unstructured-ingest 0.5.15__tar.gz → 0.5.16__tar.gz

{unstructured_ingest-0.5.15/unstructured_ingest.egg-info → unstructured_ingest-0.5.16}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: unstructured-ingest
-Version: 0.5.15
+Version: 0.5.16
 Summary: A library that prepares raw documents for downstream ML tasks.
 Home-page: https://github.com/Unstructured-IO/unstructured-ingest
 Author: Unstructured Technologies
@@ -22,13 +22,13 @@ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Requires-Python: >=3.9.0,<3.14
 Description-Content-Type: text/markdown
 License-File: LICENSE.md
-Requires-Dist: python-dateutil
 Requires-Dist: tqdm
 Requires-Dist: dataclasses_json
-Requires-Dist: pydantic>=2.7
-Requires-Dist: click
 Requires-Dist: opentelemetry-sdk
+Requires-Dist: python-dateutil
 Requires-Dist: pandas
+Requires-Dist: pydantic>=2.7
+Requires-Dist: click
 Provides-Extra: remote
 Requires-Dist: unstructured-client>=0.30.0; extra == "remote"
 Provides-Extra: csv
@@ -81,13 +81,13 @@ Requires-Dist: chromadb; extra == "chroma"
 Provides-Extra: clarifai
 Requires-Dist: clarifai; extra == "clarifai"
 Provides-Extra: confluence
-Requires-Dist: requests; extra == "confluence"
 Requires-Dist: atlassian-python-api; extra == "confluence"
+Requires-Dist: requests; extra == "confluence"
 Provides-Extra: couchbase
 Requires-Dist: couchbase; extra == "couchbase"
 Provides-Extra: delta-table
-Requires-Dist: boto3; extra == "delta-table"
 Requires-Dist: deltalake; extra == "delta-table"
+Requires-Dist: boto3; extra == "delta-table"
 Provides-Extra: discord
 Requires-Dist: discord.py; extra == "discord"
 Provides-Extra: dropbox
@@ -98,19 +98,19 @@ Requires-Dist: duckdb; extra == "duckdb"
 Provides-Extra: elasticsearch
 Requires-Dist: elasticsearch[async]; extra == "elasticsearch"
 Provides-Extra: gcs
-Requires-Dist: gcsfs; extra == "gcs"
 Requires-Dist: bs4; extra == "gcs"
+Requires-Dist: gcsfs; extra == "gcs"
 Requires-Dist: fsspec; extra == "gcs"
 Provides-Extra: github
-Requires-Dist: requests; extra == "github"
 Requires-Dist: pygithub>1.58.0; extra == "github"
+Requires-Dist: requests; extra == "github"
 Provides-Extra: gitlab
 Requires-Dist: python-gitlab; extra == "gitlab"
 Provides-Extra: google-drive
 Requires-Dist: google-api-python-client; extra == "google-drive"
 Provides-Extra: hubspot
-Requires-Dist: hubspot-api-client; extra == "hubspot"
 Requires-Dist: urllib3; extra == "hubspot"
+Requires-Dist: hubspot-api-client; extra == "hubspot"
 Provides-Extra: jira
 Requires-Dist: atlassian-python-api; extra == "jira"
 Provides-Extra: kafka
@@ -124,23 +124,23 @@ Requires-Dist: pymilvus; extra == "milvus"
 Provides-Extra: mongodb
 Requires-Dist: pymongo; extra == "mongodb"
 Provides-Extra: neo4j
-Requires-Dist: cymple; extra == "neo4j"
-Requires-Dist: networkx; extra == "neo4j"
 Requires-Dist: neo4j-rust-ext; extra == "neo4j"
+Requires-Dist: networkx; extra == "neo4j"
+Requires-Dist: cymple; extra == "neo4j"
 Provides-Extra: notion
-Requires-Dist: htmlBuilder; extra == "notion"
-Requires-Dist: httpx; extra == "notion"
 Requires-Dist: backoff; extra == "notion"
+Requires-Dist: httpx; extra == "notion"
+Requires-Dist: htmlBuilder; extra == "notion"
 Requires-Dist: notion-client; extra == "notion"
 Provides-Extra: onedrive
 Requires-Dist: bs4; extra == "onedrive"
-Requires-Dist: msal; extra == "onedrive"
 Requires-Dist: Office365-REST-Python-Client; extra == "onedrive"
+Requires-Dist: msal; extra == "onedrive"
 Provides-Extra: opensearch
 Requires-Dist: opensearch-py; extra == "opensearch"
 Provides-Extra: outlook
-Requires-Dist: msal; extra == "outlook"
 Requires-Dist: Office365-REST-Python-Client; extra == "outlook"
+Requires-Dist: msal; extra == "outlook"
 Provides-Extra: pinecone
 Requires-Dist: pinecone-client>=3.7.1; extra == "pinecone"
 Provides-Extra: postgres
@@ -155,8 +155,8 @@ Provides-Extra: s3
 Requires-Dist: s3fs; extra == "s3"
 Requires-Dist: fsspec; extra == "s3"
 Provides-Extra: sharepoint
-Requires-Dist: msal; extra == "sharepoint"
 Requires-Dist: Office365-REST-Python-Client; extra == "sharepoint"
+Requires-Dist: msal; extra == "sharepoint"
 Provides-Extra: salesforce
 Requires-Dist: simple-salesforce; extra == "salesforce"
 Provides-Extra: sftp
@@ -182,12 +182,12 @@ Requires-Dist: httpx; extra == "vectara"
 Requires-Dist: requests; extra == "vectara"
 Requires-Dist: aiofiles; extra == "vectara"
 Provides-Extra: vastdb
-Requires-Dist: ibis; extra == "vastdb"
-Requires-Dist: vastdb; extra == "vastdb"
 Requires-Dist: pyarrow; extra == "vastdb"
+Requires-Dist: vastdb; extra == "vastdb"
+Requires-Dist: ibis; extra == "vastdb"
 Provides-Extra: zendesk
-Requires-Dist: httpx; extra == "zendesk"
 Requires-Dist: bs4; extra == "zendesk"
+Requires-Dist: httpx; extra == "zendesk"
 Requires-Dist: aiofiles; extra == "zendesk"
 Provides-Extra: embed-huggingface
 Requires-Dist: sentence-transformers; extra == "embed-huggingface"

