[DocumentIntelligence] Generate from the latest typespec with service API 2024-02-29-preview (#34564)

Author: Yalin Li

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

@@ -4,6 +4,14 @@
 
 ### Features Added
 
+- Added model `AnalyzeResultOperation` and `Warning`.
+- Added property `base_classifier_id` to model `BuildDocumentClassifierRequest`.
+- Added property `base_classifier_id` and `warnings` to model `DocumentClassifierDetails`.
+- Added property `warnings` to model `DocumentModelDetails`.
+- Added property `value_selection_group` to model `DocumentField`.
+- Added value `selectionGroup` to enum `DocumentFieldType`.
+- Added value `completed` to enum `OperationStatus`.
+
 ### Breaking Changes
 
 - Changed property name from `items` to `items_property` in model `DocumentFieldSchema` and `DocumentList`.
@@ -13,6 +21,9 @@
 ### Other Changes
 
 - Changed the default polling interval from 30s to 5s.
+- Changed the default service API version to `2024-02-29-preview`.
+- Bumped minimum dependency on `azure-core` to `>=1.30.0`.
+- Bumped minimum dependency on `typing-extensions` to `>=4.6.0`.
 - Python 3.7 is no longer supported. Please use Python version 3.8 or later.
 
 ## 1.0.0b1 (2023-11-17)

sdk/documentintelligence/azure-ai-documentintelligence/README.md

@@ -575,6 +575,10 @@ print(f"\nModel ID: {my_model.model_id}")
 print(f"Description: {my_model.description}")
 print(f"Model created on: {my_model.created_date_time}")
 print(f"Model expires on: {my_model.expiration_date_time}")
+if my_model.warnings:
+    print("Warnings encountered while building the model:")
+    for warning in my_model.warnings:
+        print(f"warning code: {warning.code}, message: {warning.message}, target of the error: {warning.target}")
 ```
 
 <!-- END SNIPPET -->

sdk/documentintelligence/azure-ai-documentintelligence/assets.json

@@ -2,5 +2,5 @@
   "AssetsRepo": "Azure/azure-sdk-assets",
   "AssetsRepoPrefixPath": "python",
   "TagPrefix": "python/documentintelligence/azure-ai-documentintelligence",
-  "Tag": "python/documentintelligence/azure-ai-documentintelligence_3a7d15d3b0"
+  "Tag": "python/documentintelligence/azure-ai-documentintelligence_a54f07ab98"
 }

sdk/documentintelligence/azure-ai-documentintelligence/azure/ai/documentintelligence/_client.py

@@ -41,7 +41,7 @@ class DocumentIntelligenceClient(
     :type credential: ~azure.core.credentials.AzureKeyCredential or
      ~azure.core.credentials.TokenCredential
     :keyword api_version: The API version to use for this operation. Default value is
-     "2023-10-31-preview". Note that overriding this default value may result in unsupported
+     "2024-02-29-preview". Note that overriding this default value may result in unsupported
      behavior.
     :paramtype api_version: str
     :keyword int polling_interval: Default waiting time between two polls for LRO operations if no
@@ -123,7 +123,7 @@ class DocumentIntelligenceAdministrationClient(
     :type credential: ~azure.core.credentials.AzureKeyCredential or
      ~azure.core.credentials.TokenCredential
     :keyword api_version: The API version to use for this operation. Default value is
-     "2023-10-31-preview". Note that overriding this default value may result in unsupported
+     "2024-02-29-preview". Note that overriding this default value may result in unsupported
      behavior.
     :paramtype api_version: str
     :keyword int polling_interval: Default waiting time between two polls for LRO operations if no

sdk/documentintelligence/azure-ai-documentintelligence/azure/ai/documentintelligence/_configuration.py

@@ -31,13 +31,13 @@ class DocumentIntelligenceClientConfiguration:  # pylint: disable=too-many-insta
     :type credential: ~azure.core.credentials.AzureKeyCredential or
      ~azure.core.credentials.TokenCredential
     :keyword api_version: The API version to use for this operation. Default value is
-     "2023-10-31-preview". Note that overriding this default value may result in unsupported
+     "2024-02-29-preview". Note that overriding this default value may result in unsupported
      behavior.
     :paramtype api_version: str
     """
 
     def __init__(self, endpoint: str, credential: Union[AzureKeyCredential, "TokenCredential"], **kwargs: Any) -> None:
