Files
quality_analyse/verification/test_native_example_structure.py

82 lines
5.2 KiB
Python
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
"""A preceding measure may dominate examples, never a token's own field."""
import sys,unittest
from pathlib import Path
sys.path.insert(0,str(Path(__file__).resolve().parents[1]/'scripts'))
from nao_native_contract import native_span_structure
class ExamplePredicateStructureTests(unittest.TestCase):
def locate(self,text,token):
start=text.index(token);return native_span_structure(text).locate(start,start+len(token))
def test_actual_repair_example_inherits_only_the_measure(self):
text=('- 9月Top名单各规格亮片报废重量均高于黑片,例如 {{I13}} 为亮片 {{N115}} kg、黑片 {{N114}} kg,'
'{{I8}} 为亮片 {{N65}} kg、黑片 {{N64}} kg;8月Top名单中 {{I11}} 则以黑片 {{N82}} kg 高于亮片 {{N83}} kg。')
for token,stage in [('N115','bright'),('N114','black'),('N65','bright'),('N64','black'),('N82','black'),('N83','bright')]:
location=self.locate(text,'{{'+token+'}}')
self.assertEqual(location['metricFamily'],'scrap')
self.assertEqual(location['ownerStage'],stage)
self.assertEqual(location['fieldAssociation']['kind'],'governing_predicate')
self.assertEqual(location['fieldAssociation']['label'],'报废重量')
self.assertNotIn('9月',location['fieldAssociation']['label'])
self.assertNotIn('亮片',location['fieldAssociation']['label'])
self.assertIn('8月',self.locate(text,'{{N82}}')['prefix'])
def test_actual_initial_sort_frame_governs_the_following_observation(self):
text='**2026年9月 / 消费电子产业部 / 消费电子-3C** 按规格合计报废重量排序,{{I1}} 以 {{N9}} kg 居首,综合合格率 {{N10}}。'
location=self.locate(text,'{{N9}}')
self.assertEqual(location['metricFamily'],'scrap')
self.assertEqual(location['fieldAssociation']['label'],'报废重量')
self.assertNotIn('消费电子',location['fieldAssociation']['label'])
self.assertIsNone(location['ownerStage'])
self.assertEqual(self.locate(text,'{{N10}}')['metricFamily'],'rate')
def test_explicit_period_and_stage_do_not_inherit_from_field_frame(self):
text='9月亮片报废重量较高,8月{{I1}}黑片{{N1}} kg。'
location=self.locate(text,'{{N1}}')
self.assertEqual(location['metricFamily'],'scrap')
self.assertEqual(location['ownerStage'],'black')
self.assertNotIn('9月',location['predicateContext'])
self.assertNotIn('亮片',location['predicateContext'])
def test_local_nonmass_fields_bar_the_prior_mass_frame(self):
for label,family in [('报废率','scrap_rate'),('报废重量占比','share'),('记录数','count')]:
text='报废重量排序,'+label+'另述,{{I1}}以{{N1}} kg居首。'
self.assertEqual(self.locate(text,'{{N1}}')['metricFamily'],family)
def test_rank_table_exposes_original_cells_and_zero_based_column(self):
text='| 排名 | 规格 | 综合报废重量 | 综合合格率 |\n|---|---|---|---|\n| 1 | {{I1}} | {{N9}} kg | {{N10}} |'
location=self.locate(text,'{{N9}}')
self.assertEqual(location['columnIndex'],2)
self.assertEqual(location['rowLabel'],'1')
self.assertEqual([cell['text'].strip() for cell in location['rowCells']],['1','{{I1}}','{{N9}} kg','{{N10}}'])
for cell in location['rowCells']:
self.assertEqual(cell['text'],text[cell['start']:cell['end']])
self.assertEqual(location['metricFamily'],'scrap')
def test_cells_expose_ambiguity_without_inventing_a_subject_or_field(self):
text='| 排名 | 规格 | 其他规格 | 综合合格率 |\n|---|---|---|---|\n| 1 | {{I1}} | {{I2}} | {{N1}} |'
location=self.locate(text,'{{N1}}')
self.assertEqual(location['columnIndex'],3)
self.assertEqual(location['rowLabel'],'1')
self.assertEqual(location['metricFamily'],'rate')
self.assertEqual(location['ownerStage'],'overall')
self.assertEqual([cell['text'].strip() for cell in location['rowCells']][1:3],['{{I1}}','{{I2}}'])
def test_explicit_new_field_overrides_and_future_text_cannot_authorize(self):
text='报废重量较高,例如{{I1}}为黑片{{N1}} kg,投入重量较低,例如{{I2}}为亮片{{N2}} kg。'
self.assertEqual(self.locate(text,'{{N1}}')['metricFamily'],'scrap')
self.assertEqual(self.locate(text,'{{N2}}')['metricFamily'],'input')
future='例如{{I1}}为黑片{{N1}} kg,后面另述报废重量较高。'
self.assertIsNone(self.locate(future,'{{N1}}')['metricFamily'])
stopped='报废重量较高。另段例如{{I1}}为黑片{{N1}} kg。'
self.assertIsNone(self.locate(stopped,'{{N1}}')['metricFamily'])
def test_rate_share_and_new_list_are_not_mass_antecedents(self):
for label in ('报废率','报废重量占比','报废份额','记录数'):
text=label+'较高,例如{{I1}}为黑片{{N1}} kg。'
self.assertNotEqual(self.locate(text,'{{N1}}')['metricFamily'],'scrap')
text='- 报废重量较高,例如{{I1}}为黑片{{N1}} kg。\n- 另述例如{{I2}}为亮片{{N2}} kg。'
self.assertIsNone(self.locate(text,'{{N2}}')['metricFamily'])
if __name__=='__main__':unittest.main()