datalab-api 0.3.1__tar.gz → 0.4.0__tar.gz

{datalab_api-0.3.1/src/datalab_api.egg-info → datalab_api-0.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datalab-api
-Version: 0.3.1
+Version: 0.4.0
 Summary: A Python API for instances of the datalab data management platform (https://github.com/datalab-org/datalab).
 Author-email: Matthew Evans <matthew@ml-evs.science>
 License-Expression: MIT
@@ -13,6 +13,7 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
 Classifier: Intended Audience :: Science/Research
 Classifier: Intended Audience :: Information Technology
 Classifier: Topic :: Other/Nonlisted Topic
@@ -38,9 +39,6 @@ Requires-Dist: mkdocs-jupyter; extra == "docs"
 Provides-Extra: cli
 Requires-Dist: typer~=0.9; extra == "cli"
 Requires-Dist: click-shell~=2.1; extra == "cli"
-Provides-Extra: cheminventory-helper
-Requires-Dist: pandas~=2.2; extra == "cheminventory-helper"
-Requires-Dist: openpyxl~=3.0; extra == "cheminventory-helper"
 Provides-Extra: all
 Requires-Dist: datalab-api[cli,dev,docs]; extra == "all"
 Dynamic: license-file
@@ -61,7 +59,7 @@ A simple Python API that can interact with [*datalab*](https://github.com/datala
 
 The idea here is to provide a set of utility functions and models for manipulating samples, cells, inventory, files, users and metadata associated with *datalab* entries in an automated way.
 
-This API may not expose all the functionality available in a given *datalab* instance, and the idea would be that this package can support multiple versions of the underlying [*datalab* REST API](https://docs.datalab-org.io/en/stable/rest_api/).
+This API may not expose all the functionality available in a given *datalab* instance, and the idea would be that this package can support multiple versions of the underlying [*datalab* REST API](https://docs.datalab-org.io).
 This means that the API is primarily *functional* in nature, as opposed to object-oriented, since *datalab* instances are free to use their own custom data models.
 The available schemas are reported as instance metadata and in the future object-oriented models may be able to be genereated directly in the client (so e.g., the returned data would be Python objects like `Sample` rather than JSON data).
 

{datalab_api-0.3.1 → datalab_api-0.4.0}/README.md

@@ -14,7 +14,7 @@ A simple Python API that can interact with [*datalab*](https://github.com/datala
 
 The idea here is to provide a set of utility functions and models for manipulating samples, cells, inventory, files, users and metadata associated with *datalab* entries in an automated way.
 
-This API may not expose all the functionality available in a given *datalab* instance, and the idea would be that this package can support multiple versions of the underlying [*datalab* REST API](https://docs.datalab-org.io/en/stable/rest_api/).
+This API may not expose all the functionality available in a given *datalab* instance, and the idea would be that this package can support multiple versions of the underlying [*datalab* REST API](https://docs.datalab-org.io).
 This means that the API is primarily *functional* in nature, as opposed to object-oriented, since *datalab* instances are free to use their own custom data models.
 The available schemas are reported as instance metadata and in the future object-oriented models may be able to be genereated directly in the client (so e.g., the returned data would be Python objects like `Sample` rather than JSON data).
 

{datalab_api-0.3.1 → datalab_api-0.4.0}/pyproject.toml

@@ -16,6 +16,7 @@ classifiers = [
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
     "Intended Audience :: Science/Research",
     "Intended Audience :: Information Technology",
     "Topic :: Other/Nonlisted Topic",
@@ -52,8 +53,6 @@ cli = [
     "click-shell ~= 2.1", # REPL-like interface
 ]
 
-cheminventory-helper = ["pandas ~= 2.2", "openpyxl ~= 3.0"]
-
 all = ["datalab-api[dev,docs,cli]"]
 
 [project.scripts]

{datalab_api-0.3.1 → datalab_api-0.4.0}/src/datalab_api/__init__.py

@@ -5,7 +5,9 @@ from collections.abc import Iterable
 from pathlib import Path
 from typing import Any, Optional, Union
 
-from ._base import BaseDatalabClient, DuplicateItemError, __version__
+import httpx
+
+from ._base import BaseDatalabClient, DatalabAPIError, DuplicateItemError, __version__
 
