def convert(
cls, tree_element: model.ArticleTitleAmendment
) -> Iterable[OutgoingReference]:
assert tree_element.article is not None
assert tree_element.act_reference is not None
act_id = ActReferenceConversionHelper.get_act_id_from_parse_result(
tree_element.act_reference)
article_id = "".join(tree_element.article.id)
act_start_pos, act_end_pos = ActReferenceConversionHelper.get_act_id_pos_from_parse_result(
tree_element.act_reference)
yield OutgoingReference(act_start_pos, act_end_pos,
Reference(act=act_id))
ref_start_pos, ref_end_pos = get_subtree_start_and_end_pos(
tree_element.article)
yield OutgoingReference(ref_start_pos, ref_end_pos,
Reference(act=act_id, article=article_id))
def convert_parsed_references(
cls,
parsed_references: Iterable[OutgoingReference],
prefixlen: int,
textlen: int,
abbreviations_map: Mapping[str, str],
) -> Tuple[OutgoingReference, ...]:
# TODO: pylint is right here, should refactor.
# pylint: disable=too-many-arguments
result = []
for in_text_reference in parsed_references:
# The end of the parsed reference is inside the target string
# Checking for the end and not the beginning is important, because
# we also want partial references to work here.
if in_text_reference.end_pos <= prefixlen or in_text_reference.end_pos > textlen:
continue
result.append(
OutgoingReference(
start_pos=max(in_text_reference.start_pos - prefixlen, 0),
end_pos=(in_text_reference.end_pos - prefixlen),
reference=in_text_reference.reference.
resolve_abbreviations(abbreviations_map)))
result.sort()
return tuple(result)
def convert(
cls,
tree_element: model.BlockAmendment) -> Iterable[OutgoingReference]:
assert isinstance(tree_element.act_reference, model.ActReference)
act_id = ActReferenceConversionHelper.get_act_id_from_parse_result(
tree_element.act_reference)
act_start_pos, act_end_pos = ActReferenceConversionHelper.get_act_id_pos_from_parse_result(
tree_element.act_reference)
yield OutgoingReference(act_start_pos, act_end_pos,
Reference(act=act_id))
amended_reference = ReferenceConversionHelper.convert_potential_in_text_reference(
act_id, tree_element.amended_reference)
inserted_reference = ReferenceConversionHelper.convert_potential_in_text_reference(
act_id, tree_element.inserted_reference)
if amended_reference is not None and inserted_reference is not None:
# Act has to be cut off first, because otherwise relative_to does not do anything.
fixed_ref = attr.evolve(inserted_reference.reference,
act=None).relative_to(
amended_reference.reference)
inserted_reference = attr.evolve(inserted_reference,
reference=fixed_ref)
if amended_reference:
yield amended_reference
if inserted_reference:
yield inserted_reference
def convert(cls,
tree_element: model.Repeal) -> Iterable[OutgoingReference]:
assert tree_element.references is not None
act_id = ActReferenceConversionHelper.get_act_id_from_parse_result(
tree_element.act_reference)
act_start_pos, act_end_pos = ActReferenceConversionHelper.get_act_id_pos_from_parse_result(
tree_element.act_reference)
yield OutgoingReference(act_start_pos, act_end_pos,
Reference(act=act_id))
yield from ReferenceConversionHelper.convert_multiple_in_text_references(
act_id, tree_element.references)
def iter(self, start_override: Optional[int],
end_override: int) -> Iterable[OutgoingReference]:
self.commit_deferred_item()
ref_args = [self.act, None, None, None, None]
levels = ("articles", "paragraphs", "points", "subpoints")
for arg_pos, level in enumerate(levels, start=1):
level_vals = getattr(self, level)
if len(level_vals) == 1:
ref_args[arg_pos] = level_vals[0][0]
else:
level_vals.sort(key=lambda x: x[1])
for level_val, start, end in level_vals[:-1]:
