Интегрирован wesp в сайт

WESP_REL/app/lab/calc/__init__.py

@@ -0,0 +1,4 @@
+from .engine import calculate_ration
+from .diff import compare_master_execution
+
+__all__ = ["calculate_ration", "compare_master_execution"]

WESP_REL/app/lab/calc/compound.py

@@ -0,0 +1,112 @@
+from __future__ import annotations
+
+from typing import Any
+
+from app.lab.calc.derived import apply_content_derived
+from app.lab.calc.nutrients import norm_diff, weighted_average
+from app.lab.indicators import RATION_ALL_INDICATORS
+
+
+def _compound_active(lines: list[dict[str, Any]]) -> list[dict[str, Any]]:
+    return [
+        l
+        for l in lines
+        if l.get("in_compound") and (l.get("daily_kg") or 0) > 0
+    ]
+
+
+def _sum_kg(lines: list[dict[str, Any]]) -> float:
+    return sum(float(l.get("daily_kg") or 0) for l in lines)
+
+
+def _compound_indicators(
+    lines: list[dict[str, Any]],
+    total_kg: float,
+    norms: dict[str, dict[str, float | None]],
+) -> list[dict[str, Any]]:
+    content_by_key: dict[str, float | None] = {}
+    for defn in RATION_ALL_INDICATORS:
+        if defn.get("derived"):
+            continue
+        content_by_key[defn["key"]] = weighted_average(
+            lines,
+            total_kg,
+            defn["nutrient_keys"],
+            indicator_key=defn.get("key"),
+        )
+    for defn in RATION_ALL_INDICATORS:
+        if not defn.get("derived"):
+            continue
+        content_by_key[defn["key"]] = apply_content_derived(
+            content_by_key,
+            defn,
+            compound_mode=True,
+        )
+    rows = []
+    for defn in RATION_ALL_INDICATORS:
+        key = defn["key"]
+        content = content_by_key.get(key)
+        bounds = norms.get(key, {})
+        min_v = bounds.get("min")
+        max_v = bounds.get("max")
+        diff = norm_diff(content, min_v, max_v)
+        if content is None and min_v is None and max_v is None:
+            continue
+        rows.append(
+            {
+                "key": key,
+                "label": defn["label"],
+                "unit": defn["unit"],
+                "min": min_v,
+                "max": max_v,
+                "content": content,
+                "diff": diff,
+            }
+        )
+    return rows
+
+
+def calculate_compound_feed(
+    lines: list[dict[str, Any]],
+    norms: dict[str, dict[str, float | None]] | None = None,
+    *,
+    profile_mass_kg: float | None = None,
+    heads_per_trip: int = 1,
+) -> dict[str, Any] | None:
+    del profile_mass_kg, heads_per_trip
+    active = _compound_active(lines)
+    if not active:
+        return None
+    total_kg = _sum_kg(active)
+    cost = None
+    any_cost = False
+    for line in active:
+        kg = line.get("daily_kg")
+        price = line.get("price_per_kg")
+        if kg is None or price is None:
+            continue
+        cost = (cost or 0) + float(kg) * float(price)
+        any_cost = True
+    return {
+        "totals": [
+            {"key": "compound_kg", "label": "Масса комбикорма, кг", "value": total_kg},
+            {
+                "key": "compound_cost",
+                "label": "Стоимость комбикорма",
+                "value": cost if any_cost else None,
+            },
+        ],
+        "indicators": _compound_indicators(active, total_kg, norms or {}),
+        "lines": [
+            {
+                "ingredient_name": line.get("ingredient_name") or "—",
+                "daily_kg": line.get("daily_kg"),
+                "share_pct": (
+                    (float(line["daily_kg"]) / total_kg) * 100
+                    if total_kg > 0 and line.get("daily_kg") is not None
+                    else None
+                ),
+            }
+            for line in active
+        ],
+    }

WESP_REL/app/lab/calc/derived.py

@@ -0,0 +1,64 @@
+from __future__ import annotations
+
+from typing import Any
+
+from app.lab.calc.nutrients import rnb
+
+
+def apply_content_derived(
+    content_by_key: dict[str, float | None],
+    defn: dict[str, Any],
+    *,
+    total_kg: float = 0,
+    heads_per_trip: int = 1,
+    profile_mass_kg: float | None = None,
+    compound_mode: bool = False,
+) -> float | None:
+    derived = defn.get("derived")
+    if derived == "alias":
+        return content_by_key.get(defn.get("alias_of"))
+    if derived == "pct_of_dm":
+        src = content_by_key.get(defn.get("from_key"))
+        dm = content_by_key.get("dry_matter")
+        if src is None or not dm or dm <= 0:
+            return None
+        return src / dm * 100.0
+    if derived == "g_per_kg_dm":
+        src = content_by_key.get(defn.get("from_key"))
+        dm = content_by_key.get("dry_matter")
+        if src is None or not dm or dm <= 0:
+            return None
+        if compound_mode:
+            return src
+        return src / (dm / 1000.0)
+    if derived == "nel_per_kg_dm":
+        dm = content_by_key.get("dry_matter")
+        nel = content_by_key.get("nel")
+        if not dm or dm <= 0 or nel is None:
+            return None
+        if compound_mode:
+            return nel
+        return nel / (dm / 1000.0)
+    if derived == "ratio":
+        num = content_by_key.get(defn.get("ratio_num"))
+        den = content_by_key.get(defn.get("ratio_den"))
+        if num is None or den is None or den == 0:
+            return None
+        return num / den
+    if derived == "dm_pct_bw":
+        dm = content_by_key.get("dry_matter")
+        if dm is None or not profile_mass_kg or profile_mass_kg <= 0:
+            return None
+        return (dm / 1000.0 / profile_mass_kg) * 100.0
+    if derived == "ration_pct_bw":
+        if not profile_mass_kg or profile_mass_kg <= 0 or total_kg <= 0:
+            return None
+        heads = max(int(heads_per_trip or 1), 1)
+        kg_per_head = total_kg / heads
+        return (kg_per_head / profile_mass_kg) * 100.0
+    if derived in ("rnb", "bra_rnb"):
+        return rnb(
+            content_by_key.get("crude_protein"),
+            content_by_key.get("usp"),
+        )
+    return None

