n[green]✓[/green] All sellers published successfully!")
+
+    except typer.Exit:
+        raise
+    except Exception as e:
+        console.print(f"[red]✗[/red] Failed to publish sellers: {e}", style="bold red")
+        raise typer.Exit(code=1)
+
+
+@app.command("offerings")
+def publish_offerings(
+    data_path: Path | None = typer.Option(
+        None,
+        "--data-path",
+        "-d",
+        help="Path to service offering file or directory (default: current directory)",
+    ),
+    dryrun: bool = typer.Option(
+        False,
+        "--dryrun",
+        help="Run in dry run mode (no actual changes)",
+    ),
+):
+    """Publish service offering(s) from a file or directory."""
+    # Set data path
+    if data_path is None:
+        data_path = Path.cwd()
+
+    if not data_path.is_absolute():
+        data_path = Path.cwd() / data_path
+
+    if not data_path.exists():
+        console.print(f"[red]✗[/red] Path not found: {data_path}", style="bold red")
+        raise typer.Exit(code=1)
+
+    # Handle single file
+    if data_path.is_file():
+        console.print(f"[blue]Publishing service offering:[/blue] {data_path}")
+        console.print(f"[blue]Backend URL:[/blue] {os.getenv('UNITYSVC_BASE_URL', 'N/A')}\n")
+    else:
+        console.print(f"[blue]Scanning for service offerings in:[/blue] {data_path}")
+        console.print(f"[blue]Backend URL:[/blue] {os.getenv('UNITYSVC_BASE_URL', 'N/A')}\n")
+
+    async def _publish_offerings_async():
+        async with ServiceDataPublisher() as publisher:
+            # Handle single file
+            if data_path.is_file():
+                return await publisher.post_service_offering_async(data_path, dryrun=dryrun), True
+            # Handle directory
+            else:
+                return await publisher.publish_all_offerings(data_path, dryrun=dryrun), False
+
+    try:
+        result, is_single = asyncio.run(_publish_offerings_async())
+
+        if is_single:
+            console.print("[green]✓[/green] Service offering published successfully!")
+            console.print(f"[cyan]Response:[/cyan] {json.dumps(result, indent=2)}")
+        else:
+            console.print("\n[bold cyan]Publishing Summary[/bold cyan]")
+            table = Table(show_header=True, header_style="bold cyan", border_style="cyan")
+            table.add_column("Type", style="cyan")
+            table.add_column("Found", justify="right")
+            table.add_column("Success", justify="right")
+            table.add_column("Failed", justify="right")
+            table.add_column("Created", justify="right")
+            table.add_column("Updated", justify="right")
+            table.add_column("Unchanged", justify="right")
+
+            table.add_row(
+                "Offerings",
+                str(result["total"]),
+                f"[green]{result['success']}[/green]",
+                f"[red]{result['failed']}[/red]" if result["failed"] > 0 else "",
+                f"[green]{result['created']}[/green]" if result["created"] > 0 else "",
+                f"[blue]{result['updated']}[/blue]" if result["updated"] > 0 else "",
+                f"[dim]{result['unchanged']}[/dim]" if result["unchanged"] > 0 else "",
+            )
+
+            console.print(table)
+
+            if result["errors"]:
+                console.print("\n[bold red]Errors:[/bold red]")
+                for error in result["errors"]:
+                    console.print(f"  [red]✗[/red] {error['file']}")
+                    console.print(f"    {error['error']}")
+                raise typer.Exit(code=1)
+            else:
+                if dryrun:
+                    console.print("\n[green]✓[/green] Dry run completed successfully - no changes made!")
+                else:
+                    console.print("\n[green]✓[/green] All service offerings published successfully!")
+
+    except typer.Exit:
+        raise
+    except Exception as e:
+        console.print(f"[red]✗[/red] Failed to publish service offerings: {e}", style="bold red")
+        raise typer.Exit(code=1)
+
+
+@app.command("listings")
+def publish_listings(
+    data_path: Path | None = typer.Option(
+        None,
+        "--data-path",
+        "-d",
+        help="Path to service listing file or directory (default: current directory)",
+    ),
+    dryrun: bool = typer.Option(
+        False,
+        "--dryrun",
+        help="Run in dry run mode (no actual changes)",
+    ),
+):
+    """Publish service listing(s) from a file or directory."""
+
+    # Set data path
+    if data_path is None:
+        data_path = Path.cwd()
+
+    if not data_path.is_absolute():
+        data_path = Path.cwd() / data_path
+
+    if not data_path.exists():
+        console.print(f"[red]✗[/red] Path not found: {data_path}", style="bold red")
+        raise typer.Exit(code=1)
+
+    # Handle single file
+    if data_path.is_file():
+        console.print(f"[blue]Publishing service listing:[/blue] {data_path}")
+        console.print(f"[blue]Backend URL:[/blue] {os.getenv('UNITYSVC_BASE_URL', 'N/A')}\n")
+    else:
+        console.print(f"[blue]Scanning for service listings in:[/blue] {data_path}")
+        console.print(f"[blue]Backend URL:[/blue] {os.getenv('UNITYSVC_BASE_URL', 'N/A')}\n")
+
+    async def _publish_listings_async():
+        async with ServiceDataPublisher() as publisher:
+            # Handle single file
+            if data_path.is_file():
+                return await publisher.post_service_listing_async(data_path, dryrun=dryrun), True
+            # Handle directory
+            else:
+                return await publisher.publish_all_listings(data_path, dryrun=dryrun), False
+
+    try:
+        result, is_single = asyncio.run(_publish_listings_async())
+
+        if is_single:
+            console.print("[green]✓[/green] Service listing published successfully!")
+            console.print(f"[cyan]Response:[/cyan] {json.dumps(result, indent=2)}")
+        else:
+            console.print("\n[bold cyan]Publishing Summary[/bold cyan]")
+            table = Table(show_header=True, header_style="bold cyan", border_style="cyan")
+            table.add_column("Type", style="cyan")
+            table.add_column("Found", justify="right")
+            table.add_column("Success", justify="right")
+            table.add_column("Failed", justify="right")
+            table.add_column("Created", justify="right")
+            table.add_column("Updated", justify="right")
+            table.add_column("Unchanged", justify="right")
+
+            table.add_row(
+                "Listings",
+                str(result["total"]),
+                f"[green]{result['success']}[/green]",
+                f"[red]{result['failed']}[/red]" if result["failed"] > 0 else "",
+                f"[green]{result['created']}[/green]" if result["created"] > 0 else "",
+                f"[blue]{result['updated']}[/blue]" if result["updated"] > 0 else "",
+                f"[dim]{result['unchanged']}[/dim]" if result["unchanged"] > 0 else "",
+            )
+
+            console.print(table)
+
+            if result["errors"]:
+                console.print("\n[bold red]Errors:[/bold red]")
+                for error in result["errors"]:
+                    console.print(f"  [red]✗[/red] {error['file']}")
+                    console.print(f"    {error['error']}")
+                raise typer.Exit(code=1)
+            else:
+                if dryrun:
+                    console.print("\n[green]✓[/green] Dry run completed successfully - no changes made!")
+                else:
+                    console.print("\n[green]✓[/green] All service listings published successfully!")
+
+    except typer.Exit:
+        raise
+    except Exception as e:
+        console.print(f"[red]✗[/red] Failed to publish service listings: {e}", style="bold red")
+        raise typer.Exit(code=1)