"""Sample annotations as test fixtures for the v3 annotation schema.

These samples demonstrate correct annotation format and serve as regression
fixtures for the validator. They cover representative document types and
complexity levels from the Gold Corpus.

Schema version: 1.0.0
"""

from __future__ import annotations

from datetime import datetime, timezone

from services.intelligence_pipeline_v3.schemas.annotations import (
    AmbiguityMarker,
    AmbiguityType,
    AnnotatedDocument,
    AnnotationMetadata,
    CompanySentimentAnnotation,
    DirectEffect,
    EntityAnnotation,
    EntityType,
    EventAnnotation,
    EventClass,
    EvidenceSpanAnnotation,
    InferredExposure,
    NumericFactAnnotation,
    PeriodAnnotation,
    PeriodType,
    RelationAnnotation,
    RelationType,
    SentimentLabel,
)

# ---------------------------------------------------------------------------
# Sample 1: Simple earnings beat article (single company, fast path)
# ---------------------------------------------------------------------------

_EARNINGS_TEXT = (
    "Apple Inc. reported quarterly earnings of $1.52 per share, "
    "beating the consensus estimate of $1.43 by $0.09. "
    "Revenue came in at $94.9 billion, above expectations of $92.1 billion. "
    "The company raised its dividend by 4% to $0.26 per share."
)


def build_sample_earnings_beat() -> AnnotatedDocument:
    """Single-company earnings beat with numeric facts and clear sentiment."""
    ev_apple = EvidenceSpanAnnotation(
        id="ev-001",
        start_char=0,
        end_char=10,
        text="Apple Inc.",
    )
    ev_eps = EvidenceSpanAnnotation(
        id="ev-002",
        start_char=11,
        end_char=108,
        text="reported quarterly earnings of $1.52 per share, beating the consensus estimate of $1.43 by $0.09.",
    )
    ev_revenue = EvidenceSpanAnnotation(
        id="ev-003",
        start_char=109,
        end_char=179,
        text="Revenue came in at $94.9 billion, above expectations of $92.1 billion.",
    )
    ev_dividend = EvidenceSpanAnnotation(
        id="ev-004",
        start_char=180,
        end_char=237,
        text="The company raised its dividend by 4% to $0.26 per share.",
    )

    entity_apple = EntityAnnotation(
        id="ent-001",
        entity_type=EntityType.COMPANY,
        literal_text="Apple Inc.",
        canonical_id="aapl-uuid",
        canonical_name="AAPL",
        evidence_ids=["ev-001"],
        confidence=1.0,
        derivation="deterministic",
    )

    event_beat = EventAnnotation(
        id="evt-001",
        event_class=EventClass.EARNINGS_BEAT,
        description="Apple Q1 FY2025 earnings beat consensus by $0.09/share",
        primary_company_ids=["ent-001"],
        evidence_ids=["ev-002"],
        confidence=0.98,
        derivation="specialist",
    )

    event_dividend = EventAnnotation(
        id="evt-002",
        event_class=EventClass.DIVIDEND_CHANGE,
        description="Apple raises dividend by 4%",
        primary_company_ids=["ent-001"],
        evidence_ids=["ev-004"],
        confidence=0.95,
        derivation="specialist",
    )

    fact_eps = NumericFactAnnotation(
        id="fact-001",
        fact_type="eps",
        subject_entity_id="ent-001",
        predicate="reported",
        literal_value="$1.52 per share",
        normalized_value=1.52,
        unit="USD",
        period=PeriodAnnotation(
            period_type=PeriodType.FISCAL_QUARTER,
            fiscal_year=2025,
            fiscal_quarter=1,
            literal_text="quarterly",
        ),
        evidence_ids=["ev-002"],
        confidence=0.99,
        derivation="deterministic",
    )

    fact_revenue = NumericFactAnnotation(
        id="fact-002",
        fact_type="revenue",
        subject_entity_id="ent-001",
        predicate="reported",
        literal_value="$94.9 billion",
        normalized_value=94_900_000_000,
        unit="USD",
        evidence_ids=["ev-003"],
        confidence=0.99,
        derivation="deterministic",
    )

    sentiment = CompanySentimentAnnotation(
        id="sent-001",
        company_entity_id="ent-001",
        label=SentimentLabel.POSITIVE,
        positive_probability=0.88,
        negative_probability=0.04,
        neutral_probability=0.08,
        evidence_ids=["ev-002", "ev-003", "ev-004"],
        confidence=0.92,
        derivation="specialist",
    )

    direct = DirectEffect(
        event_id="evt-001",
        company_entity_id="ent-001",
        evidence_ids=["ev-002"],
        confidence=0.98,
    )

    return AnnotatedDocument(
        document_id="doc-sample-001",
        document_type="article",
        source_text=_EARNINGS_TEXT,
        metadata=AnnotationMetadata(
            schema_version="1.0.0",
            annotator_id="gold-annotator-1",
            annotation_date=datetime(2025, 1, 15, tzinfo=timezone.utc),
            review_status="gold",
            reviewer_id="senior-reviewer-1",
            review_date=datetime(2025, 1, 16, tzinfo=timezone.utc),
        ),
        evidence_spans=[ev_apple, ev_eps, ev_revenue, ev_dividend],
        entities=[entity_apple],
        events=[event_beat, event_dividend],
        relations=[],
        numeric_facts=[fact_eps, fact_revenue],
        sentiments=[sentiment],
        direct_effects=[direct],
        inferred_exposures=[],
        ambiguity_markers=[],
    )