WESP_REL/app/lab/calc/diff.py

@@ -0,0 +1,53 @@
+from __future__ import annotations
+
+from typing import Any
+
+from app.lab.constants import DIFF_TOLERANCE_KG
+
+
+def compare_master_execution(
+    master_lines: list[dict[str, Any]],
+    execution_lines: list[dict[str, Any]],
+) -> dict[str, Any]:
+    master_map: dict[str, float] = {}
+    for line in master_lines:
+        if not line.get("in_ration"):
+            continue
+        cid = line.get("component_id")
+        if not cid:
+            continue
+        master_map[str(cid)] = float(line.get("daily_kg") or 0)
+
+    exec_map: dict[str, float] = {}
+    for line in execution_lines:
+        cid = line.get("component_id")
+        if not cid:
+            continue
+        exec_map[str(cid)] = float(line.get("daily_kg_total") or 0)
+
+    all_ids = set(master_map) | set(exec_map)
+    diff_lines = []
+    has_changes = False
+    for cid in sorted(all_ids):
+        m = master_map.get(cid)
+        e = exec_map.get(cid)
+        reasons = []
+        if m is None:
+            reasons.append("missing_in_master")
+            has_changes = True
+        if e is None:
+            reasons.append("missing_in_execution")
+            has_changes = True
+        if m is not None and e is not None and abs(m - e) > DIFF_TOLERANCE_KG:
+            reasons.append("kg_mismatch")
+            has_changes = True
+        if reasons:
+            diff_lines.append(
+                {
+                    "component_id": cid,
+                    "master_kg": m,
+                    "execution_kg": e,
+                    "reasons": reasons,
+                }
+            )
+    return {"has_changes": has_changes, "lines": diff_lines}

WESP_REL/app/lab/calc/engine.py

@@ -0,0 +1,184 @@
+from __future__ import annotations
+
+from datetime import datetime, timezone
+from typing import Any
+
+from app.lab.calc.compound import calculate_compound_feed
+from app.lab.calc.derived import apply_content_derived
+from app.lab.calc.nutrients import daily_intake_total, get_nutrient_value, norm_diff, weighted_average
+from app.lab.constants import RATION_TOTAL_KEYS
+from app.lab.indicators import RATION_ALL_INDICATORS
+
+
+def _active_lines(lines: list[dict[str, Any]]) -> list[dict[str, Any]]:
+    return [
+        l
+        for l in lines
+        if l.get("in_ration") and (l.get("daily_kg") or 0) > 0
+    ]
+
+
+def _sum_kg(lines: list[dict[str, Any]]) -> float:
+    return sum(float(l.get("daily_kg") or 0) for l in lines)
+
+
+def _sum_cost(lines: list[dict[str, Any]]) -> float | None:
+    total = 0.0
+    any_cost = False
+    for line in lines:
+        kg = line.get("daily_kg")
+        price = line.get("price_per_kg")
+        if kg is None or price is None:
+            continue
+        total += float(kg) * float(price)
+        any_cost = True
+    return total if any_cost else None
+
+
+def _sum_ration_percent(lines: list[dict[str, Any]], total_kg: float) -> float | None:
+    if total_kg <= 0:
+        return None
+    return sum((float(l.get("daily_kg") or 0) / total_kg) * 100 for l in lines)
+
+
+def _compute_totals(
+    ration_type: str,
+    active: list[dict[str, Any]],
+    total_kg: float,
+) -> list[dict[str, Any]]:
+    defs = RATION_TOTAL_KEYS.get(ration_type, RATION_TOTAL_KEYS["BEEF"])
+    ration_kg = _sum_kg(active)
+    values = {
+        "total_kg": total_kg if total_kg > 0 else None,
+        "ration_kg": ration_kg if ration_kg > 0 else None,
+        "ration_pct_sum": _sum_ration_percent(active, total_kg),
+        "cost_total": _sum_cost(active),
+    }
+    return [{"key": d["key"], "label": d["label"], "value": values.get(d["key"])} for d in defs]
+
+
+def _indicator_content(
+    defn: dict[str, Any],
+    active: list[dict[str, Any]],
+    total_kg: float,
+    *,
+    heads_per_trip: int,
+) -> float | None:
+    key = defn.get("key")
+    if defn.get("aggregation") == "weighted_avg":
+        return weighted_average(
+            active,
+            total_kg,
+            defn["nutrient_keys"],
+            indicator_key=key,
+        )
+    return daily_intake_total(
+        active,
+        defn["nutrient_keys"],
+        heads_per_trip=heads_per_trip,
+        indicator_key=key,
+    )
+
+
+def _compute_indicators(
+    active: list[dict[str, Any]],
+    total_kg: float,
+    norms: dict[str, dict[str, float | None]],
+    *,
+    heads_per_trip: int = 1,
+    profile_mass_kg: float | None = None,
+) -> list[dict[str, Any]]:
+    content_by_key: dict[str, float | None] = {}
+    for defn in RATION_ALL_INDICATORS:
+        if defn.get("derived"):
+            continue
+        content_by_key[defn["key"]] = _indicator_content(
+            defn, active, total_kg, heads_per_trip=heads_per_trip
+        )
+
+    for defn in RATION_ALL_INDICATORS:
+        if not defn.get("derived"):
+            continue
+        content_by_key[defn["key"]] = apply_content_derived(
+            content_by_key,
+            defn,
+            total_kg=total_kg,
+            heads_per_trip=heads_per_trip,
+            profile_mass_kg=profile_mass_kg,
+        )
+
+    rows = []
+    for defn in RATION_ALL_INDICATORS:
+        key = defn["key"]
+        content = content_by_key.get(key)
+        bounds = norms.get(key, {})
+        min_v = bounds.get("min")
+        max_v = bounds.get("max")
+        diff = norm_diff(content, min_v, max_v)
+        if content is None and min_v is None and max_v is None:
+            continue
+        rows.append(
+            {
+                "key": key,
+                "label": defn["label"],
+                "unit": defn["unit"],
+                "min": min_v,
+                "max": max_v,
+                "content": content,
+                "diff": diff,
+            }
+        )
+    return rows
+
+
+def _missing_nutrient_warnings(active: list[dict[str, Any]]) -> list[str]:
+    warnings: list[str] = []
+    for line in active:
+        nutrients = line.get("nutrients") or {}
+        cp = get_nutrient_value(
+            line.get("dry_matter"),
+            nutrients,
+            ["Сыр. Протеин"],
+            indicator_key="crude_protein",
+        )
+        if cp is not None:
+            continue
+        name = line.get("ingredient_name") or line.get("component_id") or "?"
+        warnings.append(f"nutrients_missing:{name}")
+    return warnings
+
+
+def calculate_ration(
+    ration_type: str,
+    lines: list[dict[str, Any]],
+    norms: dict[str, dict[str, float | None]] | None = None,
+    *,
+    heads_per_trip: int = 1,
+    profile_mass_kg: float | None = None,
+) -> dict[str, Any]:
+    norms = norms or {}
+    errors: list[str] = []
+    active = _active_lines(lines)
+    total_kg = _sum_kg(active)
+    heads = max(int(heads_per_trip or 1), 1)
+    if not active:
+        errors.append("Нет строк сырья «в рационе» с дозировкой кг/день")
+    warnings = _missing_nutrient_warnings(active)
+    compound = calculate_compound_feed(
+        lines, norms, profile_mass_kg=profile_mass_kg, heads_per_trip=heads
+    )
+    return {
+        "calculated_at": datetime.now(timezone.utc).isoformat(),
+        "engine": "native",
+        "totals": _compute_totals(ration_type, active, total_kg),
+        "indicators": _compute_indicators(
+            active,
+            total_kg,
+            norms,
+            heads_per_trip=heads,
+            profile_mass_kg=profile_mass_kg,
+        ),
+        "compound": compound,
+        "errors": errors,
+        "warnings": warnings,
+    }