{unstructured_ingest-0.5.15 → unstructured_ingest-0.5.16}/test/integration/connectors/test_zendesk.py

@@ -1,6 +1,5 @@
 import os
 from pathlib import Path
-from typing import Optional
 
 import pytest
 
@@ -21,20 +20,20 @@ from unstructured_ingest.v2.processes.connectors.zendesk.zendesk import (
     ZendeskIndexerConfig,
 )
 
+SUBDOMAIN = "unstructuredhelp"
+EMAIL = "test@unstructured.io"
 
-async def zendesk_source_test(
-    tmp_path: Path,
-    token: Optional[str] = None,
-    email: Optional[str] = None,
-    subdomain: Optional[str] = None,
-):
 
-    access_config = ZendeskAccessConfig(api_token=token)
+@pytest.mark.asyncio
+@pytest.mark.tags(SOURCE_TAG, CONNECTOR_TYPE, UNCATEGORIZED_TAG)
+@requires_env("ZENDESK_TOKEN")
+async def test_zendesk_source_tickets(temp_dir: Path):
+    access_config = ZendeskAccessConfig(api_token=os.environ["ZENDESK_TOKEN"])
     connection_config = ZendeskConnectionConfig(
-        subdomain=subdomain, email=email, access_config=access_config
+        subdomain=SUBDOMAIN, email=EMAIL, access_config=access_config
     )
 
-    index_config = ZendeskIndexerConfig(batch_size=2, item_type="tickets")
+    index_config = ZendeskIndexerConfig(item_type="tickets")
 
     indexer = ZendeskIndexer(
         connection_config=connection_config,
@@ -43,7 +42,7 @@ async def zendesk_source_test(
     )
 
     # handle downloader.
-    download_config = ZendeskDownloaderConfig(download_dir=tmp_path)
+    download_config = ZendeskDownloaderConfig(download_dir=temp_dir)
 
     downloader = ZendeskDownloader(
         connection_config=connection_config,
@@ -57,26 +56,23 @@ async def zendesk_source_test(
         downloader=downloader,
         configs=SourceValidationConfigs(
             test_id="zendesk-tickets",
-            expected_num_files=4,
+            expected_num_files=8,
             validate_file_data=False,
             validate_downloaded_files=True,
         ),
     )
 
