Move to minimalkv

CHANGES.rst

@@ -2,6 +2,10 @@
 Changelog
 =========
 
+Kartothek 4.1.0 (2021-04-xx)
+============================
+
+* Switch from ``simplekv`` and ``storefact`` to their successor ``minimalkv`` as the library providing the store implementations.
 
 Kartothek 4.0.2 (2021-04-xx)
 ============================

asv_bench/benchmarks/index.py

@@ -9,10 +9,10 @@
 import uuid
 from functools import lru_cache
 
+import minimalkv
 import numpy as np
 import pandas as pd
 import pyarrow as pa
-import storefact
 
 from kartothek.core.index import ExplicitSecondaryIndex
 from kartothek.io_components.metapartition import MetaPartition
@@ -44,7 +44,7 @@ def setup(self, number_values, number_partitions, dtype):
             column=self.column_name, index_dct=index_dct, dtype=arrow_type
         )
         self.tmp_dir = tempfile.mkdtemp()
-        self.store = storefact.get_store_from_url("hfs://{}".format(self.tmp_dir))
+        self.store = minimalkv.get_store_from_url("hfs://{}".format(self.tmp_dir))
         self.dataset_uuid = "some_uuid"
         self.storage_key = self.ktk_index.store(self.store, self.dataset_uuid)
 

asv_bench/benchmarks/predicate_pushdown.py

@@ -1,4 +1,4 @@
-from storefact import get_store_from_url
+from minimalkv import get_store_from_url
 
 from kartothek.serialization import ParquetSerializer
 from kartothek.serialization.testing import get_dataframe_not_nested

asv_bench/benchmarks/write.py

@@ -4,7 +4,7 @@
 import tempfile
 import uuid
 
-from storefact import get_store_from_url
+from minimalkv import get_store_from_url
 
 from kartothek.core.common_metadata import make_meta
 from kartothek.core.testing import get_dataframe_alltypes

conda-requirements.txt

@@ -1,13 +1,12 @@
 dask[dataframe]
 decorator
+minimalkv>=1.2
 msgpack-python>=0.5.2
 # Currently dask and numpy==1.16.0 clash
 numpy!=1.15.0,!=1.16.0
 pandas>=0.23.0, !=1.0.0
 pyarrow>=0.17.1,!=1.0.0, <4
 simplejson
-simplekv
-storefact
 toolz
 typing_extensions # Some backports of the py3.8 typing module
 urlquote>=1.1.3

docs/conf.py

@@ -105,13 +105,11 @@
 
 intersphinx_mapping = {
     "pandas": ("https://pandas.pydata.org/docs/", None),
-    "simplekv": ("https://simplekv.readthedocs.io/en/stable/", None),
+    "minimalkv": ("https://minimalkv.readthedocs.io/en/stable/", None),
     "pyarrow": ("https://arrow.apache.org/docs/", None),
     "numpy": ("https://numpy.org/doc/stable/", None),
     "python": ("https://docs.python.org/3", None),
     "dask": ("https://docs.dask.org/en/stable/", None),
-    # Storefact isn't exposing any sphinx refs
-    # "storefact": ("https://storefact.readthedocs.io/en/stable", None),
 }
 
 # In particular type annotations are rendered as its full path to the class but

docs/environment-docs.yml

@@ -4,14 +4,13 @@ channels:
 dependencies:
   - dask[dataframe]
   - decorator
+  - minimalkv>=1.2
   - msgpack-python>=0.5.2
   # Currently dask and numpy==1.16.0 clash
   - numpy!=1.15.0,!=1.16.0
   - pandas>=0.23.0, !=1.0.0
   - pyarrow>=0.17.1,!=1.0.0, <4
   - simplejson
-  - simplekv
-  - storefact
   - toolz
   - typing_extensions # Some backports of the py3.8 typing module
   - urlquote>=1.1.3

docs/guide/cube/command_line_features.rst

@@ -17,7 +17,7 @@ Command Line Features
        </style>
 
 Kartothek Cube also features a command line interface (CLI) for some cube operations. To use it, create a ``skv.yml`` file that
-describes `storefact`_ stores:
+describes `minimalkv`_ stores:
 
 .. code-block:: yaml
 
@@ -147,5 +147,5 @@ Some information is not available when reading the schema information and requir
 
 Use ``kartothek_cube --help`` to get a list of all commands, or see :mod:`~kartothek.cli`.
 
