remove deprecated models (#171)

CHANGES.rst

@@ -1,3 +1,12 @@
+2.0.0 (unreleased)
+===================
+
+- Remove deprecated jwst.datamodels models: DataModel, DrizProductModel,
+  MIRIRampModel, MultiProductModel [#171]
+
+- Increase CRDS minimum version to 11.17.1 [#171]
+
+
 1.10.1 (2024-03-25)
 ===================
 
@@ -298,7 +307,6 @@ Other
 
 - Close for opened files [#169]
 
-
 1.5.0 (2023-05-16)
 ==================
 

docs/source/jwst/datamodels/attributes.rst

@@ -51,7 +51,7 @@ The current models are as follows:
     'NisImgPhotomModel', 'NisSossPhotomModel', 'NisWfssPhotomModel',
     'NrsFsPhotomModel', 'NrsMosPhotomModel',
     'PsfMaskModel',
-    'QuadModel', 'RampModel', 'MIRIRampModel',
+    'QuadModel', 'RampModel',
     'RampFitOutputModel', 'ReadnoiseModel',
     'ReferenceFileModel', 'ReferenceCubeModel', 'ReferenceImageModel', 'ReferenceQuadModel',
     'RegionsModel', 'ResetModel',
@@ -105,9 +105,9 @@ is stored in a separate image extension).  Full-frame data for all other
 instruments have four columns or rows of reference pixels on each edge
 of the image.
 
-DataModel Base Class
+JwstDataModel Base Class
 --------------------
 
-.. autoclass:: jwst.datamodels.DataModel
+.. autoclass:: jwst.datamodels.JwstDataModel
    :members:
    :noindex:

docs/source/jwst/datamodels/models.rst

@@ -239,7 +239,7 @@ use::
 In place of `ImageModel`, use the type of data one expects to find in
 the file.  For example, if spectrographic data is expected, use
 `SpecModel`.  If it doesn't matter (perhaps the application is only
-sorting FITS files into categories) use the base class `DataModel`.
+sorting FITS files into categories) use the base class `JwstDataModel`.
 
 An alternative is to use::
 
@@ -276,9 +276,9 @@ use the :ref:`metadata` tree.
 There is a convenience method, `find_fits_keyword` to find where a
 FITS keyword is used in the metadata tree::
 
-    >>> from stdatamodels.jwst.datamodels import DataModel
+    >>> from stdatamodels.jwst.datamodels import JwstDataModel
     >>> # First, create a model of the desired type
-    >>> model = DataModel()
+    >>> model = JwstDataModel()
     >>> model.find_fits_keyword('DATE-OBS')
     [u'meta.observation.date']
 
@@ -322,20 +322,20 @@ Environment Variables
 There are a number of environment variables that affect how models are read.
 
 PASS_INVALID_VALUES
-  Used by `~jwst.datamodels.DataModel` when instantiating
+  Used by `~jwst.datamodels.JwstDataModel` when instantiating
   a model from a file. If ``True``, values that do not validate the schema will
   still be added to the metadata. If ``False``, they will be set to ``None``.
   Default is ``False``.
 
 STRICT_VALIDATION
-  Used by `~jwst.datamodels.DataModel` when instantiating a model from a file.
+  Used by `~jwst.datamodels.JwstDataModel` when instantiating a model from a file.
   If ``True``, schema validation errors will generate an exception.
   If ``False``, they will generate a warning.
   Default is ``False``.
 
 SKIP_FITS_UPDATE
   DEPRECATED: In the future the FITS header will always be used.
-  Used by `~jwst.datamodels.DataModel` when instantiating a
+  Used by `~jwst.datamodels.JwstDataModel` when instantiating a
   model from a FITS file. When ``False``, models opened from FITS files will
   proceed and load the FITS header values into the model. When ``True`` and the
   FITS file has an ASDF extension, the loading/validation of the FITS header

docs/source/jwst/datamodels/new_model.rst

@@ -189,16 +189,16 @@ The model class
 Now, let's see how this schema is tied in with a new Python class for
 the model.
 
-First, we need to import the `DataModel` class, which is the base
+First, we need to import the `JwstDataModel` class, which is the base
 class for all models::
 
-  from stdatamodels.jwst.datamodels import DataModel
+  from stdatamodels.jwst.datamodels import JwstDataModel
 
-Then we create a new Python class that inherits from `DataModel`, and
+Then we create a new Python class that inherits from `JwstDataModel`, and
 set its `schema_url` class member to point to the schema that we just
 defined above::
 
-  class MiriBadPixelMaskModel(DataModel):
+  class MiriBadPixelMaskModel(JwstDataModel):
       schema_url = "bad_pixel_mask.schema.yaml"
 
 Here, the `schema_url` has all of the "magical" URL abilities
@@ -234,7 +234,7 @@ elements, by setting `schema_url` to::
 Within this class, we'll define a constructor.  All model constructors
 must take the highly polymorphic ``init`` value as the first argument.
 This can be a file, another model, or all kinds of other things.  See
-the docstring of `jwst.datamodels.DataModel.__init__` for more
+the docstring of `jwst.datamodels.JwstDataModel.__init__` for more
 information.  But we're going to let the base class handle that
 anyway.
 