 __all__ = ("DatalabClient", "DuplicateItemError", "__version__")
 
@@ -147,7 +149,7 @@ class DatalabClient(BaseDatalabClient):
         if collection_id is not None:
             try:
                 collection_immutable_id = self.get_collection(collection_id)[0]["immutable_id"]
-            except RuntimeError:
+            except (RuntimeError, DatalabAPIError):
                 self.create_collection(collection_id)
                 collection_immutable_id = self.get_collection(collection_id)[0]["immutable_id"]
             new_item["collections"] = new_item.get("collections", [])
@@ -285,6 +287,10 @@ class DatalabClient(BaseDatalabClient):
         if not file_path.exists():
             raise FileNotFoundError(f"File {file_path=} does not exist.")
 
+        # Use a longer read timeout for uploads, as the server may do
+        # significant processing before responding
+        upload_timeout = httpx.Timeout(self._timeout.connect, read=600.0, pool=self._timeout.pool)
+
         upload_url = f"{self.datalab_api_url}/upload-file/"
         with open(file_path, "rb") as file:
             files = {"file": (file_path.name, file)}
@@ -293,6 +299,7 @@ class DatalabClient(BaseDatalabClient):
                 files=files,
                 data={"item_id": item_id, "replace_file": None},
                 expected_status=201,
+                timeout=upload_timeout,
             )
         return upload
 
@@ -400,8 +407,18 @@ class DatalabClient(BaseDatalabClient):
             "save_to_db": True,
         }
 
-        resp = self._post(blocks_url, json=payload)
-        return resp["new_block_data"]
+        resp = self._post(blocks_url, expected_status=[200, 202], json=payload)
+
+        # If block is created synchronously, return the block data
+        if "new_block_data" in resp:
+            return resp["new_block_data"]
+
+        # Otherwise, we return the response which may contain information about the asynchronous block creation process
+        task_id = resp.get("task_id")
+        if task_id:
+            self.triggered_block_task_ids.add(task_id)
+
+        return resp
 
     def get_collection(self, collection_id: str) -> tuple[dict[str, Any], list[dict[str, Any]]]:
         """Get a collection with a given ID.

{datalab_api-0.3.1 → datalab_api-0.4.0}/src/datalab_api/_base.py

@@ -2,8 +2,9 @@ import logging
 import os
 import re
 import warnings
+from getpass import getpass
 from importlib.metadata import version
-from typing import Any, Optional
+from typing import Any, Optional, Union
 
 import httpx
 from rich.logging import RichHandler
@@ -37,6 +38,14 @@ class BaseDatalabClient(metaclass=AutoPrettyPrint):
     _headers: dict[str, str] = {}
     _timeout: httpx.Timeout = httpx.Timeout(10.0, read=60.0)
 
+    interactive: bool = True
+    """Whether the client is being used in an interactive context (e.g., CLI or notebook) or not.
+    Set to false if you want the script to error if no API key is found in the environment variables,
+    rather than prompting for input."""
+
+    triggered_block_task_ids: set[str] = set()
+    """A set of block task IDs that have been triggered during the lifetime of this client instance."""
+
     info: dict[str, Any] = {}
     """The `data` response from the `/info` endpoint of the Datalab API."""
 
@@ -189,9 +198,14 @@ class BaseDatalabClient(metaclass=AutoPrettyPrint):
                 api_key = api_key.strip("'").strip('"')
 
             if api_key is None:
-                raise ValueError(
-                    f"No API key found in environment variables {key_env_var}/<prefix>_{key_env_var}."
-                )
+                if self.interactive:
+                    api_key = getpass(
+                        f"Enter your API key for {self.datalab_api_url} (input will be hidden): "
+                    )
+                else:
+                    raise ValueError(
+                        f"No API key found in environment variables {key_env_var}/<prefix>_{key_env_var}."
+                    )
 
             self._api_key = api_key
 
@@ -215,14 +229,14 @@ class BaseDatalabClient(metaclass=AutoPrettyPrint):
             self._session.close()
 
     def _handle_response(
-        self, response: httpx.Response, url: str, expected_status: int = 200
+        self, response: httpx.Response, url: str, expected_status: Union[int, list[int]] = 200
     ) -> dict[str, Any]:
         """Handle HTTP response with consistent error handling.
 