-.. _storefact: https://github.com/blue-yonder/storefact
+.. _minimalkv: https://github.com/data-engineering-collective/minimalkv
 

docs/guide/cube/examples.rst

@@ -13,14 +13,14 @@ First, we want to create a cube for geodata:
 ...     partition_columns=["country"],
 ... )
 
-Apart from an abstract cube definition, we need a `simplekv`_-based storage backend:
+Apart from an abstract cube definition, we need a `minimalkv`_-based storage backend:
 
 >>> from functools import partial
 >>> import tempfile
->>> import storefact
+>>> import minimalkv
 >>> store_location = tempfile.mkdtemp()
 >>> store_factory = partial(
-...     storefact.get_store_from_url,
+...     minimalkv.get_store_from_url,
 ...     "hfs://" + store_location,
 ... )
 >>> store = store_factory()
@@ -424,4 +424,4 @@ geodata++time/table/_common_metadata
 .. _Dask: https://docs.dask.org/
 .. _Dask.Bag: https://docs.dask.org/en/latest/bag.html
 .. _Dask.DataFrame: https://docs.dask.org/en/latest/dataframe.html
-.. _simplekv: https://simplekv.readthedocs.io/
+.. _minimalkv: https://minimalkv.readthedocs.io/

docs/guide/cube/glossary.rst

@@ -55,8 +55,8 @@ Glossary
         Dataset that provides the groundtruth about which :term:`Cell` are in a :term:`Cube`.
 
     Store Factory
-        A callable that does not take any arguments and creates a new `simplekv`_ store when being called. Its type is
-        ``Callable[[], simplekv.KeyValueStore]``.
+        A callable that does not take any arguments and creates a new `minimalkv`_ store when being called. Its type is
+        ``Callable[[], minimalkv.KeyValueStore]``.
 
     Query
         A request for data from the cube, including things like "payload columns", "conditions", and more.
@@ -76,4 +76,4 @@ Glossary
 
 .. _Data Cubes: https://en.wikipedia.org/wiki/Data_cube
 .. _Parquet: https://parquet.apache.org/
-.. _simplekv: https://simplekv.readthedocs.io/
+.. _minimalkv: https://minimalkv.readthedocs.io/

docs/guide/getting_started.rst

@@ -47,24 +47,24 @@ We want to store this DataFrame now as a dataset. Therefore, we first need
 to connect to a storage location.
 
 We define a store factory as a callable which contains the storage information.
-We will use `storefact`_ in this example to construct such a store factory
+We will use `minimalkv`_ in this example to construct such a store factory
 for the local filesystem (``hfs://`` indicates we are using the local filesystem and
 what follows is the filepath).
 
 .. ipython:: python
 
     from functools import partial
     from tempfile import TemporaryDirectory
-    from storefact import get_store_from_url
+    from minimalkv import get_store_from_url
 
     dataset_dir = TemporaryDirectory()
 
     store_url = f"hfs://{dataset_dir.name}"
 
 .. admonition:: Storage locations
 
-    `storefact`_ offers support for several stores in Kartothek, these can be created using the
-    function `storefact.get_store_from_url` with one of the following prefixes:
+    `minimalkv`_ offers support for several stores in Kartothek, these can be created using the
+    function `minimalkv.get_store_from_url` with one of the following prefixes:
 
     - ``hfs``: Local filesystem
     - ``hazure``: AzureBlockBlobStorage
@@ -74,15 +74,11 @@ Interface
 ---------
 
 Kartothek can write to any location that
-fulfills the `simplekv.KeyValueStore interface
-<https://simplekv.readthedocs.io/en/latest/#simplekv.KeyValueStore>`_  as long as they
-support `ExtendedKeyspaceMixin
-<https://github.com/mbr/simplekv/search?q=%22class+ExtendedKeyspaceMixin%22&unscoped_q=%22class+ExtendedKeyspaceMixin%22>`_
+fulfills the `minimalkv.KeyValueStore interface
+<https://minimalkv.readthedocs.io/en/latest/#minimalkv.KeyValueStore>`_  as long as they
+support ``ExtendedKeyspaceMixin``
 (this is necessary so that ``/`` can be used in the storage key name).
 