-        api_version: str = kwargs.pop("api_version", "2023-10-31-preview")
+        api_version: str = kwargs.pop("api_version", "2024-02-29-preview")
 
         if endpoint is None:
             raise ValueError("Parameter 'endpoint' must not be None.")
@@ -86,13 +86,13 @@ class DocumentIntelligenceAdministrationClientConfiguration:  # pylint: disable=
     :type credential: ~azure.core.credentials.AzureKeyCredential or
      ~azure.core.credentials.TokenCredential
     :keyword api_version: The API version to use for this operation. Default value is
-     "2023-10-31-preview". Note that overriding this default value may result in unsupported
+     "2024-02-29-preview". Note that overriding this default value may result in unsupported
      behavior.
     :paramtype api_version: str
     """
 
     def __init__(self, endpoint: str, credential: Union[AzureKeyCredential, "TokenCredential"], **kwargs: Any) -> None:
-        api_version: str = kwargs.pop("api_version", "2023-10-31-preview")
+        api_version: str = kwargs.pop("api_version", "2024-02-29-preview")
 
         if endpoint is None:
             raise ValueError("Parameter 'endpoint' must not be None.")

sdk/documentintelligence/azure-ai-documentintelligence/azure/ai/documentintelligence/_model_base.py

@@ -5,7 +5,6 @@
 # license information.
 # --------------------------------------------------------------------------
 # pylint: disable=protected-access, arguments-differ, signature-differs, broad-except
-# pyright: reportGeneralTypeIssues=false
 
 import calendar
 import decimal
@@ -16,9 +15,11 @@
 import re
 import copy
 import typing
-import email
+import enum
+import email.utils
 from datetime import datetime, date, time, timedelta, timezone
 from json import JSONEncoder
+from typing_extensions import Self
 import isodate
 from azure.core.exceptions import DeserializationError
 from azure.core import CaseInsensitiveEnumMeta
@@ -35,6 +36,7 @@
 __all__ = ["SdkJSONEncoder", "Model", "rest_field", "rest_discriminator"]
 
 TZ_UTC = timezone.utc
+_T = typing.TypeVar("_T")
 
 
 def _timedelta_as_isostr(td: timedelta) -> str:
@@ -242,7 +244,7 @@ def _deserialize_date(attr: typing.Union[str, date]) -> date:
     # This must NOT use defaultmonth/defaultday. Using None ensure this raises an exception.
     if isinstance(attr, date):
         return attr
-    return isodate.parse_date(attr, defaultmonth=None, defaultday=None)
+    return isodate.parse_date(attr, defaultmonth=None, defaultday=None)  # type: ignore
 
 
 def _deserialize_time(attr: typing.Union[str, time]) -> time:
@@ -375,8 +377,12 @@ def get(self, key: str, default: typing.Any = None) -> typing.Any:
         except KeyError:
             return default
 
-    @typing.overload  # type: ignore
-    def pop(self, key: str) -> typing.Any:  # pylint: disable=no-member
+    @typing.overload
+    def pop(self, key: str) -> typing.Any:
+        ...
+
+    @typing.overload
+    def pop(self, key: str, default: _T) -> _T:
         ...
 
     @typing.overload
@@ -397,8 +403,8 @@ def clear(self) -> None:
     def update(self, *args: typing.Any, **kwargs: typing.Any) -> None:
         self._data.update(*args, **kwargs)
 
-    @typing.overload  # type: ignore
-    def setdefault(self, key: str) -> typing.Any:
+    @typing.overload
+    def setdefault(self, key: str, default: None = None) -> None:
         ...
 
     @typing.overload
@@ -438,6 +444,8 @@ def _serialize(o, format: typing.Optional[str] = None):  # pylint: disable=too-m
         return _serialize_bytes(o, format)
     if isinstance(o, decimal.Decimal):
         return float(o)
+    if isinstance(o, enum.Enum):
+        return o.value
     try:
         # First try datetime.datetime
         return _serialize_datetime(o, format)
@@ -462,7 +470,13 @@ def _get_rest_field(
 
 
 def _create_value(rf: typing.Optional["_RestField"], value: typing.Any) -> typing.Any:
-    return _deserialize(rf._type, value) if (rf and rf._is_model) else _serialize(value, rf._format if rf else None)
+    if not rf:
+        return _serialize(value, None)
+    if rf._is_multipart_file_input:
+        return value
+    if rf._is_model:
+        return _deserialize(rf._type, value)
+    return _serialize(value, rf._format)
 
 