WESP_REL/app/lab/calc/feed_groups.py

@@ -0,0 +1,211 @@
+"""Группы кормов для авторациона — канонический тип компонента + legacy-маппинг."""
+
+from __future__ import annotations
+
+from typing import Any
+
+from app.models import Component
+
+FEED_GROUPS: tuple[dict[str, Any], ...] = (
+    {
+        "id": "rough",
+        "label": "База — грубые",
+        "shortLabel": "База",
+        "required": True,
+        "minPick": 1,
+        "hint": "Без каркаса партия не взлетит. Я не бизнесмен — я специалист.",
+        "step": 1,
+    },
+    {
+        "id": "succulent",
+        "label": "Влага — сочные",
+        "shortLabel": "Влага",
+        "required": False,
+        "minPick": 0,
+        "hint": "Сочное сырьё. Можно не мешать — но чистота пострадает.",
+        "step": 2,
+    },
+    {
+        "id": "concentrate",
+        "label": "Энергия — концентраты",
+        "shortLabel": "Энергия",
+        "required": False,
+        "minPick": 0,
+        "hint": "Концентрат дозируй как реагент — точно. Держись подальше от моей территории.",
+        "step": 3,
+    },
+    {
+        "id": "other",
+        "label": "Добавки",
+        "shortLabel": "Добавки",
+        "required": False,
+        "minPick": 0,
+        "hint": "Минералы и премикс. Необязательно. Но Хайзенберг бы не пропустил.",
+        "step": 4,
+    },
+)
+
+# Канонические значения component.type (выбор в /components)
+FEED_COMPONENT_TYPES: tuple[dict[str, str], ...] = (
+    {
+        "value": "Грубые корма",
+        "feedGroup": "rough",
+        "description": "Сено, солома.",
+    },
+    {
+        "value": "Сочные корма",
+        "feedGroup": "succulent",
+        "description": "Силос, корнеплоды.",
+    },
+    {
+        "value": "Концентрированные",
+        "feedGroup": "concentrate",
+        "description": "Зерно, комбикорм, жмых, шрот",
+    },
+    {
+        "value": "Добавки",
+        "feedGroup": "other",
+        "description": "Премикс, минералы, витамины, КЖП",
+    },
+)
+
+_GROUP_BY_ID = {g["id"]: g for g in FEED_GROUPS}
+
+_LEGACY_TYPE_TO_GROUP: dict[str, str] = {
+    "зерновые": "concentrate",
+    "энергетические": "concentrate",
+    "белковые": "concentrate",
+    "минеральные": "other",
+    "витаминные": "other",
+}
+
+# Старый тип «Объемные корма» — уточните до Грубые/Сочные; эвристика по имени
+_ROUGH_NAME_KEYS = ("солом", "сено", "hay", "straw")
+_SUCCULENT_NAME_KEYS = ("силос", "сенаж", "сочн", "корнеплод", "свекл", "морков", "тыкв", "зелен")
+_CONCENTRATE_NAME_KEYS = ("зерн", "концентр", "комбикорм", "комбик", "жмых", "шрот", "дробин", "пивн")
+
+
+def _norm(text: str | None) -> str:
+    return (text or "").strip().lower()
+
+
+def _canonical_type_map() -> dict[str, str]:
+    return {_norm(t["value"]): t["feedGroup"] for t in FEED_COMPONENT_TYPES}
+
+
+def _canonical_values() -> list[str]:
+    return [t["value"] for t in FEED_COMPONENT_TYPES]
+
+
+def list_component_feed_types() -> list[dict[str, str]]:
+    return [
+        {
+            "value": t["value"],
+            "feedGroup": t["feedGroup"],
+            "description": t["description"],
+            "groupLabel": _GROUP_BY_ID[t["feedGroup"]]["label"],
+        }
+        for t in FEED_COMPONENT_TYPES
+    ]
+
+
+def is_canonical_feed_type(component_type: str | None) -> bool:
+    return _norm(component_type) in _canonical_type_map()
+
+
+def feed_group_for_type(component_type: str | None) -> str | None:
+    """Группа авторациона по component.type (канон или legacy)."""
+    ctype = _norm(component_type)
+    if not ctype:
+        return None
+    canonical = _canonical_type_map().get(ctype)
+    if canonical:
+        return canonical
+    if ctype in ("объемные корма", "объёмные корма"):
+        return None
+    return _LEGACY_TYPE_TO_GROUP.get(ctype)
+
+
+def classify_feed_group(comp: Component) -> str:
+    """Группа для авторациона: сначала component.type, иначе эвристика по имени (legacy)."""
+    by_type = feed_group_for_type(comp.type)
+    if by_type:
+        return by_type
+
+    name = _norm(comp.name)
+
+    def has_any(keys: tuple[str, ...]) -> bool:
+        return any(k in name for k in keys)
+
+    if has_any(_ROUGH_NAME_KEYS):
+        return "rough"
+    if has_any(_SUCCULENT_NAME_KEYS):
+        return "succulent"
+    if has_any(_CONCENTRATE_NAME_KEYS):
+        return "concentrate"
+    if _norm(comp.type) in ("объемные корма", "объёмные корма"):
+        return "succulent"
+    return "other"
+
+
+def list_feed_groups_api() -> list[dict[str, Any]]:
+    return [
+        {
+            "id": g["id"],
+            "label": g["label"],
+            "shortLabel": g["shortLabel"],
+            "required": g["required"],
+            "minPick": g["minPick"],
+            "hint": g["hint"],
+            "step": g["step"],
+        }
+        for g in FEED_GROUPS
+    ]
+
+
+def parse_group_selections(raw: Any) -> dict[str, list[str]]:
+    if not isinstance(raw, dict):
+        return {}
+    out: dict[str, list[str]] = {}
+    for gid in _GROUP_BY_ID:
+        vals = raw.get(gid) or []
+        if isinstance(vals, list):
+            out[gid] = [str(x) for x in vals if x]
+    return out
+
+
+def flatten_group_selections(selections: dict[str, list[str]]) -> list[str]:
+    seen: list[str] = []
+    for gid in _GROUP_BY_ID:
+        for cid in selections.get(gid) or []:
+            if cid not in seen:
+                seen.append(cid)
+    return seen
+
+
+def validate_group_selections(selections: dict[str, list[str]]) -> list[str]:
+    errors: list[str] = []
+    for g in FEED_GROUPS:
+        gid = g["id"]
+        picked = selections.get(gid) or []
+        if g["required"] and len(picked) < int(g["minPick"]):
+            errors.append(f"{gid}:need_{g['minPick']}")
+    total = len(flatten_group_selections(selections))
+    if total < 3:
+        errors.append("pool:need_3")
+    return errors
+
+
+def triplet_meets_group_rules(
+    triplet_ids: set[str],
+    selections: dict[str, list[str]],
+) -> bool:
+    for g in FEED_GROUPS:
+        if not g["required"]:
+            continue
+        pool = set(selections.get(g["id"]) or [])
+        if not pool:
+            continue
+        if not triplet_ids & pool:
+            return False
+    return True

