def _deactivate_alineas(
text: StructuredText, inapplicability: PotentialInapplicability,
parameter_values: Dict[Parameter, Any]) -> StructuredText:
text = copy(text)
inactive_alineas = inapplicability.alineas
all_inactive = inactive_alineas is None
warning = generate_inactive_warning(inapplicability.condition,
parameter_values,
all_alineas=all_inactive,
whole_text=False)
if inactive_alineas is not None:
inactive_alineas_set = set(inactive_alineas)
new_outer_alineas = [
replace(al, inactive=i in inactive_alineas_set)
for i, al in enumerate(text.outer_alineas)
]
else:
new_outer_alineas = [
replace(al, inactive=True) for al in text.outer_alineas
]
text.applicability = Applicability(active=not all_inactive,
warnings=[warning])
if inapplicability.subsections_are_inapplicable:
text.sections = [
_deactivate_child_section(section, all_inactive=all_inactive)
for section in text.sections
]
text.outer_alineas = new_outer_alineas
return text
def _detect_and_add_topics(section: StructuredText,
matching_depth: int,
current_depth: int = 0) -> StructuredText:
section = copy(section)
topic = _detect(
section.title.text) if matching_depth == current_depth else None
section.annotations = Annotations(topic)
section.sections = [
_detect_and_add_topics(sec, matching_depth, current_depth + 1)
for sec in section.sections
]
return section
def _get_simple_text() -> StructuredText:
row1 = Row([_cell('AA'), _cell('BB'), _cell('CC')], True)
row2 = Row([_cell('DD', rs=2), _cell('EE', cs=2)], False)
row3 = Row([_cell('FF'), _cell('GG')], False)
table = Table([row1, row2, row3])
alineas = [
_text('Alinea 1'),
_text('Alinea 2'),
EnrichedString('', table=table)
]
sections = [StructuredText(_text('Title 2'), [], [], None, None, None)]
text = StructuredText(_text('Title'), alineas, sections, None, None, None)
return text
def _deactivate_child_section(section: StructuredText,
all_inactive: bool) -> StructuredText:
section = copy(section)
if section.applicability:
section.applicability.active = not all_inactive
else:
section.applicability = Applicability(active=not all_inactive)
section.sections = [
_deactivate_child_section(sec, all_inactive)
for sec in section.sections
]
section.outer_alineas = [
replace(al, inactive=True) for al in section.outer_alineas
]
return section
def from_dict(cls, dict_: Dict[str, Any]) -> 'AlternativeSection':
return AlternativeSection(
dict_['section_id'],
StructuredText.from_dict(dict_['new_text']),
load_condition(dict_['condition']),
dict_['id'],
)
def compute_detection_performance(dataset: List[_LabelizedText], ontology: TopicOntology):
all_labels = [label for _, label in dataset]
texts = [
(titles, StructuredText(EnrichedString(''), outer_alineas, [], None)) for (titles, outer_alineas), _ in dataset
]
for topic, texts_ in _missing_topics(all_labels, texts, ontology).items():
_pretty_print(topic, texts_)
predicted_labels = [_extract_topics(text, titles, ontology) for titles, text in texts]
print(Counter([exp == pred for exp, pred in zip(all_labels, predicted_labels) if exp]))
def _node_section() -> StructuredText:
return StructuredText(
_str('abc'),
[_str('abc'), _enriched_string_table()],
[_leaf_section()],
None,
Reference('ref', 'name'),
Annotations(None),
)
def _apply_parameter_values_in_text(
text: StructuredText, parameter_values: Dict[Parameter,
Any]) -> StructuredText:
na_conditions, modifications, warnings = _extract_satisfied_objects_and_warnings(
text.parametrization, parameter_values)
text = copy(text)
if not na_conditions and not modifications:
text.sections = [
_apply_parameter_values_in_text(section, parameter_values)
for section in text.sections
]
text.applicability = Applicability()
else:
text = _apply_satisfied_modificators(text, na_conditions,
modifications, parameter_values)
all_warnings = sorted(