if start_override is not None:
start = start_override
start_override = None
ref_args[arg_pos] = level_val
yield OutgoingReference(start, end, Reference(*ref_args))
ref_args[arg_pos] = level_vals[-1][0]
if start_override is None:
start_override = level_vals[-1][1]
assert start_override is not None
yield OutgoingReference(start_override, end_override,
Reference(*ref_args))
def convert(
cls, tree_element: model.CompoundReference
) -> Iterable[OutgoingReference]:
act_id = None
if tree_element.act_reference is not None:
act_id = ActReferenceConversionHelper.get_act_id_from_parse_result(
tree_element.act_reference)
start_pos, end_pos = ActReferenceConversionHelper.get_act_id_pos_from_parse_result(
tree_element.act_reference)
yield OutgoingReference(start_pos, end_pos, Reference(act=act_id))
if tree_element.references:
for reference in tree_element.references:
assert isinstance(reference, model.Reference)
yield from ReferenceConversionHelper.convert_single_in_text_reference(
act_id, reference)
def test_semantic_reparse_abbrevs() -> None:
TEST_ACT = Act(
identifier="2050. évi XD. törvény",
publication_date=Date(2050, 3, 4),
subject="A nyelvtani tesztelésről",
preamble='',
children=(
Article(
identifier="1",
children=(Paragraph(
text=
"Fontos lesz később a tesztelés X tulajdonságiról szóló 2040. évi DX. törvény (a továbbiakban Xtv.) rövidítésének feloldása.",
), )),
Article(
identifier="2",
children=(
Paragraph(
identifier="1",
text=
"Bekeverünk a tesztelés Y tulajdonságiról szóló 2041. évi X. törvény (a továbbiakban Ytv.) dolgaival",
),
Paragraph(
identifier="2",
text=
"Itt megemlítendő az Ytv. 10. §-a és a tesztelés Z tulajdonságiról szóló 2041. évi XXX. törvény (a továbbiakban Ztv.) 1. §-a közötti különbség",
),
Paragraph(
identifier="3",
intro="Mert később használatban van",
children=(
AlphabeticPoint(identifier="a",
text="az Xtv. 1. § c) pontja, és"),
AlphabeticPoint(identifier="b",
text="az Ytv. 2. §-a."),
),
),
),
),
Article(
identifier="3",
children=(Paragraph(
text=
"Mégegyszer megemlítendő, hogy fontos az Xtv. 1. §-a, Ytv. 1. §-a, és Ztv. 1337. §-a.",
), )),
Article(
identifier="4",
children=(Paragraph(
intro=
"Az Ytv. 12. § (8) bekezdése helyébe a következő rendelkezés lép:",
children=(BlockAmendmentContainer(children=(Paragraph(
identifier='8',
text="Beillesztett referencia: 12. §, vajon lesz baj?"
), ), ), ),
), ),
),
),
)
with_semantics_1 = ActSemanticsParser.add_semantics_to_act(TEST_ACT)
assert with_semantics_1.is_semantic_parsed
assert with_semantics_1.article('1').paragraph().act_id_abbreviations == (
ActIdAbbreviation('Xtv.', '2040. évi DX. törvény'), )
assert with_semantics_1.article('2').paragraph(
'1').act_id_abbreviations == (ActIdAbbreviation(
'Ytv.', '2041. évi X. törvény'), )
assert with_semantics_1.article('2').paragraph(
'2').act_id_abbreviations == (ActIdAbbreviation(
'Ztv.', '2041. évi XXX. törvény'), )
assert with_semantics_1.article('3').paragraph().outgoing_references == (
OutgoingReference(start_pos=40,
end_pos=44,
reference=Reference(act='2040. évi DX. törvény')),
OutgoingReference(start_pos=45,
end_pos=51,
reference=Reference(act='2040. évi DX. törvény',
article='1')),
OutgoingReference(start_pos=53,
end_pos=57,
reference=Reference(act='2041. évi X. törvény')),
OutgoingReference(start_pos=58,
end_pos=64,
reference=Reference(act='2041. évi X. törvény',
article='1')),
OutgoingReference(start_pos=69,
end_pos=73,
reference=Reference(act='2041. évi XXX. törvény')),
OutgoingReference(start_pos=74,
end_pos=83,
reference=Reference(act='2041. évi XXX. törvény',
article='1337')),
)
with_semantics_2 = ActSemanticsParser.add_semantics_to_act(
with_semantics_1)