WESP_REL/app/lab/calc/formulate.py

@@ -0,0 +1,417 @@
+from __future__ import annotations
+
+import itertools
+import time
+from dataclasses import dataclass, field
+from typing import Any
+
+from app.lab.calc.engine import calculate_ration
+from app.lab.calc.feed_groups import (
+    FEED_GROUPS,
+    flatten_group_selections,
+    triplet_meets_group_rules,
+    validate_group_selections,
+)
+from app.lab.calc.feed_groups import classify_feed_group as _classify_feed_group
+from app.lab.calc.formulate_optimize import optimize_shares_for_triplet
+from app.lab.calc.formulate_score import (
+    build_calc_lines,
+    build_triplet_score_model,
+    score_model_shares,
+    violation_score,
+)
+from app.lab.calc.formulate_validate import validate_component, validate_components
+from app.lab.calc.norms_resolver import NormsParams, NormsResolveRequest, normalize_norms_method, resolve_norms
+from app.lab.calc.nutrients import get_nutrient_value
+from app.lab.indicators import RATION_ALL_INDICATORS
+from app.lab.models import LabAnimalProfile
+from app.lab.services.component_nutrients import nutrients_calc_dict_batch
+from app.lab.services.profile_norms import load_norms_dict
+from app.models import Component
+
+
+@dataclass
+class FormulateRequest:
+    profile_id: str
+    candidate_ids: list[str] = field(default_factory=list)
+    group_selections: dict[str, list[str]] = field(default_factory=dict)
+    main_feed_ids: list[str] = field(default_factory=list)  # legacy → rough
+    mass_kg: float | None = None
+    milk_yield_kg: float | None = None
+    heads_per_trip: int | None = None
+    total_kg_per_head: float = 7.3
+    optimize_keys: list[str] = field(default_factory=list)
+    objective: str = "min_cost"
+    cost_weight: float = 100.0
+    grid_step: float = 0.1
+    prefilter_k: int = 18
+    min_share: float = 0.05
+    norms_method: str = "wesp"
+    norms_params: dict[str, Any] = field(default_factory=dict)
+
+
+_DEFAULT_OPTIMIZE_KEYS = (
+    "dry_matter",
+    "usp",
+    "nel",
+    "crude_protein",
+    "rnb",
+    "nfc_pct_dm_uk",
+)
+
+
+def _indicator_def(key: str) -> dict[str, Any] | None:
+    for defn in RATION_ALL_INDICATORS:
+        if defn["key"] == key:
+            return defn
+    return None
+
+
+def _resolve_norms(profile: LabAnimalProfile, req: FormulateRequest) -> tuple[dict, dict]:
+    stored = load_norms_dict(profile.id)
+    mass = req.mass_kg if req.mass_kg is not None else profile.mass_kg
+    milk = req.milk_yield_kg if req.milk_yield_kg is not None else profile.milk_yield_kg
+    from app.lab.services.norms_params import load_norms_params
+
+    method = normalize_norms_method(req.norms_method or profile.norms_method)
+    params = load_norms_params(profile)
+    if req.norms_params:
+        merged = {
+            "milkFatPct": params.milk_fat_pct,
+            "lactationNo": params.lactation_no,
+            "lactationStage": params.lactation_stage,
+            "bodyCondition": params.body_condition,
+            "housingSystem": params.housing_system,
+            "koncOeSv": params.konc_oe_sv,
+        }
+        merged.update(req.norms_params)
+        params = NormsParams.from_dict(merged)
+    resolved, meta = resolve_norms(
+        NormsResolveRequest(
+            method=method,
+            stored=stored,
+            mass_kg=mass,
+            milk_yield_kg=milk,
+            ration_type=profile.ration_type,
+            force_dynamic=method == "wesp",
+            params=params,
+        )
+    )
+    dynamic = meta.get("dynamicNorms") or meta.get("dynamic") or {}
+    return resolved, {"normsMethod": method, "dynamicNorms": dynamic, "normsMeta": meta.get("meta")}
+
+
+def _rough_component_value(
+    comp: Component,
+    optimize_keys: list[str],
+    norms: dict[str, dict[str, float | None]],
+    nutrient_cache: dict[str, dict[str, float]],
+) -> float | None:
+    nutrients = nutrient_cache.get(comp.id, {})
+    dm_pct = comp.dry_matter
+    total = 0.0
+    counted = 0
+    for key in optimize_keys:
+        defn = _indicator_def(key)
+        if defn is None or defn.get("derived"):
+            continue
+        bounds = norms.get(key) or {}
+        target_min = bounds.get("min")
+        target_max = bounds.get("max")
+        if target_min is None and target_max is None:
+            continue
+        target = None
+        if target_min is not None and target_max is not None:
+            target = (float(target_min) + float(target_max)) / 2.0
+        elif target_max is not None:
+            target = float(target_max) * 0.9
+        elif target_min is not None:
+            target = float(target_min) * 1.1
+        if target is None or target == 0:
+            continue
+        val = get_nutrient_value(
+            dm_pct,
+            nutrients,
+            defn.get("nutrient_keys") or [],
+            indicator_key=key,
+        )
+        if val is None:
+            continue
+        rel = (float(val) - target) / abs(target)
+        total += rel * rel
+        counted += 1
+    return total / counted if counted else None
+
+
+def _prefilter_candidates(
+    candidates: list[Component],
+    optimize_keys: list[str],
+    norms: dict[str, dict[str, float | None]],
+    nutrient_cache: dict[str, dict[str, float]],
+    *,
+    prefilter_k: int,
+    must_keep_ids: set[str] | None = None,
+) -> list[Component]:
+    must_keep_ids = must_keep_ids or set()
+    pinned = [c for c in candidates if c.id in must_keep_ids]
+    rest = [c for c in candidates if c.id not in must_keep_ids]
+    slots = max(prefilter_k - len(pinned), 0)
+    if len(candidates) <= prefilter_k:
+        return candidates
+    if slots <= 0:
+        return pinned[:prefilter_k]
+
+    prices = [float(c.price) for c in rest if c.price is not None]
+    median_price = sorted(prices)[len(prices) // 2] if prices else 1.0
+    if median_price <= 0:
+        median_price = 1.0
+
+    scored: list[tuple[float, Component]] = []
+    for comp in rest:
+        price = float(comp.price) if comp.price is not None else median_price
+        cost_part = price / median_price
+        nutrient_part = _rough_component_value(comp, optimize_keys, norms, nutrient_cache)
+        if nutrient_part is None:
+            score = cost_part
+        else:
+            score = 0.4 * cost_part + 0.6 * nutrient_part
+        scored.append((score, comp))
+    scored.sort(key=lambda x: x[0])
+    return pinned + [comp for _, comp in scored[:slots]]
+
+
+def _load_profile(profile_id: str) -> LabAnimalProfile:
+    profile = LabAnimalProfile.query.filter_by(id=profile_id, is_deleted=False).first()
+    if profile is None:
+        raise LookupError("Профиль не найден")
+    return profile
+
+
+def _load_eligible_components(candidate_ids: list[str]) -> tuple[list[Component], list[dict[str, Any]]]:
+    unique = list(dict.fromkeys(candidate_ids))
+    validations = validate_components(unique)
+    ineligible = [v for v in validations if not v["eligible"]]
+    if ineligible:
+        raise ValueError("ineligible_components", ineligible)
+    comps: list[Component] = []
+    for cid in unique:
+        comp = Component.query.filter_by(id=cid, is_deleted=False).first()
+        if comp is None:
+            raise ValueError("component_not_found", cid)
+        comps.append(comp)
+    return comps, validations
+
+
+def _resolve_group_selections(req: FormulateRequest) -> dict[str, list[str]]:
+    selections = dict(req.group_selections or {})