 
-async def zendesk_source_articles_test(
-    tmp_path: Path,
-    token: Optional[str] = None,
-    email: Optional[str] = None,
-    subdomain: Optional[str] = None,
-):
-
-    access_config = ZendeskAccessConfig(api_token=token)
+@pytest.mark.asyncio
+@pytest.mark.tags(SOURCE_TAG, CONNECTOR_TYPE, UNCATEGORIZED_TAG)
+@requires_env("ZENDESK_TOKEN")
+async def test_zendesk_source_articles(temp_dir):
+    access_config = ZendeskAccessConfig(api_token=os.environ["ZENDESK_TOKEN"])
     connection_config = ZendeskConnectionConfig(
-        subdomain=subdomain, email=email, access_config=access_config
+        subdomain=SUBDOMAIN, email=EMAIL, access_config=access_config
     )
 
-    index_config = ZendeskIndexerConfig(batch_size=2, item_type="articles")
+    index_config = ZendeskIndexerConfig(item_type="articles")
 
     indexer = ZendeskIndexer(
         connection_config=connection_config,
@@ -85,7 +81,7 @@ async def zendesk_source_articles_test(
     )
 
     # handle downloader.
-    download_config = ZendeskDownloaderConfig(download_dir=tmp_path, extract_images=True)
+    download_config = ZendeskDownloaderConfig(download_dir=temp_dir, extract_images=True)
 
     downloader = ZendeskDownloader(
         connection_config=connection_config,
@@ -99,44 +95,26 @@ async def zendesk_source_articles_test(
         downloader=downloader,
         configs=SourceValidationConfigs(
             test_id="zendesk-articles",
-            expected_num_files=4,
-            validate_file_data=False,
+            expected_num_files=8,
+            validate_file_data=True,
             validate_downloaded_files=True,
         ),
     )
 
 
-@pytest.mark.asyncio
 @pytest.mark.tags(SOURCE_TAG, CONNECTOR_TYPE, UNCATEGORIZED_TAG)
-@requires_env("ZENDESK_TOKEN")
-async def test_zendesk_source(temp_dir):
-    await zendesk_source_test(
-        tmp_path=temp_dir,
-        token=os.environ["ZENDESK_TOKEN"],
-        email="test@unstructured.io",
-        subdomain="unstructuredhelp",
+def test_zendesk_source_articles_fail(temp_dir):
+    access_config = ZendeskAccessConfig(api_token="FAKE_TOKEN")
+    connection_config = ZendeskConnectionConfig(
+        subdomain=SUBDOMAIN, email=EMAIL, access_config=access_config
     )
 
+    index_config = ZendeskIndexerConfig(item_type="tickets")
 
-@pytest.mark.asyncio
-@pytest.mark.tags(SOURCE_TAG, CONNECTOR_TYPE, UNCATEGORIZED_TAG)
-@requires_env("ZENDESK_TOKEN")
-async def test_zendesk_source_articles(temp_dir):
-    await zendesk_source_articles_test(
-        tmp_path=temp_dir,
-        token=os.environ["ZENDESK_TOKEN"],
-        email="test@unstructured.io",
-        subdomain="unstructuredhelp",
+    indexer = ZendeskIndexer(
+        connection_config=connection_config,
+        index_config=index_config,
+        connector_type=CONNECTOR_TYPE,
     )
-
-
-@pytest.mark.asyncio
-@pytest.mark.tags(SOURCE_TAG, CONNECTOR_TYPE, UNCATEGORIZED_TAG)
-async def test_zendesk_source_articles_fail(temp_dir):
     with pytest.raises(expected_exception=UserAuthError):
-        await zendesk_source_articles_test(
-            tmp_path=temp_dir,
-            token="FORCE_FAIL_TOKEN",
-            email="test@unstructured.io",
-            subdomain="unstructuredhelp",
-        )
+        indexer.precheck()

{unstructured_ingest-0.5.15 → unstructured_ingest-0.5.16}/test/integration/connectors/utils/validation/source.py

@@ -103,7 +103,7 @@ def check_contents(
         file_data_path = expected_output_dir / f"{file_data.identifier}.json"
         with file_data_path.open("r") as file:
             expected_file_data_contents = json.load(file)
-        current_file_data_contents = file_data.model_dump()
+        current_file_data_contents = json.loads(file_data.model_dump_json())
         expected_file_data_contents = configs.omit_ignored_fields(expected_file_data_contents)
         current_file_data_contents = configs.omit_ignored_fields(current_file_data_contents)
         diff = DeepDiff(expected_file_data_contents, current_file_data_contents)
@@ -184,7 +184,7 @@ def update_fixtures(
         for file_data in all_file_data:
             file_data_path = file_data_output_path / f"{file_data.identifier}.json"
             with file_data_path.open(mode="w") as f:
-                json.dump(file_data.model_dump(), f, indent=2)
+                f.write(file_data.model_dump_json(indent=2))
 