@@ -252,7 +252,7 @@ technically writing a new constructor for the model is optional:
         Parameters
         ----------
         init : any
-            Any of the initializers supported by `~jwst.datamodels.DataModel`.
+            Any of the initializers supported by `~jwst.datamodels.JwstDataModel`.
 
         dq : numpy array
             The data quality array.
@@ -336,7 +336,7 @@ function.
 Lastly, remember the ``meta.bad_pixel_count`` element we defined
 above?  We need some way to make sure that whenever the file is
 written out that it has the correct value.  The model may have been
-loaded and modified.  For this, `DataModel` has the `on_save` method
+loaded and modified.  For this, `JwstDataModel` has the `on_save` method
 hook, which may be overridden by the subclass to add anything that
 should happen just before saving:
 

docs/source/jwst/datamodels/structure.rst

@@ -23,7 +23,7 @@ attribute is not defined, which in turn calls _make_default_array if
 the schema says the attribute is an array. All these methods can be
 found in properties.py.
 
-The base class for Datamodels isDataModel,  in model_base.py. It takes
+The base class for Datamodels is JwstDataModel,  in model_base.py. It takes
 several arguments, the most important of which is init, which as the
 name suggests, specifies how to initialize the primary data array of
 the model. Init is most usually the name of a file, but can be an

pyproject.toml

@@ -49,7 +49,7 @@ test = [
     "psutil",
     "pytest>=4.6.0",
     "pytest-doctestplus",
-    "crds>=11.16.14",
+    "crds>=11.17.1",
     "scipy>=1.5",
 ]
 docs = [

src/stdatamodels/jwst/datamodels/__init__.py

@@ -1,4 +1,4 @@
-from .model_base import JwstDataModel, DataModel
+from .model_base import JwstDataModel
 from .abvega_offset import ABVegaOffsetModel
 from .amilg import AmiLgModel
 from .amilgfitmodel import AmiLgFitModel
@@ -16,7 +16,6 @@
 from .dark import DarkModel
 from .darkMIRI import DarkMIRIModel
 from .drizpars import DrizParsModel
-from .drizproduct import DrizProductModel
 from .emi import EmiModel
 from .extract1dimage import Extract1dImageModel
 from .extract1d_spec import Extract1dIFUModel
@@ -37,12 +36,10 @@
 from .level1b import Level1bModel
 from .linearity import LinearityModel
 from .mask import MaskModel
-from .ramp import MIRIRampModel
 from .mrsxartcorr import MirMrsXArtCorrModel
 from .multicombinedspec import MultiCombinedSpecModel
 from .multiexposure import MultiExposureModel
 from .multiextract1d import MultiExtract1dImageModel
-from .multiprod import MultiProductModel
 from .multislit import MultiSlitModel
 from .multispec import MultiSpecModel
 from .nirspec_flat import NirspecFlatModel, NirspecQuadFlatModel
@@ -94,7 +91,7 @@
 
 __all__ = [
     'open',
-    'DataModel', 'JwstDataModel',
+    'JwstDataModel',
     'ABVegaOffsetModel',
     'AmiLgModel',
     'AmiLgFitModel',
@@ -109,7 +106,6 @@
     'DarkModel', 'DarkMIRIModel',
     'DisperserModel', 'DistortionModel', 'DistortionMRSModel',
     'DrizParsModel',
-    'DrizProductModel',
     'EmiModel',
     'Extract1dImageModel',
     'Extract1dIFUModel',
@@ -126,7 +122,6 @@
     'LinearityModel', 'MaskModel', 'MSAModel',
     'MultiCombinedSpecModel', 'MultiExposureModel',
     'MultiExtract1dImageModel', 'MultiSlitModel',
-    'MultiProductModel',
     'MultiSpecModel',
     'NIRCAMGrismModel', 'NIRISSGrismModel',
     'OTEModel',
@@ -140,7 +135,7 @@
     'NisImgPhotomModel', 'NisSossPhotomModel', 'NisWfssPhotomModel',
     'NrsFsPhotomModel', 'NrsMosPhotomModel',
     'PsfMaskModel',
-    'QuadModel', 'RampModel', 'MIRIRampModel',
+    'QuadModel', 'RampModel',
     'RampFitOutputModel', 'ReadnoiseModel',
     'ReferenceFileModel', 'ReferenceCubeModel', 'ReferenceImageModel', 'ReferenceQuadModel',
     'RegionsModel', 'ResetModel',

src/stdatamodels/jwst/datamodels/model_base.py

@@ -1,10 +1,9 @@
-import warnings
 from astropy.time import Time
 
 from stdatamodels import DataModel as _DataModel
 
 
-__all__ = ["DataModel", "JwstDataModel"]
+__all__ = ["JwstDataModel"]
 
