custom.sanity.electricity_capacity.ElectricityCapacityComparison.evaluate_df() - Code Metrics - Inspection of "feature/validation_integration" - openego/eGon-data - Measure and Improve Code Quality continuously with Scrutinizer

Passed

Pull Request — dev (#1375)

unknown

created 2026-01-19 10:52 UTC

ElectricityCapacityComparison.evaluate_df() D

↳ Parent: custom.sanity.electricity_capacity

Complexity

Conditions

Size

Total Lines	138
Code Lines	93

Duplication

Lines	52
Ratio	37.68 %

Importance

Changes

Metric	Value
eloc	93
dl	52
loc	138
rs	4.8957
c	0
b	0
f	0
cc	9
nop	3

How to fix Long Method

"""
Sanity check validation rules for electricity capacity comparison.

Validates that distributed capacities in etrago tables match input capacities
from scenario_capacities table.
"""

from egon_validation.rules.base import DataFrameRule, RuleResult, Severity
from typing import Optional, List


class ElectricityCapacityComparison(DataFrameRule):
    """
    Compare distributed capacity with input capacity for electricity components.

    Compares the total capacity in etrago tables (grid.egon_etrago_generator,
    grid.egon_etrago_storage) against the input capacity from the scenario
    capacities table (supply.egon_scenario_capacities).

    This validation ensures that capacity distribution is correct and no
    capacity is lost or incorrectly added during the distribution process.
    """

    def __init__(
        self,
        table: str,
        rule_id: str,
        scenario: str = "eGon2035",
        carrier: str = "wind_onshore",
        component_type: str = "generator",
        output_carriers: Optional[List[str]] = None,
        rtol: float = 0.10,
        **kwargs
    ):
        """
        Parameters
        ----------
        table : str
            Target table (grid.egon_etrago_generator or grid.egon_etrago_storage)
        rule_id : str
            Unique identifier for this validation rule
        scenario : str
            Scenario name ("eGon2035" or "eGon100RE")
        carrier : str
            Carrier type for the input table (supply.egon_scenario_capacities)
        component_type : str
            Type of component ("generator", "storage", or "link")
        output_carriers : List[str], optional
            List of carrier names in output table. If None, uses carrier parameter.
            Useful for biomass which maps to multiple output carriers.
        rtol : float
            Relative tolerance for capacity deviation (default: 0.10 = 10%)
        """
        super().__init__(
            rule_id=rule_id,
            table=table,
            scenario=scenario,
            carrier=carrier,
            component_type=component_type,
            output_carriers=output_carriers,
            rtol=rtol,
            **kwargs
        )
        self.kind = "sanity"
        self.scenario = scenario
        self.carrier = carrier
        self.component_type = component_type
        self.output_carriers = output_carriers or [carrier]
        self.rtol = rtol

    def get_query(self, ctx):
        """
        Query to compare input and output capacities.

        Returns a query that:
        1. Sums output capacity from etrago table for German buses
        2. Sums input capacity from scenario_capacities table
        3. Returns both values for comparison
        """
        # Build carrier filter for output table
        if len(self.output_carriers) == 1:
            carrier_filter = f"carrier = '{self.output_carriers[0]}'"
        else:
            carriers_str = "', '".join(self.output_carriers)
            carrier_filter = f"carrier IN ('{carriers_str}')"

        # Build bus filter based on component type
        # Links have bus0 and bus1, generators/storage have bus
        if self.component_type == "link":
            bus_filter = f"""
            AND (bus0 IN (
                SELECT bus_id
                FROM grid.egon_etrago_bus
                WHERE scn_name = '{self.scenario}'
                AND country = 'DE'
            ) OR bus1 IN (
                SELECT bus_id
                FROM grid.egon_etrago_bus
                WHERE scn_name = '{self.scenario}'
                AND country = 'DE'
            ))
            """
        else:
            bus_filter = f"""
            AND bus IN (
                SELECT bus_id
                FROM grid.egon_etrago_bus
                WHERE scn_name = '{self.scenario}'
                AND country = 'DE'
            )
            """

        return f"""
        WITH output_capacity AS (
            SELECT
                COALESCE(SUM(p_nom::numeric), 0) as output_capacity_mw
            FROM {self.table}
            WHERE scn_name = '{self.scenario}'
            AND {carrier_filter}
            {bus_filter}
        ),
        input_capacity AS (
            SELECT
                COALESCE(SUM(capacity::numeric), 0) as input_capacity_mw
            FROM supply.egon_scenario_capacities
            WHERE carrier = '{self.carrier}'
            AND scenario_name = '{self.scenario}'
        )
        SELECT
            o.output_capacity_mw,
            i.input_capacity_mw
        FROM output_capacity o
        CROSS JOIN input_capacity i
        """

    def evaluate_df(self, df, ctx):
        """
        Evaluate capacity comparison.