     # Record file structure of download directory
     download_files = get_files(dir_path=download_dir)
@@ -216,7 +216,9 @@ def run_all_validations(
             len(predownload_file_data) == expected_number_indexed_file_data
         ), f"expected {expected_number_indexed_file_data} but got {len(predownload_file_data)}"
     if expected_num_files := configs.expected_num_files:
-        assert len(postdownload_file_data) == expected_num_files
+        assert (
+            len(postdownload_file_data) == expected_num_files
+        ), f"expected {expected_num_files} but got {len(postdownload_file_data)}"
 
     for pre_data, post_data in zip(predownload_file_data, postdownload_file_data):
         configs.run_file_data_validation(

unstructured_ingest-0.5.16/unstructured_ingest/__version__.py

@@ -0,0 +1 @@
+__version__ = "0.5.16"  # pragma: no cover

{unstructured_ingest-0.5.15 → unstructured_ingest-0.5.16}/unstructured_ingest/v2/pipeline/steps/download.py

@@ -88,9 +88,9 @@ class DownloadStep(PipelineStep):
                 f"match size of local file: {file_size_bytes}, updating"
             )
             file_data.metadata.filesize_bytes = file_size_bytes
-        logger.debug(f"updating file data with new content: {file_data.model_dump()}")
+        logger.debug(f"updating file data with new content: {file_data.model_dump_json()}")
         with file_data_path.open("w") as file:
-            json.dump(file_data.model_dump(), file, indent=2)
+            file.write(file_data.model_dump_json(indent=2))
 
     async def _run_async(self, fn: Callable, file_data_path: str) -> list[DownloadStepResponse]:
         file_data = file_data_from_file(path=file_data_path)
@@ -173,7 +173,7 @@ class DownloadStep(PipelineStep):
         filepath = (self.cache_dir / filename).resolve()
         filepath.parent.mkdir(parents=True, exist_ok=True)
         with open(str(filepath), "w") as f:
-            json.dump(file_data.model_dump(), f, indent=2)
+            f.write(file_data.model_dump_json(indent=2))
         return str(filepath)
 
     def get_hash(self, extras: Optional[list[str]]) -> str:

{unstructured_ingest-0.5.15 → unstructured_ingest-0.5.16}/unstructured_ingest/v2/pipeline/steps/index.py

@@ -37,14 +37,14 @@ class IndexStep(PipelineStep):
     @instrument(span_name=STEP_ID)
     def run(self) -> Generator[str, None, None]:
         for file_data in self.process.run():
-            logger.debug(f"generated file data: {file_data.model_dump()}")
+            logger.debug(f"generated file data: {file_data.model_dump_json()}")
             try:
                 record_hash = self.get_hash(extras=[file_data.identifier])
                 filename = f"{record_hash}.json"
                 filepath = (self.cache_dir / filename).resolve()
                 filepath.parent.mkdir(parents=True, exist_ok=True)
                 with open(str(filepath), "w") as f:
-                    json.dump(file_data.model_dump(), f, indent=2)
+                    f.write(file_data.model_dump_json(indent=2))
                 yield str(filepath)
             except Exception as e:
                 logger.error(f"failed to create index for file data: {file_data}", exc_info=True)
@@ -54,14 +54,14 @@ class IndexStep(PipelineStep):
 
     async def run_async(self) -> AsyncGenerator[str, None]:
         async for file_data in self.process.run_async():
-            logger.debug(f"generated file data: {file_data.model_dump()}")
+            logger.debug(f"generated file data: {file_data.model_dump_json()}")
             try:
                 record_hash = self.get_hash(extras=[file_data.identifier])
                 filename = f"{record_hash}.json"
                 filepath = (self.cache_dir / filename).resolve()
                 filepath.parent.mkdir(parents=True, exist_ok=True)
                 with open(str(filepath), "w") as f:
-                    json.dump(file_data.model_dump(), f, indent=2)
+                    f.write(file_data.model_dump_json(indent=2))
                 yield str(filepath)
             except Exception as e:
                 logger.error(f"failed to create index for file data: {file_data}", exc_info=True)