 
 class JwstDataModel(_DataModel):
@@ -53,8 +52,3 @@ def on_save(self, init):
         super().on_save(init)
 
         self.meta.date = Time.now().isot
-
-
-class DataModel(JwstDataModel):
-    def __init_subclass__(cls):
-        warnings.warn("Class DataModel has been renamed JwstDataModel. It will be removed in v 1.5.", DeprecationWarning, 2)

src/stdatamodels/jwst/datamodels/ramp.py

@@ -1,9 +1,7 @@
-import warnings
-
 from .model_base import JwstDataModel
 
 
-__all__ = ['RampModel', 'MIRIRampModel']
+__all__ = ['RampModel']
 
 
 class RampModel(JwstDataModel):
@@ -43,16 +41,3 @@ def __init__(self, init=None, **kwargs):
         self.pixeldq = self.pixeldq
         self.groupdq = self.groupdq
         self.err = self.err
-
-
-class MIRIRampModel(RampModel):
-    """A data model for 4D MIRI ramps.
-
-    This model has been deprecated. Please use `RampModel` instead.
-    """
-    def __init__(self, *args, **kwargs):
-        warnings.warn(
-            '"MIRIRampModel" is deprecated and will be removed. Use RampModel',
-            category=DeprecationWarning,
-        )
-        return super(MIRIRampModel, self).__init__(*args, **kwargs)

src/stdatamodels/jwst/datamodels/tests/test_models.py

@@ -13,8 +13,7 @@
 import pytest
 
 from stdatamodels.jwst.datamodels import (JwstDataModel, ImageModel, MaskModel, AsnModel,
-                                          MultiSlitModel, SlitModel, DataModel,
-                                          DrizProductModel, MultiProductModel, MIRIRampModel,
+                                          MultiSlitModel, SlitModel,
                                           NirspecFlatModel, NirspecQuadFlatModel,
                                           SlitDataModel, IFUImageModel, ABVegaOffsetModel)
 from stdatamodels.jwst import datamodels
@@ -322,11 +321,7 @@ def test_all_datamodels_init(model):
     """
     Test that all current datamodels can be initialized.
     """
-    if model in (DrizProductModel, MultiProductModel, MIRIRampModel):
-        with pytest.warns(DeprecationWarning):
-            model()
-    else:
-        model()
+    model()
 
 
 def test_meta_date_management(tmp_path):
@@ -555,15 +550,6 @@ def test_dq_def_roundtrip(tmp_path):
     assert total_diff == 0
 
 
-def test_deprecation_data_model():
-    """ Tests that inheriting from stdatamodels.jwst.datamodels.DataModel
-    raises a DeprecationWarning."""
-
-    with pytest.deprecated_call():
-        class Dummy(DataModel):
-            pass
-
-
 @pytest.mark.parametrize("shape", [None, 10])
 @pytest.mark.parametrize("model", [NirspecFlatModel, NirspecQuadFlatModel])
 def test_nirspec_flat_table_migration(tmp_path, model, shape):

src/stdatamodels/jwst/datamodels/tests/test_open.py

@@ -2,7 +2,6 @@
 Test datamodel.open
 """
 
-import contextlib
 import os
 import os.path
 from pathlib import Path, PurePath
@@ -44,22 +43,6 @@ def test_guess(guess):
                     pass
 
 
-def test_mirirampmodel_deprecation(tmp_path):
-    """Test that a deprecated MIRIRampModel can be opened"""
-    path = str(tmp_path / "ramp.fits")
-    # Create a MIRIRampModel, working around the deprecation.
-    model = datamodels.RampModel((1, 1, 10, 10))
-    model.save(path)
-    hduls = fits.open(path, mode='update')
-    hduls[0].header['datamodl'] = 'MIRIRampModel'
-    hduls.close()
-
-    # Test it.
-    with pytest.warns(DeprecationWarning):
-        miri_ramp = datamodels.open(path)
-    assert isinstance(miri_ramp, datamodels.RampModel)
-
-
 @pytest.fixture
 def mock_get_available_memory(monkeypatch):
     def mock(include_swap=True):
@@ -255,10 +238,12 @@ def test_open_asdf_no_datamodel_class(tmp_path, suffix):
     model = DataModel()
     model.save(path)
 
-    # Note: only the fits open emits a "model_type not found" warning.  Both
-    # fits and asdf should behave the same
-    ctx = pytest.warns(util.NoTypeWarning) if suffix == 'fits' else contextlib.nullcontext()
-    with ctx:
+    # previously, only the fits file issued a NoTypeWarning
+    # this was a quirk of the deprecated (and now removed) jwst.DataModel
+    # sharing the same class name as stdatamodels.DataModel (used here)
+    # now that jwst.DataModel is removed, both the fits and asdf
+    # files correctly report NoTypeWarning
+    with pytest.warns(util.NoTypeWarning):
         with datamodels.open(path) as m:
             assert isinstance(m, DataModel)