 class Model(_MyMutableMapping):
@@ -498,7 +512,7 @@ def __init__(self, *args: typing.Any, **kwargs: typing.Any) -> None:
     def copy(self) -> "Model":
         return Model(self.__dict__)
 
-    def __new__(cls, *args: typing.Any, **kwargs: typing.Any) -> "Model":  # pylint: disable=unused-argument
+    def __new__(cls, *args: typing.Any, **kwargs: typing.Any) -> Self:  # pylint: disable=unused-argument
         # we know the last three classes in mro are going to be 'Model', 'dict', and 'object'
         mros = cls.__mro__[:-3][::-1]  # ignore model, dict, and object parents, and reverse the mro order
         attr_to_rest_field: typing.Dict[str, _RestField] = {  # map attribute name to rest_field property
@@ -540,7 +554,7 @@ def _deserialize(cls, data, exist_discriminators):
             return cls(data)
         discriminator = cls._get_discriminator(exist_discriminators)
         exist_discriminators.append(discriminator)
-        mapped_cls = cls.__mapping__.get(data.get(discriminator), cls)  # pylint: disable=no-member
+        mapped_cls = cls.__mapping__.get(data.get(discriminator), cls)  # pyright: ignore # pylint: disable=no-member
         if mapped_cls == cls:
             return cls(data)
         return mapped_cls._deserialize(data, exist_discriminators)  # pylint: disable=protected-access
@@ -557,17 +571,24 @@ def as_dict(self, *, exclude_readonly: bool = False) -> typing.Dict[str, typing.
         if exclude_readonly:
             readonly_props = [p._rest_name for p in self._attr_to_rest_field.values() if _is_readonly(p)]
         for k, v in self.items():
-            if exclude_readonly and k in readonly_props:  # pyright: ignore[reportUnboundVariable]
+            if exclude_readonly and k in readonly_props:  # pyright: ignore
                 continue
-            result[k] = Model._as_dict_value(v, exclude_readonly=exclude_readonly)
+            is_multipart_file_input = False
+            try:
+                is_multipart_file_input = next(
+                    rf for rf in self._attr_to_rest_field.values() if rf._rest_name == k
+                )._is_multipart_file_input
+            except StopIteration:
+                pass
+            result[k] = v if is_multipart_file_input else Model._as_dict_value(v, exclude_readonly=exclude_readonly)
         return result
 
     @staticmethod
     def _as_dict_value(v: typing.Any, exclude_readonly: bool = False) -> typing.Any:
         if v is None or isinstance(v, _Null):
             return None
         if isinstance(v, (list, tuple, set)):
-            return [Model._as_dict_value(x, exclude_readonly=exclude_readonly) for x in v]
+            return type(v)(Model._as_dict_value(x, exclude_readonly=exclude_readonly) for x in v)
         if isinstance(v, dict):
             return {dk: Model._as_dict_value(dv, exclude_readonly=exclude_readonly) for dk, dv in v.items()}
         return v.as_dict(exclude_readonly=exclude_readonly) if hasattr(v, "as_dict") else v
@@ -605,29 +626,22 @@ def _deserialize_model(model_deserializer: typing.Optional[typing.Callable], obj
                     return obj
                 return _deserialize(model_deserializer, obj)
 
-            return functools.partial(_deserialize_model, annotation)
+            return functools.partial(_deserialize_model, annotation)  # pyright: ignore
     except Exception:
         pass
 
     # is it a literal?
     try:
-        if sys.version_info >= (3, 8):
-            from typing import (
-                Literal,
-            )  # pylint: disable=no-name-in-module, ungrouped-imports
-        else:
-            from typing_extensions import Literal  # type: ignore  # pylint: disable=ungrouped-imports
-
-        if annotation.__origin__ == Literal:
+        if annotation.__origin__ is typing.Literal:  # pyright: ignore
             return None
     except AttributeError:
         pass
 
     # is it optional?
     try:
-        if any(a for a in annotation.__args__ if a == type(None)):
+        if any(a for a in annotation.__args__ if a == type(None)):  # pyright: ignore
             if_obj_deserializer = _get_deserialize_callable_from_annotation(
-                next(a for a in annotation.__args__ if a != type(None)), module, rf