# TODO: with_semantics_2 is with_semantics_1
assert with_semantics_2 == with_semantics_1
modified_paragraph = attr.evolve(
with_semantics_1.article("2").paragraph("1"),
text=
"Bekeverünk a tesztelés Y új tulajdonságiról szóló 2057. évi X. törvény (a továbbiakban Ytv.) dolgaival",
semantic_data=None,
outgoing_references=None,
act_id_abbreviations=None,
)
modified_article = attr.evolve(
with_semantics_1.article("2"),
children=(modified_paragraph, ) +
with_semantics_1.article("2").children[1:],
)
modified_act = attr.evolve(
with_semantics_1,
children=(with_semantics_1.children[0], modified_article,
with_semantics_1.children[2], with_semantics_1.children[3]),
)
assert not modified_act.is_semantic_parsed
modified_with_semantics = ActSemanticsParser.add_semantics_to_act(
modified_act)
assert modified_with_semantics.article('2').paragraph(
'1').act_id_abbreviations == (ActIdAbbreviation(
'Ytv.', '2057. évi X. törvény'), )
assert modified_with_semantics.article('3').paragraph(
).outgoing_references == (
OutgoingReference(start_pos=40,
end_pos=44,
reference=Reference(act='2040. évi DX. törvény')),
OutgoingReference(start_pos=45,
end_pos=51,
reference=Reference(act='2040. évi DX. törvény',
article='1')),
OutgoingReference(start_pos=53,
end_pos=57,
reference=Reference(act='2057. évi X. törvény')),
OutgoingReference(start_pos=58,
end_pos=64,
reference=Reference(act='2057. évi X. törvény',
article='1')),
OutgoingReference(start_pos=69,
end_pos=73,
reference=Reference(act='2041. évi XXX. törvény')),
OutgoingReference(start_pos=74,
end_pos=83,
reference=Reference(act='2041. évi XXX. törvény',
article='1337')),
)
assert modified_with_semantics.article('4').paragraph().semantic_data == (
BlockAmendment(position=Reference(act='2057. évi X. törvény',
article='12',
paragraph='8'), ), )
assert with_semantics_1.article('1') is modified_with_semantics.article(
'1')
# Note that because of the abbreviation change, everything else may be reparsed,
# so no asserts for e.g. article('3')
# No need to reparse BlockAmendments though
a4_children = with_semantics_1.article('4').paragraph().children
modified_a4_children = modified_with_semantics.article(
'4').paragraph().children
assert a4_children is not None
assert modified_a4_children is not None
assert a4_children[0] is modified_a4_children[0]
def test_semantic_reparse_simple() -> None:
TEST_ACT = Act(
identifier="2050. évi XD. törvény",
publication_date=Date(2050, 3, 4),
subject="A nyelvtani tesztelésről",
preamble='',
children=(
Article(
identifier="1",
children=(Paragraph(
text=
"Fontos lesz később a tesztelés X tulajdonságiról szóló 2040. évi DX. törvény (a továbbiakban Xtv.) rövidítésének feloldása.",
), )),
Article(
identifier="2",
children=(
Paragraph(
identifier="1",
text=
"Bekeverünk a tesztelés Y tulajdonságiról szóló 2041. évi X. törvény dolgaival.",
),
Paragraph(
identifier="2",
text=
"Itt megemlítendő a 1. § és a tesztelés Z tulajdonságiról szóló 2041. évi XXX. törvény (a továbbiakban Ztv.) 1. §-a közötti különbség",
),
Paragraph(
identifier="3",
intro="Az Xtv.",
wrap_up="szöveg lép.",
children=(
AlphabeticPoint(
identifier="a",
text=
"12. § (7) bekezdésében a „fontatlan” szövegrész helyébe a „nem fontos”,",
),
AlphabeticPoint(
identifier="b",
text=
"11. §-ben a „nemigazán” szövegrész helyébe a „nem”",
),
),
),
),
),
Article(
identifier="3",
children=(Paragraph(
text=
"Ez a törvény a kihirdetését követő napon lép hatályba."),
)),
Article(
identifier="4",
children=(Paragraph(
intro=
"A Ztv. 12. § (8) bekezdése helyébe a következő rendelkezés lép:",
children=(BlockAmendmentContainer(children=(Paragraph(
identifier='8',
text="Beillesztett referencia: 11. §, vajon lesz baj?"