        Parameters
        ----------
        df : pd.DataFrame
            DataFrame with output_capacity_mw and input_capacity_mw columns
        ctx : dict
            Context information

        Returns
        -------
        RuleResult
            Validation result with success/failure status
        """
        if df.empty:
            return RuleResult(
                rule_id=self.rule_id,
                task=self.task,
                table=self.table,
                kind=self.kind,
                success=False,
                message=f"No data found for {self.carrier} capacity comparison",
                severity=Severity.ERROR,
                schema=self.schema,
                table_name=self.table_name,
                rule_class=self.__class__.__name__
            )

        output_capacity = float(df["output_capacity_mw"].values[0])
        input_capacity = float(df["input_capacity_mw"].values[0])

        # Case 1: Both zero - OK, no capacity needed
        if output_capacity == 0 and input_capacity == 0:

            return RuleResult(
                rule_id=self.rule_id,
                task=self.task,
                table=self.table,
                kind=self.kind,
                success=True,
                observed=0.0,
                expected=0.0,
                message=(
                    f"No {self.carrier} {self.component_type} capacity needed "
                    f"for {self.scenario} (both input and output are zero)"
                ),
                severity=Severity.INFO,
                schema=self.schema,
                table_name=self.table_name,
                rule_class=self.__class__.__name__
            )

        # Case 2: Input > 0 but output = 0 - ERROR
        if input_capacity > 0 and output_capacity == 0:

            return RuleResult(
                rule_id=self.rule_id,
                task=self.task,
                table=self.table,
                kind=self.kind,
                success=False,
                observed=0.0,
                expected=input_capacity,
                message=(
                    f"{self.carrier} {self.component_type} capacity was not distributed at all! "
                    f"Input: {input_capacity:.2f} MW, Output: 0 MW for {self.scenario}"
                ),
                severity=Severity.ERROR,
                schema=self.schema,
                table_name=self.table_name,
                rule_class=self.__class__.__name__
            )

        # Case 3: Output > 0 but input = 0 - ERROR
        if output_capacity > 0 and input_capacity == 0:

            return RuleResult(
                rule_id=self.rule_id,
                task=self.task,
                table=self.table,
                kind=self.kind,
                success=False,
                observed=output_capacity,
                expected=0.0,
                message=(
                    f"{self.carrier} {self.component_type} capacity was distributed "
                    f"even though no input was provided! "
                    f"Output: {output_capacity:.2f} MW, Input: 0 MW for {self.scenario}"
                ),
                severity=Severity.ERROR,
                schema=self.schema,
                table_name=self.table_name,
                rule_class=self.__class__.__name__
            )

        # Case 4: Both > 0 - Check deviation
        deviation = abs(output_capacity - input_capacity) / input_capacity
        deviation_pct = deviation * 100
        error_pct = ((output_capacity - input_capacity) / input_capacity) * 100

        success = deviation <= self.rtol

        if success:
            return RuleResult(
                rule_id=self.rule_id,
                task=self.task,
                table=self.table,
                kind=self.kind,
                success=True,
                observed=output_capacity,
                expected=input_capacity,
                message=(
                    f"{self.carrier} {self.component_type} capacity valid for {self.scenario}: "
                    f"Output: {output_capacity:.2f} MW, Input: {input_capacity:.2f} MW, "
                    f"Deviation: {error_pct:+.2f}% (tolerance: ±{self.rtol*100:.2f}%)"
                ),
                severity=Severity.INFO,
                schema=self.schema,
                table_name=self.table_name,
                rule_class=self.__class__.__name__
            )
        else:
            return RuleResult(
                rule_id=self.rule_id,
                task=self.task,
                table=self.table,
                kind=self.kind,
                success=False,
                observed=output_capacity,
                expected=input_capacity,
                message=(
                    f"{self.carrier} {self.component_type} capacity deviation too large for {self.scenario}: "
                    f"Output: {output_capacity:.2f} MW, Input: {input_capacity:.2f} MW, "
                    f"Deviation: {error_pct:+.2f}% (tolerance: ±{self.rtol*100:.2f}%)"
                ),
                severity=Severity.ERROR,
                schema=self.schema,
                table_name=self.table_name,
                rule_class=self.__class__.__name__
            )


