Merge pull request #1091 from benjeffery/default-metadata

Make default metadata schema dependant

Author: mergify[bot]

.pre-commit-config.yaml

@@ -1,26 +1,26 @@
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v3.2.0
+    rev: v3.4.0
     hooks:
       - id: check-merge-conflict
       - id: debug-statements
       - id: mixed-line-ending
       - id: check-case-conflict
       - id: check-yaml
   - repo: https://github.com/benjeffery/pre-commit-clang-format
-    rev: master
+    rev: c21a74d089aaeb86c2c19df371c7e7bf40c07207
     hooks:
     - id: clang-format
       exclude: dev-tools|examples
       verbose: true
   - repo: https://github.com/asottile/reorder_python_imports
-    rev: v2.3.5
+    rev: v2.3.6
     hooks:
       - id: reorder-python-imports
         args: [--application-directories=python,
                --unclassifiable-application-module=_tskit]
   - repo: https://github.com/asottile/pyupgrade
-    rev: v2.7.2
+    rev: v2.7.4
     hooks:
       - id: pyupgrade
         args: [--py3-plus, --py36-plus]
@@ -33,13 +33,13 @@ repos:
         # loop that throws an error before black excludes it
         args: [python/tskit, python/setup.py, python/stress_lowlevel.py]
   - repo: https://gitlab.com/pycqa/flake8
-    rev: 3.8.3
+    rev: 3.8.4
     hooks:
       - id: flake8
         args: [--config=python/.flake8]
         additional_dependencies: ["flake8-bugbear==20.1.4", "flake8-builtins==1.5.2"]
   - repo: https://github.com/asottile/blacken-docs
-    rev: v1.8.0
+    rev: v1.9.1
     hooks:
       - id: blacken-docs
         args: [--skip-errors]

python/CHANGELOG.rst

@@ -11,6 +11,11 @@
   the shortest possible branch lengths instead of a fixed epsilon of
   1e-10. (:user:`jeromekelleher`, :issue:`1089`, :pr:`1090`)
 
+- Default value metadata in ``add_row`` functions is now schema-dependant, so that
+  ``metadata={}`` is no longer needed as an argument when a schema is present
+  (:user:`benjeffery`, :issue:`1084`).
+
+
 --------------------
 [0.3.4] - 2020-12-02
 --------------------

python/tests/test_tables.py

@@ -760,6 +760,21 @@ def test_default_metadata_schema(self):
                 }
             )
 
+    def test_default_metadata_add_row(self):
+        row_data = self.input_data_for_add_row()
+        del row_data["metadata"]
+
+        table = self.table_class()
+        table.add_row(**row_data)
+        assert table[0].metadata == b""
+        assert table[0].metadata == table.metadata_schema.empty_value
+
+        table = self.table_class()
+        table.metadata_schema = tskit.MetadataSchema({"codec": "json"})
+        table.add_row(**row_data)
+        assert table[0].metadata == {}
+        assert table[0].metadata == table.metadata_schema.empty_value
+
     def test_row_round_trip_metadata_schema(self):
         data = self.metadata_example_data()
         table = self.table_class()

python/tskit/metadata.py

@@ -510,6 +510,7 @@ def __init__(self, schema: Optional[Mapping[str, Any]]) -> None:
             self._validate_row = validate_bytes
             self.encode_row = NOOPCodec({}).encode
             self.decode_row = NOOPCodec({}).decode
+            self.empty_value = b""
         else:
             try:
                 TSKITMetadataSchemaValidator.check_schema(schema)
@@ -530,6 +531,7 @@ def __init__(self, schema: Optional[Mapping[str, Any]]) -> None:
             self._validate_row = TSKITMetadataSchemaValidator(schema).validate
             self.encode_row = codec_instance.encode
             self.decode_row = codec_instance.decode
+            self.empty_value = {}
 
     def __repr__(self) -> str:
         return self._string

python/tskit/tables.py

@@ -510,9 +510,13 @@ def add_row(self, flags=0, location=None, metadata=None):
             array describing the location of this individual. If not specified
             or None, a zero-dimensional location is stored.
         :param object metadata: Any object that is valid metadata for the table's schema.
+            Defaults to the default metadata value for the table's schema. This is
+            typically ``{}``. For no schema, ``None``.
         :return: The ID of the newly added node.
         :rtype: int
         """
+        if metadata is None:
+            metadata = self.metadata_schema.empty_value
         metadata = self.metadata_schema.validate_and_encode_row(metadata)
         return self.ll_table.add_row(flags=flags, location=location, metadata=metadata)
 
