Skip to content

Commit d09824d

Browse files
softwareengineerprogrammersoftwareengineerprogrammer
committed
test_case_study_documentation
1 parent 4d39011 commit d09824d

File tree

1 file changed

+185
-0
lines changed

1 file changed

+185
-0
lines changed

‎tests/geophires_x_tests/test_fervo_project_cape_4.py

Lines changed: 185 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,12 @@
1+
import re
2+
from typing import Any
3+
14
from base_test_case import BaseTestCase
5+
from geophires_x.GeoPHIRESUtils import sig_figs
6+
from geophires_x.Parameter import HasQuantity
27
from geophires_x_client import GeophiresInputParameters
38
from geophires_x_client import GeophiresXClient
9+
from geophires_x_client import GeophiresXResult
410

511

612
class FervoProjectCape4TestCase(BaseTestCase):
@@ -41,3 +47,182 @@ def test_fervo_project_cape_4_results_against_reference_values(self):
4147
]['value']
4248
self.assertGreater(pumping_power_pct, 13)
4349
self.assertLess(pumping_power_pct, 17)
50+
51+
def test_case_study_documentation(self):
52+
"""
53+
Parses result values from case study documentation markdown and checks that they match the actual result.
54+
Useful for catching when minor updates are made to the case study which need to be manually synced to the
55+
documentation.
56+
"""
57+
58+
def _parse_value_unit(raw_string: str) -> dict:
59+
"""
60+
A helper function to parse a string and extract a numerical value and its unit.
61+
It handles various formats like currency, percentages, and scientific notation.
62+
"""
63+
# First, strip any parenthetical notes, e.g., "(based on...)"
64+
clean_str = re.split(r'\(', raw_string)[0].strip()
65+
66+
# Case 1: LCOE format ($X.X/MWh -> cents/kWh)
67+
match = re.match(r'^\$(\d+\.?\d*)/MWh$', clean_str)
68+
if match:
69+
value_ = float(match.group(1))
70+
# Convert $/MWh to cents/kWh by dividing by 10
71+
return {'value': round(value_ / 10, 2), 'unit': 'cents/kWh'}
72+
73+
# Case 2: Billion dollar format ($X.XB -> MUSD)
74+
match = re.match(r'^\$(\d+\.?\d*)B$', clean_str)
75+
if match:
76+
value_ = float(match.group(1))
77+
return {'value': value_ * 1000, 'unit': 'MUSD'}
78+
79+
# Case 3: Million dollar format ($X.XM...)
80+
match = re.match(r'^\$(\d+\.?\d*)M', clean_str)
81+
if match:
82+
value_ = float(match.group(1))
83+
return {'value': value_, 'unit': 'MUSD'}
84+
85+
# Case 4: Dollar per kW format ($X/kW -> USD/kW)
86+
match = re.match(r'^\$(\d+\.?\d*)/kW$', clean_str)
87+
if match:
88+
value_ = float(match.group(1))
89+
return {'value': value_, 'unit': 'USD/kW'}
90+
91+
# Case 5: Percentage format (X.X%)
92+
match = re.match(r'^(\d+\.?\d*)%$', clean_str)
93+
if match:
94+
value_ = float(match.group(1))
95+
return {'value': value_, 'unit': '%'}
96+
97+
# Case 6: Temperature format (X℃ -> degC)
98+
match = re.match(r'^(\d+\.?\d*)℃$', clean_str)
99+
if match:
100+
value_ = float(match.group(1))
101+
return {'value': value_, 'unit': 'degC'}
102+
103+
# Case 7: Scientific notation format (X.X×10⁶ Y) # ruff: noqa: RUF003
104+
match = re.match(r'^(\d+\.?\d*)\s*[×xX]\s*10[⁶6]\s*(.*)$', clean_str)
105+
if match:
106+
base_value = float(match.group(1))
107+
unit = match.group(2).strip()
108+
return {'value': base_value * 1e6, 'unit': unit}
109+
110+
# Case 8: Standard number and unit (e.g., "503 MW")
111+
match = re.match(r'^(\d+\.?\d*)\s*([a-zA-Z²³\/]+)$', clean_str)
112+
if match:
113+
value_ = float(match.group(1))
114+
unit = match.group(2)
115+
return {'value': value_, 'unit': unit}
116+
117+
# Case 9: Dimensionless integer number (e.g., "3")
118+
match = re.match(r'^(\d+)$', clean_str)
119+
if match:
120+
value_ = int(match.group(1))
121+
return {'value': value_, 'unit': 'count'}
122+
123+
# Fallback for any unhandled formats
124+
return {'value': clean_str, 'unit': 'unknown'}
125+
126+
def parse_markdown_results_structured(markdown_text: str) -> dict:
127+
"""
128+
Parses result values from markdown into a structured dictionary with values and units.
129+
"""
