castor-extractor 0.16.15__tar.gz → 0.17.0__tar.gz

{castor_extractor-0.16.15 → castor_extractor-0.17.0}/CHANGELOG.md

@@ -1,5 +1,9 @@
 # Changelog
 
+## 0.17.0 - 2024-06-10
+
+* Uploader: redirect to the proxy, replace credentials with token
+
 ## 0.16.15 - 2024-06-07
 
 * Tableau: extract database_name for CustomSQLTables

{castor_extractor-0.16.15 → castor_extractor-0.17.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: castor-extractor
-Version: 0.16.15
+Version: 0.17.0
 Summary: Extract your metadata assets.
 Home-page: https://www.castordoc.com/
 License: EULA

{castor_extractor-0.16.15 → castor_extractor-0.17.0}/castor_extractor/commands/upload.py

@@ -13,10 +13,10 @@ logging.basicConfig(level=logging.INFO, format="%(levelname)s - %(message)s")
 def _args():
     parser = argparse.ArgumentParser()
     parser.add_argument(
-        "-c",
-        "--credentials",
+        "-k",
+        "--token",
         required=True,
-        help="""Path to credentials or credentials as string""",
+        help="""API token provided by Castor""",
     )
     parser.add_argument(
         "-s",
@@ -44,7 +44,7 @@ def _args():
     )
     parsed = parser.parse_args()
     return {
-        "credentials": parsed.credentials,
+        "token": parsed.token,
         "source_id": parsed.source_id,
         "file_path": parsed.file_path,
         "directory_path": parsed.directory_path,

{castor_extractor-0.16.15 → castor_extractor-0.17.0}/castor_extractor/uploader/constant.py

@@ -1,6 +1,13 @@
 from enum import Enum
 
-EXTRACTION_BUCKET = "extraction-storage"
+from ..utils import RetryStrategy
+
+# url of the gcs proxy
+INGEST_URL = "https://ingest.castordoc.com"
+
+RETRY_BASE_MS = 10_000
+RETRY_JITTER_MS = 1_000
+RETRY_STRATEGY = RetryStrategy.LINEAR
 
 
 class FileType(Enum):