         Args:
             response: The HTTP response object
             url: The URL that was requested
-            expected_status: The expected HTTP status code (default: 200)
+            expected_status: The expected HTTP status code(s) (default: 200)
 
         Returns:
             The JSON response data
@@ -248,7 +262,10 @@ class BaseDatalabClient(metaclass=AutoPrettyPrint):
             )
 
         # Handle other HTTP status code errors
-        if response.status_code != expected_status:
+        if isinstance(expected_status, int):
+            expected_status = [expected_status]
+
+        if response.status_code not in expected_status:
             try:
                 error_data = response.json()
                 error_message = error_data.get("message", str(response.content))
@@ -310,8 +327,27 @@ This is likely a server-side bug. Please report this issue to the datalab develo
 
         return data
 
+    def check_tasks(self) -> tuple[set[str], set[str]]:
+        """Check the status of any triggered tasks, returning any that have completed."""
+        completed_tasks = set()
+        error_tasks = set()
+
+        for task_id in self.triggered_block_task_ids:
+            try:
+                response = self._get(f"{self.datalab_api_url}/blocks/{task_id}/status")
+                task_status = response["status"]
+                if task_status == "ready":
+                    completed_tasks.add(task_id)
+            except DatalabAPIError:
+                error_tasks.add(task_id)
+
+        self.triggered_block_task_ids -= completed_tasks
+        self.triggered_block_task_ids -= error_tasks
+
+        return completed_tasks, error_tasks
+
     def _request(
-        self, method: str, url: str, expected_status: int = 200, **kwargs
+        self, method: str, url: str, expected_status: Union[int, list[int]] = 200, **kwargs
     ) -> dict[str, Any]:
         """Make an HTTP request with consistent error handling.
 
@@ -324,8 +360,11 @@ This is likely a server-side bug. Please report this issue to the datalab develo
         Returns:
             The JSON response data
         """
+        timeout = kwargs.pop("timeout", None)
         try:
-            response = self.session.request(method, url, follow_redirects=True, **kwargs)
+            response = self.session.request(
+                method, url, follow_redirects=True, timeout=timeout, **kwargs
+            )
             return self._handle_response(response, url, expected_status)
         except (httpx.RequestError, httpx.HTTPStatusError) as e:
             raise DatalabAPIError(f"Request failed for {url}: {e}")
@@ -334,7 +373,9 @@ This is likely a server-side bug. Please report this issue to the datalab develo
         """Make a GET request with error handling."""
         return self._request("GET", url, **kwargs)
 
-    def _post(self, url: str, expected_status: int = 200, **kwargs) -> dict[str, Any]:
+    def _post(
+        self, url: str, expected_status: Union[int, list[int]] = 200, **kwargs
+    ) -> dict[str, Any]:
         """Make a POST request with error handling."""
         return self._request("POST", url, expected_status, **kwargs)
 

{datalab_api-0.3.1 → datalab_api-0.4.0/src/datalab_api.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datalab-api
-Version: 0.3.1
+Version: 0.4.0
 Summary: A Python API for instances of the datalab data management platform (https://github.com/datalab-org/datalab).
 Author-email: Matthew Evans <matthew@ml-evs.science>
 License-Expression: MIT
@@ -13,6 +13,7 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
 Classifier: Intended Audience :: Science/Research
 Classifier: Intended Audience :: Information Technology
 Classifier: Topic :: Other/Nonlisted Topic
@@ -38,9 +39,6 @@ Requires-Dist: mkdocs-jupyter; extra == "docs"
 Provides-Extra: cli
 Requires-Dist: typer~=0.9; extra == "cli"
 Requires-Dist: click-shell~=2.1; extra == "cli"
-Provides-Extra: cheminventory-helper
-Requires-Dist: pandas~=2.2; extra == "cheminventory-helper"
-Requires-Dist: openpyxl~=3.0; extra == "cheminventory-helper"
 Provides-Extra: all
 Requires-Dist: datalab-api[cli,dev,docs]; extra == "all"
 Dynamic: license-file
@@ -61,7 +59,7 @@ A simple Python API that can interact with [*datalab*](https://github.com/datala
 
 The idea here is to provide a set of utility functions and models for manipulating samples, cells, inventory, files, users and metadata associated with *datalab* entries in an automated way.
 