+    if req.main_feed_ids and not selections.get("rough"):
+        selections["rough"] = list(req.main_feed_ids)
+    return selections
+
+
+def formulate(req: FormulateRequest) -> dict[str, Any]:
+    group_selections = _resolve_group_selections(req)
+    candidate_ids = list(req.candidate_ids)
+    if group_selections:
+        pool_errors = validate_group_selections(group_selections)
+        if pool_errors:
+            raise ValueError("group_selection_invalid", pool_errors)
+        candidate_ids = flatten_group_selections(group_selections)
+
+    if len(candidate_ids) < 3:
+        raise ValueError("candidate_ids_min_3")
+    if len(set(candidate_ids)) != len(candidate_ids):
+        raise ValueError("candidate_ids_duplicate")
+
+    profile = _load_profile(req.profile_id)
+    candidates, _ = _load_eligible_components(candidate_ids)
+    nutrient_cache = nutrients_calc_dict_batch(candidate_ids)
+    norms, dynamic = _resolve_norms(profile, req)
+    optimize_keys = req.optimize_keys or list(_DEFAULT_OPTIMIZE_KEYS)
+    heads = max(int(req.heads_per_trip or 10), 1)
+    herd_scale = float(req.total_kg_per_head) * heads
+    profile_mass_kg = req.mass_kg if req.mass_kg is not None else profile.mass_kg
+
+    must_keep: set[str] = set()
+    for g in FEED_GROUPS:
+        if g["required"]:
+            must_keep.update(group_selections.get(g["id"]) or [])
+
+    prefilter_applied = len(candidates) > req.prefilter_k
+    shortlist = _prefilter_candidates(
+        candidates,
+        optimize_keys,
+        norms,
+        nutrient_cache,
+        prefilter_k=req.prefilter_k,
+        must_keep_ids=must_keep,
+    )
+
+    started = time.perf_counter()
+    evaluations = 0
+    triplets_evaluated = 0
+    triplet_winners: list[dict[str, Any]] = []
+
+    def _eval_triplet(
+        triplet: tuple[Component, Component, Component],
+    ) -> dict[str, Any] | None:
+        nonlocal evaluations
+        model = build_triplet_score_model(
+            triplet,
+            nutrient_cache=nutrient_cache,
+            norms=norms,
+            optimize_keys=optimize_keys,
+            herd_scale=herd_scale,
+            heads=heads,
+            cost_weight=req.cost_weight,
+            profile_mass_kg=profile_mass_kg,
+        )
+
+        def score_fn(shares: tuple[float, float, float]) -> float:
+            _violation, _cost_head, score = score_model_shares(model, shares)
+            return score
+
+        opt = optimize_shares_for_triplet(
+            triplet,
+            score_fn,
+            min_share=req.min_share,
+            grid_step=req.grid_step,
+        )
+        if opt is None:
+            return None
+        evaluations += opt.evaluations
+        violation, cost_head, score = score_model_shares(model, opt.shares)
+        lines = build_calc_lines(triplet, opt.shares, herd_scale, nutrient_cache)
+        daily = [opt.shares[i] * herd_scale for i in range(3)]
+        return {
+            "score": score,
+            "violation": violation,
+            "costHead": cost_head,
+            "costTotal": cost_head * heads,
+            "triplet": triplet,
+            "shares": opt.shares,
+            "daily": daily,
+            "lines": lines,
+        }
+
+    for triplet in itertools.combinations(shortlist, 3):
+        triplet_ids = {c.id for c in triplet}
+        if not triplet_meets_group_rules(triplet_ids, group_selections):
+            continue
+        triplets_evaluated += 1
+        best_triplet_result = _eval_triplet(triplet)
+        if best_triplet_result is None:
+            continue
+        triplet_winners.append(best_triplet_result)
+
+    if not triplet_winners:
+        if group_selections:
+            raise ValueError("no_feasible_solution_groups")
+        raise ValueError("no_feasible_solution")
+
+    triplet_winners.sort(key=lambda x: x["score"])
+    best = triplet_winners[0]
+
+    full_calc = calculate_ration(
+        profile.ration_type or "DAIRY",
+        best["lines"],
+        norms,
+        heads_per_trip=heads,
+        profile_mass_kg=profile_mass_kg,
+    )
+    best["calc"] = full_calc
+    best["violation"] = violation_score(
+        full_calc.get("indicators") or [],
+        optimize_keys,
+        norms,
+    )
+    best["score"] = best["violation"] * req.cost_weight + best["costHead"]
+
+    duration_ms = int((time.perf_counter() - started) * 1000)
+
+    alternatives = [
+        {
+            "componentIds": [c.id for c in item["triplet"]],
+            "names": [c.name for c in item["triplet"]],
+            "score": item["score"],
+            "violation": item["violation"],
+            "costPerHead": item["costHead"],
+        }
+        for item in triplet_winners[:3]
+    ]
+    alternatives.sort(key=lambda x: x["score"])
+    alternatives = alternatives[:3]
+
+    result_lines = []
+    for i, comp in enumerate(best["triplet"]):
+        s1, s2, s3 = best["shares"]
+        share = (s1, s2, s3)[i]
+        result_lines.append(
+            {
+                "componentId": comp.id,
+                "name": comp.name,
+                "dailyKg": round(best["daily"][i], 4),
+                "sharePct": round(share * 100, 2),
+                "pricePerKg": comp.price,
+                "dryMatterPct": comp.dry_matter,
+            }
+        )
+
+    violation = best["violation"]
+    return {
+        "lines": result_lines,
+        "candidatePoolSize": len(candidates),
+        "groupSelections": group_selections,
+        "shortlistedIds": [c.id for c in shortlist],
+        "costTotal": best["costTotal"],
+        "costPerHead": best["costHead"],
+        "score": best["score"],
+        "violation": violation,
+        "feasible": violation < 0.01,
+        "indicators": best["calc"].get("indicators") or [],
+        "totals": best["calc"].get("totals") or [],
+        "optimizeKeys": optimize_keys,
+        "dynamicNorms": dynamic.get("dynamicNorms") or None,
+        "normsMethod": dynamic.get("normsMethod"),
+        "normsMeta": dynamic.get("normsMeta"),
+        "alternatives": alternatives,
+        "searchStats": {
+            "evaluations": evaluations,
+            "tripletsEvaluated": triplets_evaluated,
+            "durationMs": duration_ms,
+            "prefilterApplied": prefilter_applied,
+            "prefilterK": req.prefilter_k,
+            "scoreEngine": "fast",
+            "optimizer": "slsqp",
+            "normsMethod": dynamic.get("normsMethod"),
+        },
+    }
+
+
+def list_formulate_components() -> list[dict[str, Any]]:
+    rows = (
+        Component.query.filter_by(is_active=True, is_deleted=False)
+        .order_by(Component.name)
+        .all()
+    )
+    out: list[dict[str, Any]] = []
+    for comp in rows:
+        v = validate_component(comp.id)
+        feed_group = _classify_feed_group(comp)
+        out.append(
+            {
+                "id": comp.id,
+                "name": comp.name,
+                "type": comp.type,
+                "feedGroup": feed_group,
+                "eligible": v["eligible"],
+                "missing": v["missing"],
+                "warnings": v["warnings"],
+                "dryMatterPct": v["dryMatterPct"],
+                "hasPrice": v["hasPrice"],
+                "price": v["price"],
+                "mainFeedDmGPerKg": v.get("mainFeedDmGPerKg"),
+                "isMainFeed": v.get("isMainFeed", False),
+            }
+        )
+    return out