-For more information, take a look out at the `storefact documentation
-<https://storefact.readthedocs.io/en/latest/reference/storefact.html>`_.
-
 
 Writing data to storage
 =======================
@@ -232,5 +228,5 @@ function but returns a collection of ``dask.delayed`` objects.
 
         read_table("a_unique_dataset_identifier", store_url, predicates=[[("A", "<", 2.5)]])
 
-.. _storefact: https://github.com/blue-yonder/storefact
+.. _minimalkv: https://github.com/data-engineering-collective/minimalkv
 .. _dask: https://docs.dask.org/en/latest/

docs/guide/mutating_datasets.rst

@@ -19,7 +19,7 @@ some data there with Kartothek.
     import pandas as pd
     from functools import partial
     from tempfile import TemporaryDirectory
-    from storefact import get_store_from_url
+    from minimalkv import get_store_from_url
 
     from kartothek.api.dataset import store_dataframes_as_dataset
 
@@ -236,7 +236,7 @@ When garbage collection is called, the files are removed.
 .. ipython:: python
 
     from kartothek.api.dataset import garbage_collect_dataset
-    from storefact import get_store_from_url
+    from minimalkv import get_store_from_url
 
     store = get_store_from_url(store_url)
 
@@ -246,7 +246,7 @@ When garbage collection is called, the files are removed.
 
     files_before.difference(store.keys())  # Show files removed
 
-.. _storefact: https://github.com/blue-yonder/storefact
+.. _minimalkv: https://github.com/data-engineering-collective/minimalkv
 
 
 Mutating indexed datasets

docs/guide/partitioning.rst

@@ -30,7 +30,7 @@ first and store the data there with Kartothek:
     import pandas as pd
     from functools import partial
     from tempfile import TemporaryDirectory
-    from storefact import get_store_from_url
+    from minimalkv import get_store_from_url
 
     from kartothek.api.dataset import store_dataframes_as_dataset
 

docs/spec/store_interface.rst

@@ -4,7 +4,7 @@
 KeyValueStore Interface
 =======================
 
-All storage interaction use ``simplekv.KeyValueStore`` as an storage layer
+All storage interaction use ``minimalkv.KeyValueStore`` as an storage layer
 abstraction. This allows convenient access to many different common Key-Value
 stores (ABS, S3, GCS, local filesystem, etc.) and allows an easy switch between
 the storage backends to facilitate a simpler test setup.
@@ -13,7 +13,7 @@ Generally, all of our public functions accepting a ``store`` argument accept a
 multitude of different input types and we generally accept all kinds of stores
 inheriting from ``KeyValueStore``, assuming they implement the pickle protocol.
 However, there are storages which simply cannot be distributed across processes
-or network nodes sensibly. A prime Example is the ``simplekv.memory.DictStore``
+or network nodes sensibly. A prime Example is the ``minimalkv.memory.DictStore``
 which uses a simple python dictionary as a backend store. It is technically
 possible to (de-)serialize the store but once it is deserialized in another
 process, or another node, the store looses its meaning since the stores are
@@ -25,9 +25,8 @@ protocol, or some more complex logic is required to initialize it, kartothek
 also accepts _factories_ which must be a callable returning a ``KeyValueStore``
 (see also ``kartothek.core.typing.StoreFactory``).
 
-For convenience we also offer a `storefact`_ integration and accept store urls
+For convenience we also offer an integration that accepts store urls
 which proves another easy level of access and is well suited for ad-hoc
 investigations.
 
-.. _simplekv: https://simplekv.readthedocs.io/
-.. _storefact: https://storefact.readthedocs.io/
+.. _minimalkv: https://minimalkv.readthedocs.io/

kartothek/cli/__init__.py

@@ -4,7 +4,7 @@
 .. important::
     This module does not contain any public APIs.
 
