Dataset "schema" v0.3

docs/api.rst

@@ -50,6 +50,48 @@ Methods
    Tajimas_D
    pc_relate
 
+Variables
+=========
+
+.. autosummary::
+   :toctree: generated/
+
+    variables.base_prediction_spec
+    variables.call_allele_count_spec
+    variables.call_dosage_spec
+    variables.call_dosage_mask_spec
+    variables.call_genotype_spec
+    variables.call_genotype_mask_spec
+    variables.call_genotype_phased_spec
+    variables.call_genotype_probability_spec
+    variables.call_genotype_probability_mask_spec
+    variables.covariates_spec
+    variables.dosage_spec
+    variables.genotype_counts_spec
+    variables.loco_prediction_spec
+    variables.meta_prediction_spec
+    variables.pc_relate_phi_spec
+    variables.sample_id_spec
+    variables.sample_pcs_spec
+    variables.traits_spec
+    variables.variant_allele_spec
+    variables.variant_allele_count_spec
+    variables.variant_allele_frequency_spec
+    variables.variant_allele_total_spec
+    variables.variant_beta_spec
+    variables.variant_call_rate_spec
+    variables.variant_contig_spec
+    variables.variant_hwe_p_value_spec
+    variables.variant_id_spec
+    variables.variant_n_called_spec
+    variables.variant_n_het_spec
+    variables.variant_n_hom_alt_spec
+    variables.variant_n_hom_ref_spec
+    variables.variant_n_non_ref_spec
+    variables.variant_p_value_spec
+    variables.variant_position_spec
+    variables.variant_t_value_spec
+
 Utilities
 =========
 

docs/conf.py

@@ -6,6 +6,8 @@
 
 # -- Path setup --------------------------------------------------------------
 
+import logging as pylogging
+
 # If extensions (or modules to document with autodoc) are in another directory,
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
@@ -15,6 +17,7 @@
 from pathlib import Path
 
 import xarray
+from sphinx.util import logging
 
 sys.path.insert(0, os.path.abspath(".."))
 
@@ -52,6 +55,26 @@
     *[p.stem for p in (HERE / "extensions").glob("*.py")],
 ]
 
+
+# Workaround https://github.com/agronholm/sphinx-autodoc-typehints/issues/123
+# When this https://github.com/agronholm/sphinx-autodoc-typehints/pull/153
+# gets merged, we can remove this
+class FilterForIssue123(pylogging.Filter):
+    def filter(self, record: pylogging.LogRecord) -> bool:
+        msg = record.getMessage()
+        return not (
+            msg.startswith("Cannot treat a function")
+            and any(
+                s in msg
+                for s in ["sgkit.variables.Spec", "sgkit.variables.ArrayLikeSpec"]
+            )
+        )
+
+
+logging.getLogger("sphinx_autodoc_typehints").logger.addFilter(FilterForIssue123())
+# End of workaround
+
+
 # Add any paths that contain templates here, relative to this directory.
 templates_path = ["_templates"]
 

docs/extensions/typed_returns.py

@@ -7,8 +7,8 @@
 import re
 from typing import Iterator, List
 
-from sphinx.application import Sphinx  # type: ignore
-from sphinx.ext.napoleon import NumpyDocstring  # type: ignore
+from sphinx.application import Sphinx
+from sphinx.ext.napoleon import NumpyDocstring
 
 
 def process_return(lines: List[str]) -> Iterator[str]:

setup.cfg

@@ -111,6 +111,8 @@ ignore_missing_imports = True
 ignore_missing_imports = True
 [mypy-bed_reader.*]
 ignore_missing_imports = True
+[mypy-sphinx.*]
+ignore_missing_imports = True
 [mypy-sgkit.*]
 allow_redefinition = True
 [mypy-sgkit.*.tests.*]

sgkit/__init__.py

@@ -37,4 +37,5 @@
     "Tajimas_D",
     "pc_relate",
     "simulate_genotype_call_dataset",
+    "variables",
 ]

sgkit/model.py

@@ -3,8 +3,8 @@
 import numpy as np
 import xarray as xr
 
+from . import variables
 from .typing import ArrayLike
-from .utils import check_array_like
 
 DIM_VARIANT = "variants"
 DIM_SAMPLE = "samples"
@@ -57,11 +57,6 @@ def create_genotype_call_dataset(
     -------
     The dataset of genotype calls.
     """
-    check_array_like(variant_contig, kind="i", ndim=1)
-    check_array_like(variant_position, kind="i", ndim=1)
-    check_array_like(variant_alleles, kind={"S", "O"}, ndim=2)
-    check_array_like(sample_id, kind={"U", "O"}, ndim=1)
-    check_array_like(call_genotype, kind="i", ndim=3)
     data_vars: Dict[Hashable, Any] = {
         "variant_contig": ([DIM_VARIANT], variant_contig),
         "variant_position": ([DIM_VARIANT], variant_position),
@@ -74,16 +69,14 @@ def create_genotype_call_dataset(
         ),
     }
     if call_genotype_phased is not None:
-        check_array_like(call_genotype_phased, kind="b", ndim=2)
         data_vars["call_genotype_phased"] = (
             [DIM_VARIANT, DIM_SAMPLE],
             call_genotype_phased,
         )
     if variant_id is not None:
-        check_array_like(variant_id, kind={"U", "O"}, ndim=1)
         data_vars["variant_id"] = ([DIM_VARIANT], variant_id)
     attrs: Dict[Hashable, Any] = {"contigs": variant_contig_names}
-    return xr.Dataset(data_vars=data_vars, attrs=attrs)
+    return variables.validate(xr.Dataset(data_vars=data_vars, attrs=attrs))
 
 
 def create_genotype_dosage_dataset(
@@ -132,12 +125,6 @@ def create_genotype_dosage_dataset(
     The dataset of genotype calls.
 
     """
-    check_array_like(variant_contig, kind="i", ndim=1)
-    check_array_like(variant_position, kind="i", ndim=1)
-    check_array_like(variant_alleles, kind={"S", "O"}, ndim=2)
-    check_array_like(sample_id, kind={"U", "O"}, ndim=1)
-    check_array_like(call_dosage, kind="f", ndim=2)
-    check_array_like(call_genotype_probability, kind="f", ndim=3)
     data_vars: Dict[Hashable, Any] = {
         "variant_contig": ([DIM_VARIANT], variant_contig),
         "variant_position": ([DIM_VARIANT], variant_position),
@@ -155,7 +142,6 @@ def create_genotype_dosage_dataset(
         ),
     }
     if variant_id is not None:
-        check_array_like(variant_id, kind={"U", "O"}, ndim=1)
         data_vars["variant_id"] = ([DIM_VARIANT], variant_id)
     attrs: Dict[Hashable, Any] = {"contigs": variant_contig_names}
-    return xr.Dataset(data_vars=data_vars, attrs=attrs)
+    return variables.validate(xr.Dataset(data_vars=data_vars, attrs=attrs))