WESP_REL/app/lab/calc/formulate_optimize.py

@@ -0,0 +1,112 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any, Callable
+
+from scipy.optimize import minimize
+
+from app.models import Component
+
+ShareTuple = tuple[float, float, float]
+ScoreFn = Callable[[ShareTuple], float]
+
+
+@dataclass
+class OptimizeSharesResult:
+    shares: ShareTuple
+    score: float
+    evaluations: int
+
+
+def _normalize_shares(s1: float, s2: float, min_share: float) -> ShareTuple | None:
+    s3 = 1.0 - s1 - s2
+    ms = max(min_share, 0.0)
+    if s1 < ms - 1e-9 or s2 < ms - 1e-9 or s3 < ms - 1e-9:
+        return None
+    if abs(s1 + s2 + s3 - 1.0) > 1e-6:
+        return None
+    return (round(s1, 8), round(s2, 8), round(s3, 8))
+
+
+def _start_points(min_share: float, grid_step: float) -> list[tuple[float, float]]:
+    """Multi-start seeds: simplex center, corners, and a few grid_step hints."""
+    ms = max(min_share, 0.0)
+    max_pair = max(1.0 - 2 * ms, ms)
+    center = round((1.0 - ms) / 3.0, 6)
+    points: list[tuple[float, float]] = [
+        (center, center),
+        (ms, ms),
+        (max_pair, ms),
+        (ms, max_pair),
+        (max_pair, max_pair),
+    ]
+    step = max(grid_step, 0.1)
+    if ms <= step <= max_pair:
+        points.append((step, ms))
+        points.append((ms, step))
+    deduped: list[tuple[float, float]] = []
+    seen: set[tuple[float, float]] = set()
+    for s1, s2 in points:
+        if s1 + s2 > 1.0 - ms + 1e-9:
+            continue
+        key = (round(s1, 6), round(s2, 6))
+        if key in seen:
+            continue
+        seen.add(key)
+        deduped.append(key)
+    return deduped
+
+
+def optimize_shares_for_triplet(
+    triplet: tuple[Component, Component, Component],
+    score_fn: ScoreFn,
+    *,
+    min_share: float,
+    grid_step: float = 0.1,
+) -> OptimizeSharesResult | None:
+    del triplet
+    ms = max(min_share, 0.0)
+    max_s1 = max(1.0 - 2 * ms, ms)
+    evaluations = 0
+
+    def objective(x: Any) -> float:
+        nonlocal evaluations
+        evaluations += 1
+        shares = _normalize_shares(float(x[0]), float(x[1]), ms)
+        if shares is None:
+            return 1e18
+        return score_fn(shares)
+
+    bounds = [(ms, max_s1), (ms, max_s1)]
+    constraints = [{"type": "ineq", "fun": lambda x: 1.0 - ms - float(x[0]) - float(x[1])}]
+
+    best_score: float | None = None
+    best_shares: ShareTuple | None = None
+
+    for s1, s2 in _start_points(ms, grid_step):
+        if s1 + s2 > 1.0 - ms + 1e-9:
+            continue
+        try:
+            res = minimize(
+                objective,
+                [s1, s2],
+                method="SLSQP",
+                bounds=bounds,
+                constraints=constraints,
+                options={"ftol": 1e-8, "maxiter": 40},
+            )
+        except Exception:
+            continue
+        if not res.success and res.fun >= 1e17:
+            continue
+        shares = _normalize_shares(float(res.x[0]), float(res.x[1]), ms)
+        if shares is None:
+            continue
+        score = float(res.fun)
+        if best_score is None or score < best_score:
+            best_score = score
+            best_shares = shares
+
+    if best_shares is None or best_score is None:
+        return None
+    return OptimizeSharesResult(shares=best_shares, score=best_score, evaluations=evaluations)