castor_extractor-0.17.0/castor_extractor/uploader/upload.py

@@ -0,0 +1,133 @@
+#!/usr/bin/env python3
+import logging
+import ntpath
+from datetime import datetime
+from typing import Dict, Iterable, Optional, Tuple
+from uuid import UUID
+
+import requests
+
+from ..utils.retry import retry
+from .constant import (
+    INGEST_URL,
+    PATH_TEMPLATES,
+    RETRY_BASE_MS,
+    RETRY_JITTER_MS,
+    RETRY_STRATEGY,
+    FileType,
+)
+from .env import get_blob_env
+from .utils import iter_files
+
+logger = logging.getLogger(__name__)
+
+_EXCEPTIONS = (
+    requests.exceptions.Timeout,
+    requests.exceptions.ConnectTimeout,
+)
+
+
+def _path_and_url(
+    source_id: UUID,
+    file_type: FileType,
+    file_path: str,
+) -> Tuple[str, str]:
+
+    now = datetime.utcnow()
+    timestamp = int(now.timestamp())
+    filename = ntpath.basename(file_path)
+    path_template = PATH_TEMPLATES[file_type]
+    path = path_template.format(
+        timestamp=timestamp,
+        source_id=source_id,
+        filename=filename,
+    )
+
+    url = f"{INGEST_URL}/{path}"
+
+    return path, url
+
+
+def _headers(token: str) -> Dict:
+    return {
+        "Authorization": f"Token {token}",
+        "Accept": "text/csv, application/json",
+    }
+
+
+def _upload(
+    token: str,
+    source_id: UUID,
+    file_path: str,
+    file_type: FileType,
+) -> None:
+    """
+    Upload the given file to Google Cloud Storage (GCS)
+    - Don't call GCS API directly
+    - Call the ingestion proxy which handles authorisation and uploading
+    """
+    path, url = _path_and_url(source_id, file_type, file_path)
+    headers = _headers(token)
+    timeout, max_retries = get_blob_env()
+
+    with open(file_path, "rb") as file_content:
+
+        @retry(
+            exceptions=_EXCEPTIONS,
+            max_retries=max_retries,
+            base_ms=RETRY_BASE_MS,
+            jitter_ms=RETRY_JITTER_MS,
+            strategy=RETRY_STRATEGY,
+        )
+        def _request_post():
+            response = requests.post(
+                url=url,
+                headers=headers,
+                data=file_content,
+                timeout=timeout,
+            )
+            response.raise_for_status()
+
+        _request_post()
+
+    logger.info(f"Uploaded {file_path} as {file_type.value} to {path}")
+
+
+def upload_manifest(
+    token: str,
+    source_id: UUID,
+    file_path: str,
+) -> None:
+    """
+    credentials: path to file or dict
+    source_id: id for the source
+    file_path: path to the local manifest to upload
+    """
+    _upload(token, source_id, file_path, FileType.DBT)
+
+
+def upload(
+    token: str,
+    source_id: UUID,
+    file_type: FileType,
+    file_path: Optional[str] = None,
+    directory_path: Optional[str] = None,
+) -> None:
+    """
+    credentials: path to file or dict
+    source_id: id for the source
+    file_type: type of file(s) uploaded - see FileType Enum
+    file_path: path to the local visualization or warehouse file to upload
+    directory_path: path to the local directory containing files to upload
+    """
+    files: Iterable[str]
+    if directory_path:
+        files = iter_files(directory_path)
+    elif file_path:
+        files = [file_path]
+    else:
+        message = "either file_path or directory_path should be defined"
+        raise ValueError(message)
+
+    for file_ in files:
+        _upload(token, source_id, file_, file_type)

{castor_extractor-0.16.15 → castor_extractor-0.17.0}/castor_extractor/uploader/upload_test.py

@@ -1,7 +1,7 @@
 from uuid import UUID
 
-from .constant import FileType
-from .upload import _path
+from .constant import INGEST_URL, FileType
+from .upload import _path_and_url
 
 
 def test__path():
@@ -9,5 +9,6 @@ def test__path():
     file_type = FileType.VIZ
     file_path = "filename"
 
-    path = _path(source_id, file_type, file_path)
+    path, url = _path_and_url(source_id, file_type, file_path)
     assert path == f"visualization-{source_id}/{file_path}"
+    assert url == f"{INGEST_URL}/{path}"

{castor_extractor-0.16.15 → castor_extractor-0.17.0}/castor_extractor/uploader/utils.py

@@ -1,13 +1,23 @@
+import logging
 import os
 from typing import Iterator
 
+logger = logging.getLogger(__name__)
+
+_ALLOWED_EXTENSION = (".json", ".csv")
+
 
 def iter_files(repository_path: str) -> Iterator[str]:
     """
     Given a repository path yield all files in that repository
+    Removes file whose extension is not allowed
     """
 
     for file in os.listdir(repository_path):
+        _, ext = os.path.splitext(file)
+        if ext not in _ALLOWED_EXTENSION:
+            logger.info(f"Forbidden file extension : skipping {file}")
+            continue
         file_path = os.path.join(repository_path, file)
 
         if os.path.isfile(file_path):

{castor_extractor-0.16.15 → castor_extractor-0.17.0}/castor_extractor/warehouse/abstract/asset.py

@@ -32,6 +32,7 @@ class WarehouseAsset(ExternalAsset):
             WarehouseAsset.ADDITIONAL_TABLE_LINEAGE,
             WarehouseAsset.EXTERNAL_COLUMN_LINEAGE,
             WarehouseAsset.EXTERNAL_TABLE_LINEAGE,
+            WarehouseAsset.FUNCTION,
         }
 
 