), ), ), ),
), ),
),
),
)
with_semantics_1 = ActSemanticsParser.add_semantics_to_act(TEST_ACT)
assert with_semantics_1.is_semantic_parsed
assert with_semantics_1.article('1').paragraph().act_id_abbreviations == (
ActIdAbbreviation('Xtv.', '2040. évi DX. törvény'), )
assert with_semantics_1.article('2').paragraph(
'2').act_id_abbreviations == (ActIdAbbreviation(
'Ztv.', '2041. évi XXX. törvény'), )
assert with_semantics_1.article('1').paragraph().outgoing_references == (
OutgoingReference(start_pos=55,
end_pos=76,
reference=Reference(act='2040. évi DX. törvény')), )
assert with_semantics_1.article('2').paragraph(
'1').outgoing_references == (OutgoingReference(
start_pos=47,
end_pos=67,
reference=Reference(act='2041. évi X. törvény')), )
assert with_semantics_1.article('2').paragraph('3').point(
'a').semantic_data == (TextAmendment(position=Reference(
act='2040. évi DX. törvény', article='12', paragraph='7'),
original_text='fontatlan',
replacement_text='nem fontos'), )
assert with_semantics_1.article('3').paragraph().semantic_data == (
EnforcementDate(position=None, date=DaysAfterPublication(days=1)), )
assert with_semantics_1.article('4').paragraph().semantic_data == (
BlockAmendment(position=Reference(act='2041. évi XXX. törvény',
article='12',
paragraph='8'), ), )
a4_children = with_semantics_1.article('4').paragraph().children
assert a4_children is not None
the_block_amendment = a4_children[0]
assert isinstance(the_block_amendment, BlockAmendmentContainer)
assert the_block_amendment.semantic_data is None
assert the_block_amendment.outgoing_references is None
assert the_block_amendment.act_id_abbreviations is None
with_semantics_2 = ActSemanticsParser.add_semantics_to_act(
with_semantics_1)
assert with_semantics_2 is with_semantics_1
modified_paragraph = attr.evolve(
with_semantics_1.article("2").paragraph("1"),
text="Az 1. § és 3. § egészen fontos.",
semantic_data=None,
outgoing_references=None,
act_id_abbreviations=None,
)
modified_article = attr.evolve(
with_semantics_1.article("2"),
children=(modified_paragraph, ) +
with_semantics_1.article("2").children[1:],
)
modified_act = attr.evolve(
with_semantics_1,
children=(with_semantics_1.children[0], modified_article,
with_semantics_1.children[2], with_semantics_1.children[3]),
)
assert modified_act.article('1').is_semantic_parsed
assert not modified_act.article('2').is_semantic_parsed
assert modified_act.article('3').is_semantic_parsed
assert modified_act.article('4').is_semantic_parsed
assert not modified_act.is_semantic_parsed
modified_with_semantics = ActSemanticsParser.add_semantics_to_act(
modified_act)
assert modified_with_semantics.is_semantic_parsed
assert modified_with_semantics.article('2').paragraph(
'1').outgoing_references == (
OutgoingReference(start_pos=3,
end_pos=7,
reference=Reference(act=None, article='1')),
OutgoingReference(start_pos=11,
end_pos=15,
reference=Reference(act=None, article='3')),
)
# Check if nothing else was touched but the modified part.
assert with_semantics_1.article('1') is modified_with_semantics.article(
'1')
assert with_semantics_1.article('2').paragraph(
'2') is modified_with_semantics.article('2').paragraph('2')
assert with_semantics_1.article('2').paragraph(
'3') is modified_with_semantics.article('2').paragraph('3')
assert with_semantics_1.article('3') is modified_with_semantics.article(
'3')
assert with_semantics_1.article('4') is modified_with_semantics.article(
'4')
]
CASES_WITH_POSITIONS: List[Tuple[str, Tuple[Tuple[
Reference, OutgoingReference], ...]]] = [
("""
1. § A tesztelésről szóló 2345. évi I. törvény
a) 8. §
aa) (5) bekezdése,
ab) (6) bekezdés a) pontja,
b) 10. §-a, és
c) egy totál másik dolog 1. § c) pontja, még utána ezzel szöveggel
jól van feldolgozva.
""", (
(
ref(None, "1"),
OutgoingReference(21, 41, ref("2345. évi I. törvény")),
),
(
ref(None, "1", None, "a"),
OutgoingReference(0, 4, ref("2345. évi I. törvény", "8")),
),
(
ref(None, "1", None, "a", "aa"),
OutgoingReference(0, 13, ref("2345. évi I. törvény", "8",
"5")),
),
(
ref(None, "1", None, "a", "ab"),
OutgoingReference(0, 22,
ref("2345. évi I. törvény", "8", "6", "a")),
),