Merge pull request #672 from opencobra/chore-upgrade-goodtables

upgrade goodtables

setup.cfg

@@ -63,7 +63,7 @@ install_requires =
 	importlib_resources
 	numpydoc
 	pylru
-	goodtables ==1.0.0
+	goodtables ~=2.0
 	depinfo
 	requests
 	equilibrator_api <0.2;python_version>='3.5'

src/memote/experimental/checks.py

@@ -19,113 +19,64 @@
 
 from __future__ import absolute_import
 
-from functools import partial
+from goodtables import check, Error
 
-from goodtables import check
 
-
-def check_partial(func, *args, **kwargs):
-    """Create a partial to be used by goodtables."""
-    new_func = partial(func, *args, **kwargs)
-    new_func.check = func.check
-    return new_func
-
-
-@check('gene-not-in-model', type='custom', context='body')
-def gene_id_check(genes, errors, columns, row_number):
-    """
-    Validate gene identifiers against a known set.
-
-    Parameters
-    ----------
-    genes : set
-        The known set of gene identifiers.
-    errors :
-        Passed by goodtables.
-    columns :
-        Passed by goodtables.
-    row_number :
-        Passed by goodtables.
-
-    """
-    message = ("Gene '{value}' in column {col} and row {row} does not "
-               "appear in the metabolic model.")
-    for column in columns:
-        if "gene" in column['header'] and column['value'] not in genes:
-            message = message.format(
-                value=column['value'],
-                row=row_number,
-                col=column['number'])
-            errors.append({
-                'code': 'bad-value',
-                'message': message,
-                'row-number': row_number,
-                'column-number': column['number'],
-            })
-
-
-@check('reaction-not-in-model', type='custom', context='body')
-def reaction_id_check(reactions, errors, columns, row_number):
+@check('unknown-identifier', type='custom', context='body')
+class UnknownIdentifier:
     """
-    Validate reactions identifiers against a known set.
+    Validate data identifiers against a known set.
 
-    Parameters
+    Attributes
     ----------
-    reactions : set
-        The known set of reaction identifiers.
-    errors :
-        Passed by goodtables.
-    columns :
-        Passed by goodtables.
-    row_number :
-        Passed by goodtables.
-
-    """
-    message = ("Reaction '{value}' in column {col} and row {row} does not "
-               "appear in the metabolic model.")
-    for column in columns:
-        if "reaction" in column['header'] and column['value'] not in reactions:
-            message = message.format(
-                value=column['value'],
-                row=row_number,
-                col=column['number'])
-            errors.append({
-                'code': 'bad-value',
-                'message': message,
-                'row-number': row_number,
-                'column-number': column['number'],
-            })
+    column : str
+        The header of the data column to check.
+    identifiers : iterable of str
+        The known set of identifiers.
 
-
-@check('metabolite-not-in-model', type='custom', context='body')
-def metabolite_id_check(metabolites, errors, columns, row_number):
     """
-    Validate metabolite identifiers against a known set.
-
-    Parameters
-    ----------
-    metabolites : set
-        The known set of metabolite identifiers.
-    errors :
-        Passed by goodtables.
-    columns :
-        Passed by goodtables.
-    row_number :
-        Passed by goodtables.
 
-    """
-    message = ("Metabolite '{value}' in column {col} and row {row} does not "
-               "appear in the metabolic model.")
-    for column in columns:
-        if "metabolite" in column['header'] and \
-                column['value'] not in metabolites:
-            message = message.format(
-                value=column['value'],
-                row=row_number,
-                col=column['number'])
-            errors.append({
-                'code': 'bad-value',
-                'message': message,
-                'row-number': row_number,
-                'column-number': column['number'],
-            })
+    def __init__(self, column, identifiers, **_):
+        """
+        Initialize the custom identfier check.
+
+        Parameters
+        ----------
+        column : str
+            The header of the data column to check.
+        identifiers : iterable of str
+            The known set of identifiers.
+
+        """
+        self.column = column
+        self.identifiers = frozenset(identifiers)
+
+    def check_row(self, cells):
+        """Check each row in the data table."""
+        cell = None
+        for item in cells:
+            if item['header'] == self.column:
+                cell = item
+                break
+
+        if cell is None:
+            error = Error(
+                'unknown-identifier',
+                row_number=cells[0]['row-number'],
+                message="Checking identifiers requires the column "
+                        "'{column}' to exist.".format(column=self.column)
+            )
+            return [error]
+
+        value = cell.get('value')
+        if value not in self.identifiers:
+            error = Error(
+                'unknown-identifier',
+                cell,
+                message="Value '{value}' in column {header} on row "
+                        "{row_number} is an unknown identifier.",
+                message_substitutions={
+                    'value': value,
+                }
+            )
+            return [error]

src/memote/experimental/essentiality.py

@@ -24,7 +24,6 @@
 from cobra.flux_analysis import single_gene_deletion
 
 from memote.experimental.experiment import Experiment
-from memote.experimental.checks import check_partial, gene_id_check
 
 
 __all__ = ("EssentialityExperiment",)
@@ -70,7 +69,12 @@ def load(self, dtype_conversion=None):
     def validate(self, model, checks=[]):
         """Use a defined schema to validate the medium table format."""
         custom = [
-            check_partial(gene_id_check, frozenset(g.id for g in model.genes))
+            {
+                "unknown-identifier": {
+                    "column": "gene",
+                    "identifiers": {g.id for g in model.genes}
+                }
+            }
         ]
         super(EssentialityExperiment, self).validate(
             model=model, checks=checks + custom)

src/memote/experimental/experimental_base.py

@@ -27,6 +27,9 @@
 
 import memote.experimental.schemata
 from memote.experimental.tabular import read_tabular
+# The following import is necessary in order to register the custom check.
+from memote.experimental.checks import UnknownIdentifier  # noqa: F401
+
 
 __all__ = ("ExperimentalBase",)
 
@@ -93,7 +96,7 @@ def validate(self, model, checks=[]):
         self.evaluate_report(
             validate(records, headers=list(records[0]),
                      preset='table', schema=self.schema,
-                     order_fields=True, custom_checks=checks))
+                     order_fields=True, checks=checks))
 
     @staticmethod
     def evaluate_report(report):

src/memote/experimental/medium.py

@@ -22,7 +22,7 @@
 import logging
 
 from memote.experimental.experimental_base import ExperimentalBase
-from memote.experimental.checks import check_partial, reaction_id_check
+
 
 __all__ = ("Medium",)
 
@@ -48,8 +48,12 @@ def __init__(self, **kwargs):
     def validate(self, model, checks=[]):
         """Use a defined schema to validate the medium table format."""
         custom = [
-            check_partial(reaction_id_check,
-                          frozenset(r.id for r in model.reactions))
+            {
+                "unknown-identifier": {
+                    "column": "exchange",
+                    "identifiers": {r.id for r in model.reactions}
+                }
+            }
         ]
         super(Medium, self).validate(model=model, checks=checks + custom)