WESP_REL/app/lab/calc/formulate_score.py

@@ -0,0 +1,371 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any
+
+from app.lab.calc.derived import apply_content_derived
+from app.lab.calc.nutrients import daily_intake_total, get_nutrient_value, norm_diff, weighted_average
+from app.lab.indicators import RATION_ALL_INDICATORS, indicator_by_key
+from app.models import Component
+
+
+def resolve_score_closure(optimize_keys: list[str]) -> frozenset[str]:
+    """Collect base + derived indicator keys needed to score optimize_keys."""
+    needed: set[str] = set(optimize_keys)
+    changed = True
+    while changed:
+        changed = False
+        for key in list(needed):
+            defn = indicator_by_key(key)
+            if defn is None:
+                continue
+            derived = defn.get("derived")
+            if derived == "alias":
+                dep = defn.get("alias_of")
+                if dep and dep not in needed:
+                    needed.add(dep)
+                    changed = True
+            elif derived in ("pct_of_dm", "g_per_kg_dm", "nel_per_kg_dm"):
+                dep = defn.get("from_key")
+                if dep and dep not in needed:
+                    needed.add(dep)
+                    changed = True
+            elif derived in ("rnb", "bra_rnb"):
+                for dep in ("crude_protein", "usp"):
+                    if dep not in needed:
+                        needed.add(dep)
+                        changed = True
+            elif derived == "ratio":
+                for dep in (defn.get("ratio_num"), defn.get("ratio_den")):
+                    if dep and dep not in needed:
+                        needed.add(dep)
+                        changed = True
+            elif derived == "dm_pct_bw":
+                if "dry_matter" not in needed:
+                    needed.add("dry_matter")
+                    changed = True
+    return frozenset(needed)
+
+
+def _indicator_content(
+    defn: dict[str, Any],
+    active: list[dict[str, Any]],
+    total_kg: float,
+    *,
+    heads_per_trip: int,
+) -> float | None:
+    key = defn.get("key")
+    if defn.get("aggregation") == "weighted_avg":
+        return weighted_average(
+            active,
+            total_kg,
+            defn["nutrient_keys"],
+            indicator_key=key,
+        )
+    return daily_intake_total(
+        active,
+        defn["nutrient_keys"],
+        heads_per_trip=heads_per_trip,
+        indicator_key=key,
+    )
+
+
+def compute_score_indicators(
+    active: list[dict[str, Any]],
+    total_kg: float,
+    norms: dict[str, dict[str, float | None]],
+    optimize_keys: list[str],
+    *,
+    heads_per_trip: int = 1,
+    profile_mass_kg: float | None = None,
+) -> list[dict[str, Any]]:
+    closure = resolve_score_closure(optimize_keys)
+    content_by_key: dict[str, float | None] = {}
+
+    for defn in RATION_ALL_INDICATORS:
+        key = defn["key"]
+        if key not in closure or defn.get("derived"):
+            continue
+        content_by_key[key] = _indicator_content(
+            defn, active, total_kg, heads_per_trip=heads_per_trip
+        )
+
+    for defn in RATION_ALL_INDICATORS:
+        key = defn["key"]
+        if key not in closure or not defn.get("derived"):
+            continue
+        content_by_key[key] = apply_content_derived(
+            content_by_key,
+            defn,
+            total_kg=total_kg,
+            heads_per_trip=heads_per_trip,
+            profile_mass_kg=profile_mass_kg,
+        )
+
+    rows: list[dict[str, Any]] = []
+    for key in optimize_keys:
+        if key not in closure:
+            continue
+        defn = indicator_by_key(key)
+        if defn is None:
+            continue
+        content = content_by_key.get(key)
+        bounds = norms.get(key, {})
+        min_v = bounds.get("min")
+        max_v = bounds.get("max")
+        diff = norm_diff(content, min_v, max_v)
+        if content is None and min_v is None and max_v is None:
+            continue
+        rows.append(
+            {
+                "key": key,
+                "label": defn["label"],
+                "unit": defn["unit"],
+                "min": min_v,
+                "max": max_v,
+                "content": content,
+                "diff": diff,
+            }
+        )
+    return rows
+
+
+def violation_score(
+    indicators: list[dict[str, Any]],
+    optimize_keys: list[str],
+    norms: dict[str, dict[str, float | None]],
+) -> float:
+    by_key = {row.get("key"): row for row in indicators if row.get("key")}
+    total = 0.0
+    counted = 0
+    for key in optimize_keys:
+        bounds = norms.get(key) or {}
+        if bounds.get("min") is None and bounds.get("max") is None:
+            continue
+        row = by_key.get(key)
+        if row is None:
+            continue
+        diff = row.get("diff")
+        if diff is None:
+            diff = norm_diff(row.get("content"), bounds.get("min"), bounds.get("max"))
+        if diff is None:
+            continue
+        total += float(diff) ** 2
+        counted += 1
+    return total if counted else 0.0
+
+
+def _violation_from_content(
+    content_by_key: dict[str, float | None],
+    optimize_keys: list[str],
+    norms: dict[str, dict[str, float | None]],
+) -> float:
+    total = 0.0
+    counted = 0
+    for key in optimize_keys:
+        bounds = norms.get(key) or {}
+        if bounds.get("min") is None and bounds.get("max") is None:
+            continue
+        content = content_by_key.get(key)
+        diff = norm_diff(content, bounds.get("min"), bounds.get("max"))