-Kartothek comes with a CLI tool named ``kartothek_cube``. To use it, create an YAML file that contains a dictionary of `storefact`_
+Kartothek comes with a CLI tool named ``kartothek_cube``. To use it, create an YAML file that contains a dictionary of `minimalkv`_
 stores (keys are names of the store and the values are dicts that contain the store config). ``Kartothek`` uses a `YAML`_
 file called ``skv.yml`` and a store called ``dataset`` by default, but you may pass ``--skv`` and ``--store`` to change
 these. An example file could look like:
@@ -30,7 +30,7 @@
 
 
 .. _Dask: https://docs.dask.org/
-.. _storefact: https://github.com/blue-yonder/storefact
+.. _minimalkv: https://github.com/data-engineering-collective/minimalkv
 .. _YAML: https://yaml.org/
 """
 import logging

kartothek/cli/_utils.py

@@ -2,7 +2,7 @@
 from functools import partial
 
 import click
-import storefact
+import minimalkv
 import yaml
 
 from kartothek.api.discover import discover_cube
@@ -18,7 +18,7 @@ def get_cube(store, uuid_prefix):
     ----------
     uuid_prefix: str
         Dataset UUID prefix.
-    store: Union[Callable[[], simplekv.KeyValueStore], simplekv.KeyValueStore]
+    store: Union[Callable[[], minimalkv.KeyValueStore], minimalkv.KeyValueStore]
         KV store.
 
     Returns
@@ -41,18 +41,18 @@ def get_cube(store, uuid_prefix):
 
 def get_store(skv, store):
     """
-    Get simplekv store from storefact config file.
+    Get minimalkv store from minimalkv config file.
 
     Parameters
     ----------
     skv: str
-        Name of the storefact yaml. Normally ``'skv.yml'``.
+        Name of the minimalkv yaml. Normally ``'skv.yml'``.
     store: str
         ID of the store.
 
     Returns
     -------
-    store_factory: Callable[[], simplekv.KeyValueStore]
+    store_factory: Callable[[], minimalkv.KeyValueStore]
         Store object.
 
     Raises
@@ -73,7 +73,7 @@ def get_store(skv, store):
             "Could not find store {store} in {skv}".format(store=store, skv=skv)
         )
 
-    return partial(storefact.get_store, **store_cfg[store])
+    return partial(minimalkv.get_store, **store_cfg[store])
 
 
 def _match_pattern(what, items, pattern):

kartothek/core/common_metadata.py

@@ -12,7 +12,7 @@
 import pyarrow as pa
 import pyarrow.parquet as pq
 import simplejson
-from simplekv import KeyValueStore
+from minimalkv import KeyValueStore
 
 from kartothek.core import naming
 from kartothek.core._compat import load_json

kartothek/core/docs.py

@@ -7,11 +7,11 @@
 
 _PARAMETER_MAPPING = {
     "store": """
-    store: Callable or str or simplekv.KeyValueStore
+    store: Callable or str or minimalkv.KeyValueStore
         The store where we can find or store the dataset.
 
-        Can be either ``simplekv.KeyValueStore``, a storefact store url or a
-        generic Callable producing a ``simplekv.KeyValueStore``""",
+        Can be either ``minimalkv.KeyValueStore``, a minimalkv store url or a
+        generic Callable producing a ``minimalkv.KeyValueStore``""",
     "overwrite": """
     overwrite: Optional[bool]
         If True, allow overwrite of an existing dataset.""",
@@ -70,12 +70,12 @@
         `merge_datasets__pipeline` key that contains the source dataset uuids for
         the merge.""",
     "output_store": """
-    output_store : Union[Callable, str, simplekv.KeyValueStore]
+    output_store : Union[Callable, str, minimalkv.KeyValueStore]
         If given, the resulting dataset is written to this store. By default
         the input store.
 
-        Can be either `simplekv.KeyValueStore`, a storefact store url or a
-        generic Callable producing a ``simplekv.KeyValueStore``""",
+        Can be either `minimalkv.KeyValueStore`, a minimalkv store url or a
+        generic Callable producing a ``minimalkv.KeyValueStore``""",
     "metadata": """
     metadata : Optional[Dict]
         A dictionary used to update the dataset metadata.""",

kartothek/core/factory.py

@@ -8,7 +8,7 @@
 from kartothek.core.utils import lazy_store
 
 if TYPE_CHECKING:
-    from simplekv import KeyValueStore
+    from minimalkv import KeyValueStore
 
 __all__ = ("DatasetFactory",)
 
@@ -38,7 +38,7 @@ def __init__(
         .. code::
 
             from functools import partial
-            from storefact import get_store_from_url
+            from minimalkv import get_store_from_url
             from kartothek.io.eager import read_table
 
             ds_factory = DatasetFactory(