set(_ensure_applicabiliy(text.applicability).warnings + warnings))
text.applicability = replace(_ensure_applicabiliy(text.applicability),
warnings=all_warnings)
return text
def test_add_topics():
sub_section_1 = StructuredText(EnrichedString('Section 1.1'), [], [], None)
section_1 = StructuredText(EnrichedString('Section 1'), [],
[sub_section_1], None)
section_2 = StructuredText(EnrichedString('Section 2'), [], [], None)
am = ArreteMinisteriel(EnrichedString('arrete du 10/10/10'),
[section_1, section_2], [],
None,
id='FAKE_ID')
am_with_topics = add_topics(
am, {
(0, ): TopicName.INCENDIE,
(0, 0): TopicName.INCENDIE,
(1, ): TopicName.BRUIT_VIBRATIONS
})
sections = am_with_topics.sections
assert sections[0].annotations.topic == TopicName.INCENDIE # type: ignore
assert sections[0].sections[
0].annotations.topic == TopicName.INCENDIE # type: ignore
assert sections[
1].annotations.topic == TopicName.BRUIT_VIBRATIONS # type: ignore
def test_extract_text_lines():
assert _get_simple_text().text_lines() == [
'AM',
'alinea',
'foo',
'# Section 1',
'## Section 1.1',
'# Section 2',
'bar',
]
assert StructuredText(EnrichedString(' A'), [], [], None).text_lines() == ['A']
assert StructuredText(EnrichedString(' A'), [EnrichedString('')], [], None).text_lines() == ['A', '']
assert _TEXT_A.text_lines() == [
'6. Schématisation des différents types de joints mentionnés :',
'Vous pouvez consulter les schémas dans le',
'JO',
'n° 265 du 16/11/2010 texte numéro 21',
]
assert _TEXT_B.text_lines() == [
'6. Schématisation des différents types de joints mentionnés :',
'Vous pouvez consulter les schémas dans le',
'JO n° 265 du 16/11/2010 texte numéro 21',
]
def _build_am() -> ArreteMinisteriel:
subsection1 = StructuredText(EnrichedString('A'), [], [], None)
subsection2 = StructuredText(EnrichedString('A'), [], [], None)
sections = [
StructuredText(EnrichedString('Art. 1'), [], [subsection1], None),
StructuredText(EnrichedString('Art. 2'), [], [subsection2], None),
StructuredText(EnrichedString('Conditions d\'application'), [], [],
None),
StructuredText(EnrichedString('Art. 3'), [], [], None),
]
return ArreteMinisteriel(EnrichedString('arrete du 10/10/10'),
sections, [],
None,
id='FAKE_ID')
def test_check_parametrization_consistency():
date_ = ParameterEnum.DATE_INSTALLATION.value
dt_1 = datetime.now()
dt_2 = dt_1 + timedelta(days=1)
cd_1 = Range(date_, dt_1, dt_2)
cd_3 = Littler(date_, dt_1)
cd_4 = Greater(date_, dt_2)
cd_5 = Littler(date_, dt_2)
cd_6 = Greater(date_, dt_1)
new_text = StructuredText(_str('Art. 2'), [_str('version modifiée')], [],
None)
Parametrization(
[_NAC('', None, cd_1), _NAC('', None, cd_3)],
[AlternativeSection('', new_text, cd_4)], []).check_consistency()
Parametrization([_NAC('', None, cd_3)], [], []).check_consistency()
with pytest.raises(ParametrizationError):
Parametrization(
[_NAC('', None, cd_1), _NAC('', None, cd_5)],
[AlternativeSection('', new_text, cd_6)], []).check_consistency()
with pytest.raises(ParametrizationError):
Parametrization(
[_NAC('', None, cd_1), _NAC('', None, cd_1)], [],
[]).check_consistency()
assert not _is_probably_cid('JORFTEX')
def _get_simple_text() -> StructuredText:
sub_section_1 = StructuredText(_str('Section 1.1'), [], [], None)
section_1 = StructuredText(_str('Section 1'), [], [sub_section_1], None)
section_2 = StructuredText(_str('Section 2'), [_str('bar')], [], None)
return StructuredText(_str('AM '), [_str('alinea'), _str('foo')], [section_1, section_2], None)
_TEXT_A = StructuredText(
title=EnrichedString(text='6. Schématisation des différents types de joints mentionnés :'),
outer_alineas=[
EnrichedString(text='Vous pouvez consulter les schémas dans le'),
EnrichedString(text='JO\nn° 265 du 16/11/2010 texte numéro 21'),
],
sections=[],
applicability=None,