@@ -718,9 +722,13 @@ def add_row(self, flags=0, time=0, population=-1, individual=-1, metadata=None):
         :param int individual: The ID of the individual in which the new node was born.
             Defaults to :data:`tskit.NULL`.
         :param object metadata: Any object that is valid metadata for the table's schema.
+            Defaults to the default metadata value for the table's schema. This is
+            typically ``{}``. For no schema, ``None``.
         :return: The ID of the newly added node.
         :rtype: int
         """
+        if metadata is None:
+            metadata = self.metadata_schema.empty_value
         metadata = self.metadata_schema.validate_and_encode_row(metadata)
         return self.ll_table.add_row(flags, time, population, individual, metadata)
 
@@ -913,9 +921,13 @@ def add_row(self, left, right, parent, child, metadata=None):
         :param int parent: The ID of parent node.
         :param int child: The ID of child node.
         :param object metadata: Any object that is valid metadata for the table's schema.
+            Defaults to the default metadata value for the table's schema. This is
+            typically ``{}``. For no schema, ``None``.
         :return: The ID of the newly added edge.
         :rtype: int
         """
+        if metadata is None:
+            metadata = self.metadata_schema.empty_value
         metadata = self.metadata_schema.validate_and_encode_row(metadata)
         return self.ll_table.add_row(left, right, parent, child, metadata)
 
@@ -1127,9 +1139,13 @@ def add_row(self, left, right, node, source, dest, time, metadata=None):
         :param int dest: The ID of the destination population.
         :param float time: The time of the migration event.
         :param object metadata: Any object that is valid metadata for the table's schema.
+            Defaults to the default metadata value for the table's schema. This is
+            typically ``{}``. For no schema, ``None``.
         :return: The ID of the newly added migration.
         :rtype: int
         """
+        if metadata is None:
+            metadata = self.metadata_schema.empty_value
         metadata = self.metadata_schema.validate_and_encode_row(metadata)
         return self.ll_table.add_row(left, right, node, source, dest, time, metadata)
 
@@ -1334,9 +1350,13 @@ def add_row(self, position, ancestral_state, metadata=None):
         :param float position: The position of this site in genome coordinates.
         :param str ancestral_state: The state of this site at the root of the tree.
         :param object metadata: Any object that is valid metadata for the table's schema.
+            Defaults to the default metadata value for the table's schema. This is
+            typically ``{}``. For no schema, ``None``.
         :return: The ID of the newly added site.
         :rtype: int
         """
+        if metadata is None:
+            metadata = self.metadata_schema.empty_value
         metadata = self.metadata_schema.validate_and_encode_row(metadata)
         return self.ll_table.add_row(position, ancestral_state, metadata)
 
@@ -1560,11 +1580,15 @@ def add_row(self, site, node, derived_state, parent=-1, metadata=None, time=None
         :param int parent: The ID of the parent mutation. If not specified,
             defaults to :attr:`NULL`.
         :param object metadata: Any object that is valid metadata for the table's schema.
+            Defaults to the default metadata value for the table's schema. This is
+            typically ``{}``. For no schema, ``None``.
         :return: The ID of the newly added mutation.
         :param float time: The occurrence time for the new mutation. If not specified,
             defaults to ``UNKNOWN_TIME``, indicating the time is unknown.
         :rtype: int
         """
+        if metadata is None:
+            metadata = self.metadata_schema.empty_value
         metadata = self.metadata_schema.validate_and_encode_row(metadata)
         return self.ll_table.add_row(
             site,
@@ -1763,9 +1787,13 @@ def add_row(self, metadata=None):
         :attr:`metadata_schema<tskit.PopulationTable.metadata_schema>`.
 
         :param object metadata: Any object that is valid metadata for the table's schema.
+            Defaults to the default metadata value for the table's schema. This is
+            typically ``{}``. For no schema, ``None``.
         :return: The ID of the newly added population.
         :rtype: int
         """
+        if metadata is None:
+            metadata = self.metadata_schema.empty_value
         metadata = self.metadata_schema.validate_and_encode_row(metadata)
         return self.ll_table.add_row(metadata=metadata)