130+
raw_results = {}
131+
table_pattern = re.compile(r'^\s*\|\s*(?!-)([^|]+?)\s*\|\s*([^|]+?)\s*\|', re.MULTILINE)
132+
133+
try:
134+
results_start_index = markdown_text.index('## Results')
135+
search_area = markdown_text[results_start_index:]
136+
137+
matches = table_pattern.findall(search_area)
138+
139+
# Use key_ and value_ to avoid shadowing
140+
for match in matches:
141+
key_ = match[0].strip()
142+
value_ = match[1].strip()
143+
if key_.lower() not in ('metric', 'parameter'):
144+
raw_results[key_] = value_
145+
except ValueError:
146+
print("Warning: '## Results' section not found.")
147+
return {}
148+
149+
# Consistency check
150+
special_case_pattern = re.compile(r'LCOE\s*=\s*(\S+)\s*and\s*CAPEX\s*=\s*(\S+)')
151+
special_case_match = special_case_pattern.search(markdown_text)
152+
if special_case_match:
153+
lcoe_text = special_case_match.group(1).rstrip('.,;')
154+
lcoe_table_base = raw_results.get('LCOE', '').split('(')[0].strip()
155+
if lcoe_text != lcoe_table_base:
156+
raise ValueError(
157+
f'LCOE mismatch: Text value ({lcoe_text}) does not match table value ({lcoe_table_base}).'
158+
)
159+
160+
# Now, process the raw results into the structured format
161+
structured_results = {}
162+
# Use key_ and value_ to avoid shadowing
163+
for key_, value_ in raw_results.items():
164+
if key_ in [
165+
'After-tax IRR',
166+
'Average Production Temperature',
167+
'LCOE',
168+
'Maximum Total Electricity Generation',
169+
'Minimum Net Electricity Generation',
170+
'Number of times redrilling',
171+
'Project capital costs: Total CAPEX',
172+
'Project capital costs: $/kW',
173+
'WACC',
174+
'Well Drilling and Completion Cost',
175+
]:
176+
structured_results[key_] = _parse_value_unit(value_)
177+
178+
return structured_results
179+
180+
results_in_markdown = parse_markdown_results_structured(
181+
'\n'.join(self._get_test_file_content('../../docs/Fervo_Project_Cape-4.md'))
182+
)
183+
184+
self.assertEqual(3.96, results_in_markdown['Well Drilling and Completion Cost']['value'])
185+
self.assertEqual('MUSD', results_in_markdown['Well Drilling and Completion Cost']['unit'])
186+
187+
class Q(HasQuantity):
188+
def __init__(self, vu: dict[str, Any]):
189+
self.value = vu['value']
190+
191+
# https://stackoverflow.com/questions/2280334/shortest-way-of-creating-an-object-with-arbitrary-attributes-in-python
192+
self.CurrentUnits = type('', (), {})()
193+
194+
self.CurrentUnits.value = vu['unit']
195+
196+
capex_q = Q(results_in_markdown['Project capital costs: Total CAPEX']).quantity()
197+
markdown_capex_USD_per_kW = (
198+
capex_q.to('USD').magnitude
199+
/ Q(results_in_markdown['Maximum Total Electricity Generation']).quantity().to('kW').magnitude
200+
)
201+
self.assertAlmostEqual(
202+
sig_figs(markdown_capex_USD_per_kW, 3), results_in_markdown['Project capital costs: $/kW']['value']
203+
)
204+
205+
field_mapping = {
206+
'LCOE': 'Electricity breakeven price',
207+
'Project capital costs: Total CAPEX': 'Total CAPEX',
208+
'Well Drilling and Completion Cost': 'Drilling and completion costs per well',
209+
}
210+
211+
ignore_keys = ['Project capital costs: $/kW', 'Total fracture surface area per production well']
212+
213+
example_result = GeophiresXResult(self._get_test_file_path('../examples/Fervo_Project_Cape-4.out'))
214+
example_result_values_in_documentation = {}
215+
for key, _ in results_in_markdown.items():
216+
if key not in ignore_keys:
217+
mapped_key = field_mapping.get(key) if key in field_mapping else key
218+
entry = example_result._get_result_field(mapped_key)
219+
if entry is not None and 'value' in entry:
220+
entry['value'] = sig_figs(entry['value'], 3)
221+
222+
example_result_values_in_documentation[key] = entry
223+
224+
for ignore_key in ignore_keys:
225+
if ignore_key in results_in_markdown:
226+
del results_in_markdown[ignore_key]
227+
228+
self.assertDictAlmostEqual(results_in_markdown, example_result_values_in_documentation, places=3)

0 commit comments

Comments
 (0)