reference=Reference('ref', 'name'),
annotations=None,
id='0bEB0b14A96f',
)
_TEXT_B = StructuredText(
title=EnrichedString(text='6. Schématisation des différents types de joints mentionnés :'),
outer_alineas=[
EnrichedString(text='Vous pouvez consulter les schémas dans le'),
EnrichedString(text='JO n° 265 du 16/11/2010 texte numéro 21'),
],
sections=[],
applicability=None,
reference=None,
annotations=None,
def test_structured_text():
dict_ = _node_section().to_dict()
new_dict = StructuredText.from_dict(dict_).to_dict()
assert new_dict == dict_
def _leaf_section() -> StructuredText:
app = Applicability(True, True, ['beware'], StructuredText(_str('abc'), [], [], None))
annotations = Annotations(TopicName.AIR_ODEURS)
return StructuredText(_str('abc'), [_str('abc')], [], app, Reference('ref', 'name'), annotations)
def test_add_references():
sub_sub_sections = [
StructuredText(estr('1.1. azeaze'), [], [], None, None),
StructuredText(estr('1. 2. azeaze'), [], [], None, None),
]
sub_sections = [
StructuredText(estr('1. azeaze'), [], sub_sub_sections, None, None)
]
sections = [
StructuredText(estr('Article 1. efzefz'), [], sub_sections, None),
StructuredText(estr('2. zefez'), [], [], None),
StructuredText(estr('A. zefze'), [], [], None),
StructuredText(estr('a) zefze'), [], [], None),
StructuredText(estr('V. zefze'), [], [], None),
StructuredText(estr('ANNEXE I zefze'), [], [], None),
StructuredText(estr('Article 18.1'), [], [], None),
StructuredText(estr('Article 1'), [], [], None),
]
am = ArreteMinisteriel(estr('Arrete du 10/10/10'),
sections, [],
None,
id='FAKE_ID')
am_with_references = add_references(am)
secs = am_with_references.sections
assert ensure_not_none(secs[0].reference).nb == 'Article 1.'
assert ensure_not_none(
secs[0].sections[0].reference).nb == 'Article 1. > 1.'
assert ensure_not_none(
secs[0].sections[0].sections[0].reference).nb == 'Article 1. > 1.1.'
assert ensure_not_none(
secs[0].sections[0].sections[1].reference).nb == 'Article 1. > 1.2.'
assert ensure_not_none(secs[1].reference).nb == '2.'
assert ensure_not_none(secs[2].reference).nb == 'A.'
assert ensure_not_none(secs[3].reference).nb == 'a)'
assert ensure_not_none(secs[4].reference).nb == 'V.'
assert ensure_not_none(secs[5].reference).nb == 'Annexe I'
assert ensure_not_none(secs[6].reference).nb == 'Article 18.1'
assert ensure_not_none(secs[7].reference).nb == 'Article 1'
def _add_parametrization_in_section(text: StructuredText,
parametrization: Parametrization) -> None:
text.parametrization = _init_section_parametrization(
parametrization, text.id)
def _get_simple_text() -> StructuredText:
sub_section_1 = StructuredText(_str('Section 1.1'), [], [], None)
section_1 = StructuredText(_str('Section 1'), [], [sub_section_1], None)
section_2 = StructuredText(_str('Section 2'), [_str('bar')], [], None)
return StructuredText(_str('AM '), [_str('alinea'), _str('foo')], [section_1, section_2], None)
from envinorma.models.text_elements import EnrichedString
from envinorma.parametrization.combinations import _generate_options_dicts, generate_exhaustive_combinations
from envinorma.parametrization.models.parametrization import (
AlternativeSection,
AMWarning,
InapplicableSection,
Parametrization,
extract_conditions_from_parametrization,
)
_DATE = Parameter(id='date-d-installation', type=ParameterType.DATE)
_NEW_TEXT = StructuredText(
title=EnrichedString(text='Article 2.1'),
outer_alineas=[EnrichedString(text='Contenu nouveau')],
sections=[],
applicability=None,
reference=None,
annotations=None,
id='d16d0fE7C7fc',
)
_PARAMETRIZATION = Parametrization(
inapplicable_sections=[
InapplicableSection(
section_id='abcdef',
alineas=None,
condition=AndCondition(
conditions=frozenset([Littler(parameter=_DATE, target=date(2021, 1, 1), strict=True)])
),
)
],
alternative_sections=[