-This API may not expose all the functionality available in a given *datalab* instance, and the idea would be that this package can support multiple versions of the underlying [*datalab* REST API](https://docs.datalab-org.io/en/stable/rest_api/).
+This API may not expose all the functionality available in a given *datalab* instance, and the idea would be that this package can support multiple versions of the underlying [*datalab* REST API](https://docs.datalab-org.io).
 This means that the API is primarily *functional* in nature, as opposed to object-oriented, since *datalab* instances are free to use their own custom data models.
 The available schemas are reported as instance metadata and in the future object-oriented models may be able to be genereated directly in the client (so e.g., the returned data would be Python objects like `Sample` rather than JSON data).
 

{datalab_api-0.3.1 → datalab_api-0.4.0}/src/datalab_api.egg-info/SOURCES.txt

@@ -4,7 +4,6 @@ pyproject.toml
 src/datalab_api/__init__.py
 src/datalab_api/_base.py
 src/datalab_api/cli.py
-src/datalab_api/helpers.py
 src/datalab_api/utils.py
 src/datalab_api.egg-info/PKG-INFO
 src/datalab_api.egg-info/SOURCES.txt

{datalab_api-0.3.1 → datalab_api-0.4.0}/src/datalab_api.egg-info/requires.txt

@@ -8,10 +8,6 @@ bokeh~=2.4
 [all]
 datalab-api[cli,dev,docs]
 
-[cheminventory-helper]
-pandas~=2.2
-openpyxl~=3.0
-
 [cli]
 typer~=0.9
 click-shell~=2.1

datalab_api-0.3.1/src/datalab_api/helpers.py

@@ -1,60 +0,0 @@
-import random
-import warnings
-from pathlib import Path
-
-from datalab_api import DatalabClient, DuplicateItemError
-
-
-def import_cheminventory(filename: Path, client: DatalabClient):
-    """Helper function to import a ChemInventory xlsx export into datalab, using the
-    'starting_materials' item type.
-
-    Migrated from the original version in the main datalab repository:
-        https://github.com/the-grey-group/datalab/blob/43764fb494c2cc1bf9f7dc90c25594aeb79d5767/pydatalab/tasks.py#L350-L413
-
-    """
-
-    warnings.warn(
-        DeprecationWarning(
-            "This helper is deprecated in favour of the full-fledged plugin at: https://github.com/datalab-industries/datalab-cheminventory-plugin"
-        )
-    )
-
-    def _generate_random_startingmaterial_id():
-        """Generate 'XX' + a random 15-length string for use as an id for starting materials
-        that don't have a barcode.
-        """
-        yield "".join(["XX"] + random.choices("abcdefghijklmnopqrstuvwxyz0123456789", k=15))
-
-    try:
-        import pandas as pd
-
-        inventory_df = pd.read_excel(filename)
-    except ImportError as exc:
-        raise ImportError(
-            "Please install pandas + openpyxl to use this helper function, via `pip install datalab-api[cheminventory-helper]`"
-        ) from exc
-    inventory_df["type"] = "starting_materials"
-    inventory_df["item_id"] = inventory_df["Barcode"]
-    # Fill missing barcodes
-    inventory_df["item_id"] = inventory_df["item_id"].fillna(
-        inventory_df["item_id"].apply(lambda _: next(_generate_random_startingmaterial_id()))
-    )
-    inventory_df["Molecular Weight"] = inventory_df["Molecular Weight"].replace(" ", float("nan"))
-
-    counts = {"success": 0, "duplicate": 0, "failed": 0}
-
-    for item in inventory_df.to_dict(orient="records"):
-        try:
-            client.create_item(
-                item["item_id"], item["type"], item, collection_id="jul24-inventory-import"
-            )
-            counts["success"] += 1
-        except DuplicateItemError:
-            counts["duplicate"] += 1
-        except Exception as exc:
-            counts["failed"] += 1
-            print(f"Failed to import item: {item}. Error: {exc}")
-            continue
-
-    print(f"Done: {counts=}")