Add v0.1.0b5 for azure-ai-ml (Azure#25042)

* 2.6.0 release for azure-ai-ml

* Updated changelog for azure-ai-ml

* Applied hotfixes to pass pipelines

* Revised version number

* Typo fix

* Added missing dependency for ml

* Replaced failing test files for ml

* Updated release cut to fix pipeline bugs for ml

* Added missing test configuration files

* Fixed anonymous asset test to use proper version

* Added py.typed file

* Moved py.typed file to namespace root

* Typo fix from previous commit

* Moved py.typed file to furthest non-code depth of ml package

* Removed extra comments

* Fixing version number

* Fixes from code review

* Updated 0.1.0b5 with hotfixes

* Updated release date on changelog

sdk/ml/azure-ai-ml/CHANGELOG.md

@@ -1,15 +1,32 @@
-## Release History
+# Release History
 
-### 0.1.0b4 (unreleased)
+## 0.1.0b5 (2022-07-15)
 
-#### Features Added
+### Features Added
 
-#### Breaking Changes
+- Allow Input/Output objects to be used by CommandComponent.
+- Added MoonCake cloud support.
+- Unified inputs/outputs building and validation logic in BaseNode.
+- Allow Git repo URLs to be used as code for jobs and components.
+- Updated AutoML YAML schema to use InputSchema.
+- Added end_time to job schedule.
+- MIR and pipeline job now support registry assets.
 
-#### Bugs Fixed
+### Breaking Changes
 
-#### Other Changes
+### Bugs Fixed
 
+- Have mldesigner use argparser to parse incoming args.
+- Bumped pyjwt version to <3.0.0.
+- Reverted "upload support for symlinks".
+- Error message improvement when a YAML UnionField fails to match.
+- Reintroduced support for symlinks when uploading.
+- Hard coded registry base URL to eastus region to support preview.
+
+## 0.1.0b4 (2022-06-16)
+
+## 0.1.0b3 (2022-05-24)
+
+### Features Added
 
-### 0.1.0b3 (2022-05-24)
 - First preview.

sdk/ml/azure-ai-ml/azure/ai/ml/_arm_deployments/arm_deployment_executor.py

@@ -3,7 +3,12 @@
 # ---------------------------------------------------------
 
 from typing import Dict, Any, Optional
-from azure.ai.ml._azure_environments import ENDPOINT_URLS, _get_cloud_details, resource_to_scopes
+from azure.ai.ml._azure_environments import (
+    _get_cloud_details,
+    _get_base_url_from_metadata,
+    _resource_to_scopes,
+    _get_azure_portal_id_from_metadata,
+)
 from azure.core.polling import LROPoller
 from azure.ai.ml._arm_deployments.arm_helper import deployment_message_mapping
 from azure.ai.ml._utils._arm_id_utils import get_arm_id_object_from_id
@@ -39,8 +44,8 @@ def __init__(
         self._resource_group_name = resource_group_name
         self._deployment_name = deployment_name
         self._cloud = _get_cloud_details()
-        management_hostname = self._cloud.get(ENDPOINT_URLS.RESOURCE_MANAGER_ENDPOINT).strip("/")
-        credential_scopes = resource_to_scopes(management_hostname)
+        management_hostname = _get_base_url_from_metadata()
+        credential_scopes = _resource_to_scopes(management_hostname)
         kwargs.pop("base_url", None)
         if credential_scopes is not None:
             kwargs["credential_scopes"] = credential_scopes
@@ -82,7 +87,7 @@ def deploy_resource(
             )
             module_logger.info(
                 ENDPOINT_DEPLOYMENT_START_MSG.format(
-                    self._cloud.get(ENDPOINT_URLS.AZURE_PORTAL_ENDPOINT),
+                    _get_azure_portal_id_from_metadata(),
                     self._subscription_id,
                     self._resource_group_name,
                     self._deployment_name,