+                next(a for a in annotation.__args__ if a != type(None)), module, rf  # pyright: ignore
             )
 
             def _deserialize_with_optional(if_obj_deserializer: typing.Optional[typing.Callable], obj):
@@ -640,7 +654,13 @@ def _deserialize_with_optional(if_obj_deserializer: typing.Optional[typing.Calla
         pass
 
     if getattr(annotation, "__origin__", None) is typing.Union:
-        deserializers = [_get_deserialize_callable_from_annotation(arg, module, rf) for arg in annotation.__args__]
+        # initial ordering is we make `string` the last deserialization option, because it is often them most generic
+        deserializers = [
+            _get_deserialize_callable_from_annotation(arg, module, rf)
+            for arg in sorted(
+                annotation.__args__, key=lambda x: hasattr(x, "__name__") and x.__name__ == "str"  # pyright: ignore
+            )
+        ]
 
         def _deserialize_with_union(deserializers, obj):
             for deserializer in deserializers:
@@ -653,8 +673,10 @@ def _deserialize_with_union(deserializers, obj):
         return functools.partial(_deserialize_with_union, deserializers)
 
     try:
-        if annotation._name == "Dict":
-            value_deserializer = _get_deserialize_callable_from_annotation(annotation.__args__[1], module, rf)
+        if annotation._name == "Dict":  # pyright: ignore
+            value_deserializer = _get_deserialize_callable_from_annotation(
+                annotation.__args__[1], module, rf  # pyright: ignore
+            )
 
             def _deserialize_dict(
                 value_deserializer: typing.Optional[typing.Callable],
@@ -671,8 +693,8 @@ def _deserialize_dict(
     except (AttributeError, IndexError):
         pass
     try:
-        if annotation._name in ["List", "Set", "Tuple", "Sequence"]:
-            if len(annotation.__args__) > 1:
+        if annotation._name in ["List", "Set", "Tuple", "Sequence"]:  # pyright: ignore
+            if len(annotation.__args__) > 1:  # pyright: ignore
 
                 def _deserialize_multiple_sequence(
                     entry_deserializers: typing.List[typing.Optional[typing.Callable]],
@@ -686,10 +708,13 @@ def _deserialize_multiple_sequence(
                     )
 
                 entry_deserializers = [
-                    _get_deserialize_callable_from_annotation(dt, module, rf) for dt in annotation.__args__
+                    _get_deserialize_callable_from_annotation(dt, module, rf)
+                    for dt in annotation.__args__  # pyright: ignore
                 ]
                 return functools.partial(_deserialize_multiple_sequence, entry_deserializers)
-            deserializer = _get_deserialize_callable_from_annotation(annotation.__args__[0], module, rf)
+            deserializer = _get_deserialize_callable_from_annotation(
+                annotation.__args__[0], module, rf  # pyright: ignore
+            )
 
             def _deserialize_sequence(
                 deserializer: typing.Optional[typing.Callable],
@@ -769,6 +794,7 @@ def __init__(
         visibility: typing.Optional[typing.List[str]] = None,
         default: typing.Any = _UNSET,
         format: typing.Optional[str] = None,
+        is_multipart_file_input: bool = False,
     ):
         self._type = type
         self._rest_name_input = name
@@ -778,6 +804,11 @@ def __init__(
         self._is_model = False
         self._default = default
         self._format = format
+        self._is_multipart_file_input = is_multipart_file_input
+
+    @property
+    def _class_type(self) -> typing.Any:
+        return getattr(self._type, "args", [None])[0]
 
     @property
     def _rest_name(self) -> str:
@@ -823,8 +854,16 @@ def rest_field(
     visibility: typing.Optional[typing.List[str]] = None,
     default: typing.Any = _UNSET,
     format: typing.Optional[str] = None,
+    is_multipart_file_input: bool = False,
 ) -> typing.Any:
-    return _RestField(name=name, type=type, visibility=visibility, default=default, format=format)
+    return _RestField(
+        name=name,
+        type=type,
+        visibility=visibility,
+        default=default,
+        format=format,
+        is_multipart_file_input=is_multipart_file_input,
+    )
 
 
 def rest_discriminator(