1		"""
2		Sanity check validation rules for electricity capacity comparison.
3
4		Validates that distributed capacities in etrago tables match input capacities
5		from scenario_capacities table.
6		"""
7
8		from egon_validation.rules.base import DataFrameRule, RuleResult, Severity
9		from typing import Optional, List
10
11
12		class ElectricityCapacityComparison(DataFrameRule):
13		"""
14		Compare distributed capacity with input capacity for electricity components.
15
16		Compares the total capacity in etrago tables (grid.egon_etrago_generator,
17		grid.egon_etrago_storage) against the input capacity from the scenario
18		capacities table (supply.egon_scenario_capacities).
19
20		This validation ensures that capacity distribution is correct and no
21		capacity is lost or incorrectly added during the distribution process.
22		"""
23
24		def __init__(
25		self,
26		table: str,
27		rule_id: str,
28		scenario: str = "eGon2035",
29		carrier: str = "wind_onshore",
30		component_type: str = "generator",
31		output_carriers: Optional[List[str]] = None,
32		rtol: float = 0.10,
33		**kwargs
34		):
35		"""
36		Parameters
37		----------
38		table : str
39		Target table (grid.egon_etrago_generator or grid.egon_etrago_storage)
40		rule_id : str
41		Unique identifier for this validation rule
42		scenario : str
43		Scenario name ("eGon2035" or "eGon100RE")
44		carrier : str
45		Carrier type for the input table (supply.egon_scenario_capacities)
46		component_type : str
47		Type of component ("generator", "storage", or "link")
48		output_carriers : List[str], optional
49		List of carrier names in output table. If None, uses carrier parameter.
50		Useful for biomass which maps to multiple output carriers.
51		rtol : float
52		Relative tolerance for capacity deviation (default: 0.10 = 10%)
53		"""
54		super().__init__(
55		rule_id=rule_id,
56		table=table,
57		scenario=scenario,
58		carrier=carrier,
59		component_type=component_type,
60		output_carriers=output_carriers,
61		rtol=rtol,
62		**kwargs
63		)
64		self.kind = "sanity"
65		self.scenario = scenario
66		self.carrier = carrier
67		self.component_type = component_type
68		self.output_carriers = output_carriers or [carrier]
69		self.rtol = rtol
70
71		def get_query(self, ctx):
72		"""
73		Query to compare input and output capacities.
74
75		Returns a query that:
76		1. Sums output capacity from etrago table for German buses
77		2. Sums input capacity from scenario_capacities table
78		3. Returns both values for comparison
79		"""
80		# Build carrier filter for output table
81		if len(self.output_carriers) == 1:
82		carrier_filter = f"carrier = '{self.output_carriers[0]}'"
83		else:
84		carriers_str = "', '".join(self.output_carriers)
85		carrier_filter = f"carrier IN ('{carriers_str}')"
86
87		# Build bus filter based on component type
88		# Links have bus0 and bus1, generators/storage have bus
89		if self.component_type == "link":
90		bus_filter = f"""
91		AND (bus0 IN (
92		SELECT bus_id
93		FROM grid.egon_etrago_bus
94		WHERE scn_name = '{self.scenario}'
95		AND country = 'DE'
96		) OR bus1 IN (
97		SELECT bus_id
98		FROM grid.egon_etrago_bus
99		WHERE scn_name = '{self.scenario}'
100		AND country = 'DE'
101		))
102		"""
103		else:
104		bus_filter = f"""
105		AND bus IN (
106		SELECT bus_id
107		FROM grid.egon_etrago_bus
108		WHERE scn_name = '{self.scenario}'
109		AND country = 'DE'
110		)
111		"""
112
113		return f"""
114		WITH output_capacity AS (
115		SELECT
116		COALESCE(SUM(p_nom::numeric), 0) as output_capacity_mw
117		FROM {self.table}
118		WHERE scn_name = '{self.scenario}'
119		AND {carrier_filter}
120		{bus_filter}
121		),
122		input_capacity AS (
123		SELECT
124		COALESCE(SUM(capacity::numeric), 0) as input_capacity_mw
125		FROM supply.egon_scenario_capacities
126		WHERE carrier = '{self.carrier}'
127		AND scenario_name = '{self.scenario}'
128		)
129		SELECT
130		o.output_capacity_mw,
131		i.input_capacity_mw
132		FROM output_capacity o
133		CROSS JOIN input_capacity i
134		"""
135
136		def evaluate_df(self, df, ctx):
137		"""
138		Evaluate capacity comparison.
139
140		Parameters
141		----------
142		df : pd.DataFrame
143		DataFrame with output_capacity_mw and input_capacity_mw columns
144		ctx : dict
145		Context information
146
147		Returns
148		-------
149		RuleResult
150		Validation result with success/failure status
151		"""
152		if df.empty:
153		return RuleResult(
154		rule_id=self.rule_id,
155		task=self.task,
156		table=self.table,