{unstructured_ingest-0.5.15 → unstructured_ingest-0.5.16}/unstructured_ingest/v2/processes/connectors/databricks/volumes_table.py

@@ -61,7 +61,7 @@ class DatabricksVolumeDeltaTableUploader(Uploader):
                         self.upload_config.database, ", ".join(databases)
                     )
                 )
-            cursor.execute("SHOW TABLES")
+            cursor.execute(f"SHOW TABLES IN {self.upload_config.database}")
             table_names = [r[1] for r in cursor.fetchall()]
             if self.upload_config.table_name not in table_names:
                 raise ValueError(

unstructured_ingest-0.5.16/unstructured_ingest/v2/processes/connectors/zendesk/client.py

@@ -0,0 +1,308 @@
+from dataclasses import dataclass, field
+from datetime import datetime
+from typing import TYPE_CHECKING, Any, AsyncGenerator, Literal, Optional, Union
+
+from pydantic import BaseModel, Field, HttpUrl
+
+from unstructured_ingest.utils.dep_check import requires_dependencies
+from unstructured_ingest.utils.string_and_date_utils import fix_unescaped_unicode
+from unstructured_ingest.v2.errors import ProviderError, RateLimitError, UserAuthError, UserError
+from unstructured_ingest.v2.logger import logger
+
+if TYPE_CHECKING:
+    from httpx import AsyncClient, Client
+
+
+class Attachment(BaseModel):
+    # https://developer.zendesk.com/api-reference/ticketing/tickets/ticket-attachments/#json-format
+    content_type: Optional[str] = None
+
+
+class Via(BaseModel):
+    # https://developer.zendesk.com/documentation/ticketing/reference-guides/via-object-reference/
+    channel: Union[int, str]
+    source: dict = Field(default_factory=dict)
+
+
+class ZendeskComment(BaseModel):
+    # https://developer.zendesk.com/api-reference/ticketing/tickets/ticket_comments/#json-format
+    attachments: list[Attachment] = Field(default_factory=list)
+    audit_id: Optional[int] = None
+    author_id: Optional[int] = None
+    body: Optional[str] = None
+    created_at: Optional[datetime] = None
+    html_body: Optional[str] = None
+    id: Optional[int] = None
+    metadata: Optional[dict] = None
+    plain_body: Optional[str] = None
+    public: Optional[bool] = None
+    comment_type: Literal["Comment", "VoiceComment"] = Field(alias="type")
+    uploads: list[str] = Field(default_factory=list)
+    via: Optional[Via] = None
+
+    def as_text(self) -> str:
+        all_data = self.model_dump()
+        filtered_data = {
+            k: v
+            for k, v in all_data.items()
+            if k in ["id", "author_id", "body", "created_at"] and v is not None
+        }
+        return "".join(
+            [f"{v}\n" for v in ["comment"] + [f"{k}: {v}" for k, v in filtered_data.items()]]
+        )
+
+
+class ZendeskTicket(BaseModel):
+    # https://developer.zendesk.com/api-reference/ticketing/tickets/tickets/#json-format
+    allow_attachments: bool = True
+    allow_channelback: bool = True
+    assignee_email: Optional[str] = None
+    assignee_id: Optional[int] = None
+    attribute_value_ids: list[int] = Field(default_factory=list)
+    brand_id: Optional[int] = None
+    collaborator_ids: list[int] = Field(default_factory=list)
+    collaborators: list[Union[int, str, dict[str, str]]] = Field(default_factory=list)
+    comment: Optional[ZendeskComment] = None
+    created_at: Optional[datetime] = None
+    custom_fields: list[dict[str, Any]] = Field(default_factory=list)
+    custom_status_id: Optional[int] = None
+    description: Optional[str] = None
+    due_at: Optional[datetime] = None
+    email_cc_ids: list[int] = Field(default_factory=list)
+    email_ccs: list[dict[str, str]] = Field(default_factory=list)
+    external_id: Optional[str] = None
+    follower_ids: list[int] = Field(default_factory=list)
+    followers: list[dict[str, str]] = Field(default_factory=list)
+    followup_ids: list[int] = Field(default_factory=list)
+    forum_topic_id: Optional[int] = None
+    from_messaging_channel: bool
+    generated_timestamp: Optional[datetime] = None
+    group_id: Optional[int] = None
+    has_incidents: bool = False
+    id: Optional[int] = None
+    is_public: bool = False
+    macro_id: Optional[int] = None
+    macro_ids: list[int] = Field(default_factory=list)
+    metadata: dict[str, Any] = Field(default_factory=dict)