sdk/ml/azure-ai-ml/azure/ai/ml/_artifacts/_artifact_utilities.py

@@ -8,7 +8,7 @@
 from pathlib import Path
 from datetime import datetime, timedelta
 import uuid
-from azure.ai.ml._azure_environments import ENDPOINT_URLS, _get_cloud_details
+from azure.ai.ml._azure_environments import _get_storage_endpoint_from_metadata
 
 from azure.storage.blob import generate_blob_sas, BlobSasPermissions
 from azure.storage.filedatalake import generate_file_sas, FileSasPermissions
@@ -68,8 +68,7 @@ def get_datastore_info(operations: DatastoreOperations, name: str) -> Dict[str,
     else:
         datastore = operations.get_default(include_secrets=True)
 
-    cloud_details = _get_cloud_details()
-    storage_endpoint = cloud_details.get(ENDPOINT_URLS.STORAGE_ENDPOINT)
+    storage_endpoint = _get_storage_endpoint_from_metadata()
     credentials = datastore.credentials
     datastore_info["storage_type"] = datastore.type
     datastore_info["storage_account"] = datastore.account_name

sdk/ml/azure-ai-ml/azure/ai/ml/_artifacts/_blob_storage_helper.py

@@ -6,35 +6,26 @@
 import logging
 import time
 import os
-import warnings
-from contextlib import suppress
-from typing import Optional, Dict, Any, List, TYPE_CHECKING
+from typing import Dict, List, TYPE_CHECKING
 from pathlib import PurePosixPath, Path
-from multiprocessing import cpu_count
 from colorama import Fore
-from concurrent.futures import ThreadPoolExecutor, as_completed
-from tqdm import tqdm, TqdmWarning
-from platform import system
 import sys
 
-from azure.ai.ml._utils._exception_utils import EmptyDirectoryError
 from azure.storage.blob import BlobServiceClient, ContainerClient
 from azure.core.exceptions import ResourceNotFoundError
 from azure.ai.ml._utils._asset_utils import (
     generate_asset_id,
-    traverse_directory,
+    upload_directory,
+    upload_file,
     AssetNotChangedError,
     _build_metadata_dict,
     IgnoreFile,
-    FileUploadProgressBar,
     get_directory_size,
 )
 from azure.ai.ml._artifacts._constants import (
     UPLOAD_CONFIRMATION,
     ARTIFACT_ORIGIN,
     LEGACY_ARTIFACT_DIRECTORY,
-    EMPTY_DIRECTORY_ERROR,
-    PROCESSES_PER_CORE,
     MAX_CONCURRENCY,
     FILE_SIZE_WARNING,
     BLOB_DATASTORE_IS_HDI_FOLDER_KEY,
@@ -103,11 +94,18 @@ def upload(
 
             # start upload
             if os.path.isdir(source):
-                self.upload_dir(source, asset_id, msg, show_progress, ignore_file=ignore_file)
+                upload_directory(
+                    storage_client=self,
+                    source=source,
+                    dest=asset_id,
+                    msg=msg,
+                    show_progress=show_progress,
+                    ignore_file=ignore_file,
+                )
             else:
                 self.indicator_file = dest
                 self.check_blob_exists()
-                self.upload_file(source, dest, msg, show_progress)
+                upload_file(storage_client=self, source=source, dest=dest, msg=msg, show_progress=show_progress)
             print(Fore.RESET + "\n", file=sys.stderr)
 