# ---------------------------------------------------------------------------
# Sample 2: Multi-company competitive article (requires adjudication)
# ---------------------------------------------------------------------------

_MULTI_COMPANY_TEXT = (
    "Microsoft announced a $10 billion investment in OpenAI, "
    "intensifying competition with Google in the AI space. "
    "Analysts expect this deal to pressure Alphabet's cloud revenue growth, "
    "though some see it as validation of the broader AI investment thesis."
)


def build_sample_multi_company_competitive() -> AnnotatedDocument:
    """Multi-company article with competing sentiments and inferred exposure."""
    ev_msft = EvidenceSpanAnnotation(
        id="ev-101",
        start_char=0,
        end_char=9,
        text="Microsoft",
    )
    ev_deal = EvidenceSpanAnnotation(
        id="ev-102",
        start_char=10,
        end_char=55,
        text="announced a $10 billion investment in OpenAI,",
    )
    ev_competition = EvidenceSpanAnnotation(
        id="ev-103",
        start_char=56,
        end_char=109,
        text="intensifying competition with Google in the AI space.",
    )
    ev_pressure = EvidenceSpanAnnotation(
        id="ev-104",
        start_char=110,
        end_char=180,
        text="Analysts expect this deal to pressure Alphabet's cloud revenue growth,",
    )
    ev_validation = EvidenceSpanAnnotation(
        id="ev-105",
        start_char=181,
        end_char=250,
        text="though some see it as validation of the broader AI investment thesis.",
    )

    ent_msft = EntityAnnotation(
        id="ent-101",
        entity_type=EntityType.COMPANY,
        literal_text="Microsoft",
        canonical_id="msft-uuid",
        canonical_name="MSFT",
        evidence_ids=["ev-101"],
        confidence=1.0,
        derivation="deterministic",
    )
    ent_goog = EntityAnnotation(
        id="ent-102",
        entity_type=EntityType.COMPANY,
        literal_text="Google",
        canonical_id="googl-uuid",
        canonical_name="GOOGL",
        evidence_ids=["ev-103"],
        confidence=0.98,
        derivation="deterministic",
    )
    ent_alphabet = EntityAnnotation(
        id="ent-103",
        entity_type=EntityType.COMPANY,
        literal_text="Alphabet",
        canonical_id="googl-uuid",
        canonical_name="GOOGL",
        evidence_ids=["ev-104"],
        confidence=0.97,
        derivation="specialist",
    )

    event_ma = EventAnnotation(
        id="evt-101",
        event_class=EventClass.MA_ANNOUNCEMENT,
        description="Microsoft $10B investment in OpenAI",
        primary_company_ids=["ent-101"],
        evidence_ids=["ev-102"],
        confidence=0.96,
        derivation="specialist",
    )

    rel_competes = RelationAnnotation(
        id="rel-101",
        relation_type=RelationType.COMPETES_WITH,
        source_id="ent-101",
        target_id="ent-102",
        evidence_ids=["ev-103"],
        confidence=0.90,
        derivation="specialist",
    )

    fact_amount = NumericFactAnnotation(
        id="fact-101",
        fact_type="investment_amount",
        subject_entity_id="ent-101",
        predicate="invested",
        literal_value="$10 billion",
        normalized_value=10_000_000_000,
        unit="USD",
        evidence_ids=["ev-102"],
        confidence=0.99,
        derivation="deterministic",
    )

    sentiment_msft = CompanySentimentAnnotation(
        id="sent-101",
        company_entity_id="ent-101",
        label=SentimentLabel.POSITIVE,
        positive_probability=0.75,
        negative_probability=0.05,
        neutral_probability=0.20,
        evidence_ids=["ev-102"],
        confidence=0.85,
        derivation="specialist",
    )

    sentiment_goog = CompanySentimentAnnotation(
        id="sent-102",
        company_entity_id="ent-102",
        label=SentimentLabel.MIXED,
        positive_probability=0.30,
        negative_probability=0.45,
        neutral_probability=0.25,
        evidence_ids=["ev-103", "ev-104", "ev-105"],
        confidence=0.70,
        derivation="specialist",
    )

    direct_msft = DirectEffect(
        event_id="evt-101",
        company_entity_id="ent-101",
        evidence_ids=["ev-102"],
        confidence=0.96,
    )