+    organization_id: Optional[int] = None
+    priority: Optional[Literal["urgent", "high", "normal", "low"]] = None
+    problem_id: Optional[int] = None
+    raw_subject: Optional[str] = None
+    recipient: Optional[str] = None
+    requester: dict[str, str] = Field(default_factory=dict)
+    requester_id: int
+    safe_update: Optional[bool] = None
+    satisfaction_rating: Optional[Union[str, dict[str, Any]]] = None
+    sharing_agreement_ids: list[int] = Field(default_factory=list)
+    status: Optional[Literal["new", "open", "pending", "hold", "solved", "closed"]] = None
+    subject: Optional[str] = None
+    submitter_id: Optional[int] = None
+    tags: list[str] = Field(default_factory=list)
+    ticket_form_id: Optional[int] = None
+    ticket_type: Optional[Literal["problem", "incident", "question", "task"]] = Field(
+        default=None, alias="type"
+    )
+    updated_at: Optional[datetime] = None
+    updated_stamp: Optional[str] = None
+    url: Optional[HttpUrl] = None
+    via: Optional[Via] = None
+    via_followup_source_id: Optional[int] = None
+    via_id: Optional[int] = None
+    voice_comment: Optional[dict] = None
+
+    def as_text(self) -> str:
+        all_data = self.model_dump()
+        filtered_data = {
+            k: v
+            for k, v in all_data.items()
+            if k in ["id", "subject", "description", "created_at"] and v is not None
+        }
+        return "".join(
+            [f"{v}\n" for v in ["ticket"] + [f"{k}: {v}" for k, v in filtered_data.items()]]
+        )
+
+
+class ZendeskArticle(BaseModel):
+    # https://developer.zendesk.com/api-reference/help_center/help-center-api/articles/#json-format
+    author_id: Optional[int] = None
+    body: Optional[str] = None
+    comments_disabled: bool = False
+    content_tag_ids: list[str] = Field(default_factory=list)
+    created_at: Optional[datetime] = None
+    draft: bool = False
+    edited_at: Optional[datetime] = None
+    html_url: Optional[HttpUrl] = None
+    id: int
+    label_names: list[str] = Field(default_factory=list)
+    locale: str
+    outdated: bool = False
+    outdated_locales: list[str] = Field(default_factory=list)
+    permission_group_id: int
+    position: Optional[int] = None
+    promoted: bool = False
+    section_id: Optional[int] = None
+    source_locale: Optional[str] = None
+    title: str
+    updated_at: Optional[datetime] = None
+    url: Optional[HttpUrl] = None
+    user_segment_id: Optional[int] = None
+    user_segment_ids: list[int] = Field(default_factory=list)
+    vote_count: Optional[int] = None
+    vote_sum: Optional[int] = None
+
+    def as_html(self) -> str:
+        html = self.body
+        if title := self.title:
+            html = f"<h1>{title}</h1>{html}"
+        return fix_unescaped_unicode(f"<body class='Document' >{html}</body>")
+
+
+class ZendeskArticleAttachment(BaseModel):
+    # https://developer.zendesk.com/api-reference/help_center/help-center-api/article_attachments/#json-format
+    article_id: Optional[int] = None
+    content_type: Optional[str] = None
+    content_url: Optional[HttpUrl] = None
+    created_at: Optional[datetime] = None
+    guide_media_id: Optional[str] = None
+    id: Optional[int] = None
+    inline: bool = False
+    locale: Optional[str] = None
+    size: Optional[int] = None
+    updated_at: Optional[datetime] = None
+    url: Optional[HttpUrl] = None
+
+
+@dataclass
+class ZendeskClient:
+    token: str
+    subdomain: str
+    email: str
+    max_page_size: int = 100
+    _async_client: "AsyncClient" = field(init=False, default=None)
+    _client: "Client" = field(init=False, default=None)
+    _base_url: str = field(init=False, default=None)
+
+    async def __aenter__(self) -> "ZendeskClient":
+        return self
+
+    async def __aexit__(self, exc_type, exc_val, exc_tb):
+        await self._async_client.aclose()
+
+    @requires_dependencies(["httpx"], extras="zendesk")
+    def __post_init__(self):
+        import httpx
+
+        auth = f"{self.email}/token", self.token
+        self._client = httpx.Client(auth=auth)
+        self._async_client = httpx.AsyncClient(auth=auth)
+        self._base_url = f"https://{self.subdomain}.zendesk.com/api/v2"
+
+        # Run check
+        try:
+            url_to_check = f"{self._base_url}/groups.json"