             # upload must be completed before we try to generate confirmation file
@@ -124,98 +122,6 @@ def upload(
 
         return artifact_info
 
-    def upload_file(
-        self,
-        source: str,
-        dest: str,
-        msg: Optional[str] = None,
-        show_progress: Optional[bool] = None,
-        in_directory: bool = False,
-        callback: Any = None,
-    ) -> None:
-        """
-        Upload a single file to a path inside the container
-        """
-        validate_content = os.stat(source).st_size > 0  # don't do checksum for empty files
-
-        with open(source, "rb") as data:
-            if show_progress and not in_directory:
-                file_size, _ = get_directory_size(source)
-                file_size_in_mb = file_size / 10**6
-                if file_size_in_mb < 1:
-                    msg += Fore.GREEN + " (< 1 MB)"
-                else:
-                    msg += Fore.GREEN + f" ({round(file_size_in_mb, 2)} MBs)"
-                cntx_manager = FileUploadProgressBar(msg=msg)
-            else:
-                cntx_manager = suppress()
-
-            with cntx_manager as c:
-                callback = c.update_to if (show_progress and not in_directory) else None
-                self.container_client.upload_blob(
-                    name=dest,
-                    data=data,
-                    validate_content=validate_content,
-                    overwrite=self.overwrite,
-                    raw_response_hook=callback,
-                    max_concurrency=MAX_CONCURRENCY,
-                )
-
-        self.uploaded_file_count += 1
-
-    def upload_dir(self, source: str, dest: str, msg: str, show_progress: bool, ignore_file: IgnoreFile) -> None:
-        """
-        Upload a directory to a path inside the container
-
-        Azure Blob doesn't allow metadata setting at the directory level, so the first
-        file in the directory is designated as the file where the confirmation metadata
-        will be added at the end of the upload.
-        """
-        source_path = Path(source).resolve()
-        prefix = "" if dest == "" else dest + "/"
-        prefix += os.path.basename(source_path) + "/"
-
-        # get all paths in directory and each file's size
-        upload_paths = []
-        size_dict = {}
-        total_size = 0
-        for root, _, files in os.walk(source_path):
-            upload_paths += list(traverse_directory(root, files, source_path, prefix, ignore_file=ignore_file))
-
-        for path, _ in upload_paths:
-            path_size = os.path.getsize(path)
-            size_dict[path] = path_size
-            total_size += path_size
-
-        upload_paths = sorted(upload_paths)
-        if len(upload_paths) == 0:
-            raise EmptyDirectoryError(
-                message=EMPTY_DIRECTORY_ERROR.format(source),
-                no_personal_data_message=msg.format("[source]"),
-                target=ErrorTarget.ARTIFACT,
-                error_category=ErrorCategory.USER_ERROR,
-            )
-
-        self.indicator_file = upload_paths[0][1]
-        self.check_blob_exists()
-        self.total_file_count = len(upload_paths)
-
-        # submit paths to workers for upload
-        num_cores = int(cpu_count()) * PROCESSES_PER_CORE
-        with ThreadPoolExecutor(max_workers=num_cores) as ex:
-            futures_dict = {
-                ex.submit(self.upload_file, src, dest, in_directory=True, show_progress=show_progress): (src, dest)
-                for (src, dest) in upload_paths
-            }
-            if show_progress:
-                warnings.simplefilter("ignore", category=TqdmWarning)
-                msg += f" ({round(total_size/10**6, 2)} MBs)"
-                ascii = system() == "Windows"  # Default unicode progress bar doesn't display well on Windows
-                with tqdm(total=total_size, desc=msg, ascii=ascii) as pbar:
-                    for future in as_completed(futures_dict):
-                        file_path_name = futures_dict[future][0]
-                        pbar.update(size_dict.get(file_path_name) or 0)
-
     def check_blob_exists(self) -> None:
         """
         Throw error if blob already exists.

sdk/ml/azure-ai-ml/azure/ai/ml/_artifacts/_constants.py

@@ -30,3 +30,5 @@
     "{jsonSchemaErrorPath}{jsonSchemaMessage}\n{invalidMLTableMsg}:\n{invalidSchemaSnippet}"
 )
 BLOB_DATASTORE_IS_HDI_FOLDER_KEY = "hdi_isfolder"
+BLOB_STORAGE_CLIENT_NAME = "BlobStorageClient"
+GEN2_STORAGE_CLIENT_NAME = "Gen2StorageClient"