{castor_extractor-0.16.15 → castor_extractor-0.17.0}/pyproject.toml

@@ -3,7 +3,7 @@ requires = ["setuptools>=61.2"]
 
 [tool.poetry]
 name = "castor-extractor"
-version = "0.16.15"
+version = "0.17.0"
 description = "Extract your metadata assets."
 authors = ["Castor <support@castordoc.com>"]
 license = "EULA"

castor_extractor-0.16.15/castor_extractor/uploader/upload.py

@@ -1,116 +0,0 @@
-#!/usr/bin/env python3
-import json
-import logging
-import ntpath
-from datetime import datetime
-from typing import Iterable, Optional, Union
-from uuid import UUID
-
-from google.cloud import storage  # type: ignore
-
-from .constant import EXTRACTION_BUCKET, PATH_TEMPLATES, FileType
-from .env import get_blob_env
-from .utils import file_exist, iter_files
-
-logger = logging.getLogger(__name__)
-
-
-def _client(credentials: Union[str, dict]) -> storage.Client:
-    """supports dict, string or path to the JSON file"""
-    if isinstance(credentials, dict):
-        return storage.Client.from_service_account_info(credentials)
-    if file_exist(credentials):
-        return storage.Client.from_service_account_json(credentials)
-    if isinstance(credentials, str):
-        credentials = json.loads(credentials)
-        return storage.Client.from_service_account_info(credentials)
-    raise ValueError("needs path or dict for credentials")
-
-
-def _path(source_id: UUID, file_type: FileType, file_path: str) -> str:
-    now = datetime.utcnow()
-    timestamp = int(now.timestamp())
-    filename = ntpath.basename(file_path)
-
-    path_template = PATH_TEMPLATES[file_type]
-    return path_template.format(
-        timestamp=timestamp,
-        source_id=source_id,
-        filename=filename,
-    )
-
-
-def _get_blob(
-    credentials: Union[str, dict],
-    source_id: UUID,
-    file_path: str,
-    file_type: FileType,
-) -> storage.Blob:
-    """get the target blob to upload to"""
-    client = _client(credentials)
-    path = _path(source_id, file_type, file_path)
-
-    bucket = client.bucket(EXTRACTION_BUCKET)
-    return bucket.blob(path)
-
-
-def _upload(
-    credentials: Union[str, dict],
-    source_id: UUID,
-    file_path: str,
-    file_type: FileType,
-) -> None:
-    """
-    credentials: path to file or dict
-    source_id: id for the source
-    file_type: type of file to upload
-    file_path: path to the local file to upload
-    """
-    timeout, retries = get_blob_env()
-
-    blob = _get_blob(credentials, source_id, file_path, file_type)
-    with open(file_path, "rb") as f:
-        blob.upload_from_file(f, timeout=timeout, num_retries=retries)
-    logger.info(
-        f"uploaded {file_path} as {file_type.value} to {blob.public_url}",
-    )
-
-
-def upload_manifest(
-    credentials: Union[str, dict],
-    source_id: UUID,
-    file_path: str,
-) -> None:
-    """
-    credentials: path to file or dict
-    source_id: id for the source
-    file_path: path to the local manifest to upload
-    """
-    _upload(credentials, source_id, file_path, FileType.DBT)
-
-
-def upload(
-    credentials: Union[str, dict],
-    source_id: UUID,
-    file_type: FileType,
-    file_path: Optional[str] = None,
-    directory_path: Optional[str] = None,
-) -> None:
-    """
-    credentials: path to file or dict
-    source_id: id for the source
-    file_type: type of file(s) uploaded - see FileType Enum
-    file_path: path to the local visualization or warehouse file to upload
-    directory_path: path to the local directory containing files to upload
-    """
-    files: Iterable[str]
-    if directory_path:
-        files = iter_files(directory_path)
-    elif file_path:
-        files = [file_path]
-    else:
-        message = "either file_path or directory_path should be defined"
-        raise ValueError(message)
-
-    for file_ in files:
-        _upload(credentials, source_id, file_, file_type)