+            resp = self._client.head(url_to_check)
+            resp.raise_for_status()
+        except Exception as e:
+            raise self.wrap_error(e=e)
+
+    @requires_dependencies(["httpx"], extras="zendesk")
+    def wrap_error(self, e: Exception) -> Exception:
+        import httpx
+
+        if not isinstance(e, httpx.HTTPStatusError):
+            logger.error(f"unhandled exception from Zendesk client: {e}", exc_info=True)
+            return e
+        url = e.request.url
+        response_code = e.response.status_code
+        if response_code == 401:
+            logger.error(
+                f"Failed to connect via auth,"
+                f"{url} using zendesk response, status code {response_code}"
+            )
+            return UserAuthError(e)
+        if response_code == 429:
+            logger.error(
+                f"Failed to connect via rate limits"
+                f"{url} using zendesk response, status code {response_code}"
+            )
+            return RateLimitError(e)
+        if 400 <= response_code < 500:
+            logger.error(
+                f"Failed to connect to {url} using zendesk response, status code {response_code}"
+            )
+            return UserError(e)
+        if response_code > 500:
+            logger.error(
+                f"Failed to connect to {url} using zendesk response, status code {response_code}"
+            )
+            return ProviderError(e)
+        logger.error(f"unhandled http status error from Zendesk client: {e}", exc_info=True)
+        return e
+
+    async def fetch_content(self, url: str, content_key: str) -> AsyncGenerator[dict, None]:
+        url = f"{url}?page[size]={self.max_page_size}"
+        while True:
+            try:
+                response = await self._async_client.get(url)
+                response.raise_for_status()
+            except Exception as e:
+                raise self.wrap_error(e=e)
+
+            data = response.json()
+            for content in data[content_key]:
+                yield content
+
+            has_more = data.get("meta", {}).get("has_more", False)
+            if not has_more:
+                break
+
+            url = data["links"]["next"]
+
+    async def get_articles(self) -> AsyncGenerator[ZendeskArticle, None]:
+        """
+        Retrieves article content from Zendesk asynchronously.
+        """
+        article_url = f"https://{self.subdomain}.zendesk.com/api/v2/help_center/articles.json"
+
+        try:
+            async for article_dict in self.fetch_content(url=article_url, content_key="articles"):
+                yield ZendeskArticle.model_validate(article_dict)
+        except Exception as e:
+            raise self.wrap_error(e=e)
+
+    async def get_comments(self, ticket_id: int) -> AsyncGenerator[ZendeskComment, None]:
+        comments_url = f"https://{self.subdomain}.zendesk.com/api/v2/tickets/{ticket_id}/comments"
+
+        try:
+            async for comment_dict in self.fetch_content(url=comments_url, content_key="comments"):
+                yield ZendeskComment.model_validate(comment_dict)
+        except Exception as e:
+            raise self.wrap_error(e=e)
+
+    async def get_tickets(self) -> AsyncGenerator[ZendeskTicket, None]:
+        tickets_url = f"https://{self.subdomain}.zendesk.com/api/v2/tickets"
+
+        try:
+            async for ticket_dict in self.fetch_content(url=tickets_url, content_key="tickets"):
+                yield ZendeskTicket.model_validate(ticket_dict)
+        except Exception as e:
+            raise self.wrap_error(e=e)
+
+    async def get_article_attachments(
+        self, article_id: int
+    ) -> AsyncGenerator[ZendeskArticleAttachment, None]:
+        """
+        Handles article attachments such as images and stores them as UTF-8 encoded bytes.
+        """
+        article_attachment_url = (
+            f"https://{self.subdomain}.zendesk.com/api/v2/help_center/"
+            f"articles/{article_id}/attachments"
+        )
+
+        try:
+            async for attachment_dict in self.fetch_content(
+                url=article_attachment_url, content_key="article_attachments"
+            ):
+                yield ZendeskArticleAttachment.model_validate(attachment_dict)
+        except Exception as e:
+            raise self.wrap_error(e=e)