    inferred_goog = InferredExposure(
        event_id="evt-101",
        company_entity_id="ent-102",
        reasoning="Competitive pressure from Microsoft's AI investment threatens Google's cloud market share",
        evidence_ids=["ev-103", "ev-104"],
        confidence=0.72,
    )

    ambiguity = AmbiguityMarker(
        ambiguity_type=AmbiguityType.CONFLICTING_SENTIMENT,
        description="Alphabet sentiment is mixed — competitive pressure vs. AI thesis validation",
        affected_entity_ids=["ent-102", "ent-103"],
        severity="medium",
    )

    return AnnotatedDocument(
        document_id="doc-sample-002",
        document_type="article",
        source_text=_MULTI_COMPANY_TEXT,
        metadata=AnnotationMetadata(
            schema_version="1.0.0",
            annotator_id="gold-annotator-2",
            annotation_date=datetime(2025, 1, 20, tzinfo=timezone.utc),
            review_status="gold",
            reviewer_id="senior-reviewer-1",
            review_date=datetime(2025, 1, 21, tzinfo=timezone.utc),
        ),
        evidence_spans=[ev_msft, ev_deal, ev_competition, ev_pressure, ev_validation],
        entities=[ent_msft, ent_goog, ent_alphabet],
        events=[event_ma],
        relations=[rel_competes],
        numeric_facts=[fact_amount],
        sentiments=[sentiment_msft, sentiment_goog],
        direct_effects=[direct_msft],
        inferred_exposures=[inferred_goog],
        ambiguity_markers=[ambiguity],
    )


# ---------------------------------------------------------------------------
# Sample 3: Macro event with inferred sector exposure
# ---------------------------------------------------------------------------

_MACRO_TEXT = (
    "The Federal Reserve raised interest rates by 25 basis points to 5.50%, "
    "citing persistent inflation concerns. Markets sold off broadly, "
    "with technology stocks leading the decline."
)


def build_sample_macro_event() -> AnnotatedDocument:
    """Macro event with sector-level inferred exposure and no single primary company."""
    ev_rate = EvidenceSpanAnnotation(
        id="ev-202",
        start_char=0,
        end_char=70,
        text="The Federal Reserve raised interest rates by 25 basis points to 5.50%,",
    )
    ev_inflation = EvidenceSpanAnnotation(
        id="ev-203",
        start_char=71,
        end_char=108,
        text="citing persistent inflation concerns.",
    )
    ev_selloff = EvidenceSpanAnnotation(
        id="ev-204",
        start_char=109,
        end_char=178,
        text="Markets sold off broadly, with technology stocks leading the decline.",
    )

    ent_fed = EntityAnnotation(
        id="ent-201",
        entity_type=EntityType.COMPANY,
        literal_text="The Federal Reserve",
        canonical_id=None,
        canonical_name="Federal Reserve",
        evidence_ids=["ev-202"],
        confidence=1.0,
        derivation="deterministic",
    )

    event_macro = EventAnnotation(
        id="evt-201",
        event_class=EventClass.MACRO_EVENT,
        description="Fed raises rates 25bps to 5.50%",
        primary_company_ids=[],
        evidence_ids=["ev-202", "ev-203", "ev-204"],
        confidence=0.99,
        derivation="deterministic",
    )

    fact_rate = NumericFactAnnotation(
        id="fact-201",
        fact_type="interest_rate_change",
        subject_entity_id="ent-201",
        predicate="raised_by",
        literal_value="25 basis points",
        normalized_value=0.25,
        unit="percentage_points",
        evidence_ids=["ev-202"],
        confidence=0.99,
        derivation="deterministic",
    )

    fact_level = NumericFactAnnotation(
        id="fact-202",
        fact_type="interest_rate_level",
        subject_entity_id="ent-201",
        predicate="to",
        literal_value="5.50%",
        normalized_value=5.50,
        unit="%",
        evidence_ids=["ev-202"],
        confidence=0.99,
        derivation="deterministic",
    )

    return AnnotatedDocument(
        document_id="doc-sample-003",
        document_type="macro_event",
        source_text=_MACRO_TEXT,
        metadata=AnnotationMetadata(
            schema_version="1.0.0",
            annotator_id="gold-annotator-1",
            annotation_date=datetime(2025, 2, 1, tzinfo=timezone.utc),
            review_status="gold",
        ),
        evidence_spans=[ev_rate, ev_inflation, ev_selloff],
        entities=[ent_fed],
        events=[event_macro],
        relations=[],
        numeric_facts=[fact_rate, fact_level],
        sentiments=[],
        direct_effects=[],
        inferred_exposures=[],
        ambiguity_markers=[],
    )


# All sample builders for easy iteration
SAMPLE_BUILDERS = [
    build_sample_earnings_beat,
    build_sample_multi_company_competitive,
    build_sample_macro_event,
]