157		kind=self.kind,
158		success=False,
159		message=f"No data found for {self.carrier} capacity comparison",
160		severity=Severity.ERROR,
161		schema=self.schema,
162		table_name=self.table_name,
163		rule_class=self.__class__.__name__
164		)
165
166		output_capacity = float(df["output_capacity_mw"].values[0])
167		input_capacity = float(df["input_capacity_mw"].values[0])
168
169		# Case 1: Both zero - OK, no capacity needed
170	View Code Duplication	if output_capacity == 0 and input_capacity == 0:
		0 ignored issues – show Duplication introduced 2026-01-16 09:20 UTC by Report Bug Copy Issue Report This code seems to be duplicated in your project. Loading history...
171		return RuleResult(
172		rule_id=self.rule_id,
173		task=self.task,
174		table=self.table,
175		kind=self.kind,
176		success=True,
177		observed=0.0,
178		expected=0.0,
179		message=(
180		f"No {self.carrier} {self.component_type} capacity needed "
181		f"for {self.scenario} (both input and output are zero)"
182		),
183		severity=Severity.INFO,
184		schema=self.schema,
185		table_name=self.table_name,
186		rule_class=self.__class__.__name__
187		)
188
189		# Case 2: Input > 0 but output = 0 - ERROR
190	View Code Duplication	if input_capacity > 0 and output_capacity == 0:
		0 ignored issues – show Duplication introduced 2026-01-16 09:20 UTC by Report Bug Copy Issue Report This code seems to be duplicated in your project. Loading history...
191		return RuleResult(
192		rule_id=self.rule_id,
193		task=self.task,
194		table=self.table,
195		kind=self.kind,
196		success=False,
197		observed=0.0,
198		expected=input_capacity,
199		message=(
200		f"{self.carrier} {self.component_type} capacity was not distributed at all! "
201		f"Input: {input_capacity:.2f} MW, Output: 0 MW for {self.scenario}"
202		),
203		severity=Severity.ERROR,
204		schema=self.schema,
205		table_name=self.table_name,
206		rule_class=self.__class__.__name__
207		)
208
209		# Case 3: Output > 0 but input = 0 - ERROR
210	View Code Duplication	if output_capacity > 0 and input_capacity == 0:
		0 ignored issues – show Duplication introduced 2026-01-16 09:20 UTC by Report Bug Copy Issue Report This code seems to be duplicated in your project. Loading history...
211		return RuleResult(
212		rule_id=self.rule_id,
213		task=self.task,
214		table=self.table,
215		kind=self.kind,
216		success=False,
217		observed=output_capacity,
218		expected=0.0,
219		message=(
220		f"{self.carrier} {self.component_type} capacity was distributed "
221		f"even though no input was provided! "
222		f"Output: {output_capacity:.2f} MW, Input: 0 MW for {self.scenario}"
223		),
224		severity=Severity.ERROR,
225		schema=self.schema,
226		table_name=self.table_name,
227		rule_class=self.__class__.__name__
228		)
229
230		# Case 4: Both > 0 - Check deviation
231		deviation = abs(output_capacity - input_capacity) / input_capacity
232		deviation_pct = deviation * 100
233		error_pct = ((output_capacity - input_capacity) / input_capacity) * 100
234
235		success = deviation <= self.rtol
236
237		if success:
238		return RuleResult(
239		rule_id=self.rule_id,
240		task=self.task,
241		table=self.table,
242		kind=self.kind,
243		success=True,
244		observed=output_capacity,
245		expected=input_capacity,
246		message=(
247		f"{self.carrier} {self.component_type} capacity valid for {self.scenario}: "
248		f"Output: {output_capacity:.2f} MW, Input: {input_capacity:.2f} MW, "
249		f"Deviation: {error_pct:+.2f}% (tolerance: ±{self.rtol*100:.2f}%)"
250		),
251		severity=Severity.INFO,
252		schema=self.schema,
253		table_name=self.table_name,
254		rule_class=self.__class__.__name__
255		)
256		else:
257		return RuleResult(
258		rule_id=self.rule_id,
259		task=self.task,
260		table=self.table,
261		kind=self.kind,
262		success=False,
263		observed=output_capacity,
264		expected=input_capacity,
265		message=(
266		f"{self.carrier} {self.component_type} capacity deviation too large for {self.scenario}: "
267		f"Output: {output_capacity:.2f} MW, Input: {input_capacity:.2f} MW, "
268		f"Deviation: {error_pct:+.2f}% (tolerance: ±{self.rtol*100:.2f}%)"
269		),
270		severity=Severity.ERROR,
271		schema=self.schema,
272		table_name=self.table_name,
273		rule_class=self.__class__.__name__
274		)
275

openego / eGon-data

Pull Request — dev (#1375)

ElectricityCapacityComparison.evaluate_df() D

Complexity

Size

Duplication

Importance

How to fix Long Method

Long Method

Duplication Side-by-Side

Filter issues like