+        if diff is None:
+            continue
+        total += float(diff) ** 2
+        counted += 1
+    return total if counted else 0.0
+
+
+@dataclass
+class TripletScoreModel:
+    triplet: tuple[Component, Component, Component]
+    nutrient_cache: dict[str, dict[str, float]]
+    herd_scale: float
+    heads: int
+    cost_weight: float
+    profile_mass_kg: float | None
+    optimize_keys: list[str]
+    norms: dict[str, dict[str, float | None]]
+    closure: frozenset[str]
+    intake_unit: dict[str, tuple[float | None, float | None, float | None]]
+    weighted_vals: dict[str, tuple[float | None, float | None, float | None]]
+    prices: tuple[float, float, float]
+
+
+def build_triplet_score_model(
+    triplet: tuple[Component, Component, Component],
+    *,
+    nutrient_cache: dict[str, dict[str, float]],
+    norms: dict[str, dict[str, float | None]],
+    optimize_keys: list[str],
+    herd_scale: float,
+    heads: int,
+    cost_weight: float,
+    profile_mass_kg: float | None,
+) -> TripletScoreModel:
+    closure = resolve_score_closure(optimize_keys)
+    kg_per_share = herd_scale / max(heads, 1)
+    intake_unit: dict[str, tuple[float | None, float | None, float | None]] = {}
+    weighted_vals: dict[str, tuple[float | None, float | None, float | None]] = {}
+
+    for defn in RATION_ALL_INDICATORS:
+        key = defn["key"]
+        if key not in closure or defn.get("derived"):
+            continue
+        vals: list[float | None] = []
+        for comp in triplet:
+            nutrients = nutrient_cache.get(comp.id, {})
+            vals.append(
+                get_nutrient_value(
+                    comp.dry_matter,
+                    nutrients,
+                    defn["nutrient_keys"],
+                    indicator_key=key,
+                )
+            )
+        tup = (vals[0], vals[1], vals[2])
+        if defn.get("aggregation") == "weighted_avg":
+            weighted_vals[key] = tup
+        else:
+            intake_unit[key] = tuple(v * kg_per_share if v is not None else None for v in vals)
+
+    prices = tuple(
+        float(comp.price) if comp.price is not None else 0.0 for comp in triplet
+    )
+    return TripletScoreModel(
+        triplet=triplet,
+        nutrient_cache=nutrient_cache,
+        herd_scale=herd_scale,
+        heads=heads,
+        cost_weight=cost_weight,
+        profile_mass_kg=profile_mass_kg,
+        optimize_keys=optimize_keys,
+        norms=norms,
+        closure=closure,
+        intake_unit=intake_unit,
+        weighted_vals=weighted_vals,
+        prices=prices,
+    )
+
+
+def _content_from_shares(
+    model: TripletScoreModel,
+    shares: tuple[float, float, float],
+) -> dict[str, float | None]:
+    content_by_key: dict[str, float | None] = {}
+    total_kg = model.herd_scale
+
+    for key, coeffs in model.intake_unit.items():
+        parts = [
+            shares[i] * coeffs[i]
+            for i in range(3)
+            if coeffs[i] is not None
+        ]
+        content_by_key[key] = sum(parts) if parts else None
+
+    for key, vals in model.weighted_vals.items():
+        num = 0.0
+        den = 0.0
+        for i in range(3):
+            if vals[i] is None:
+                continue
+            num += shares[i] * vals[i]
+            den += shares[i]
+        content_by_key[key] = (num / den) if den > 0 else None
+
+    for defn in RATION_ALL_INDICATORS:
+        key = defn["key"]
+        if key not in model.closure or not defn.get("derived"):
+            continue
+        content_by_key[key] = apply_content_derived(
+            content_by_key,
+            defn,
+            total_kg=total_kg,
+            heads_per_trip=model.heads,
+            profile_mass_kg=model.profile_mass_kg,
+        )
+    return content_by_key
+
+
+def score_model_shares(
+    model: TripletScoreModel,
+    shares: tuple[float, float, float],
+) -> tuple[float, float, float]:
+    """Return (violation, cost_head, score) without building line dicts."""
+    content = _content_from_shares(model, shares)
+    violation = _violation_from_content(content, model.optimize_keys, model.norms)
+    cost_total = sum(shares[i] * model.prices[i] * model.herd_scale for i in range(3))
+    cost_head = cost_total / max(model.heads, 1)
+    score = violation * model.cost_weight + cost_head
+    return violation, cost_head, score
+
+
+def build_calc_lines(
+    triplet: tuple[Component, Component, Component],
+    shares: tuple[float, float, float],
+    herd_scale: float,
+    nutrient_cache: dict[str, dict[str, float]],
+) -> list[dict[str, Any]]:
+    lines: list[dict[str, Any]] = []
+    for i, comp in enumerate(triplet):
+        daily_kg = shares[i] * herd_scale
+        lines.append(
+            {
+                "component_id": comp.id,
+                "ingredient_name": comp.name,
+                "daily_kg": daily_kg,
+                "in_ration": True,
+                "in_compound": False,
+                "dry_matter": comp.dry_matter,
+                "price_per_kg": comp.price,
+                "nutrients": nutrient_cache.get(comp.id, {}),
+            }
+        )
+    return lines
+
+
+def cost_per_head(
+    triplet: tuple[Component, Component, Component],
+    shares: tuple[float, float, float],
+    herd_scale: float,
+    heads: int,
+) -> float:
+    total = 0.0
+    any_cost = False
+    for i, comp in enumerate(triplet):
+        kg = shares[i] * herd_scale
+        price = comp.price
+        if kg is None or price is None:
+            continue
+        total += float(kg) * float(price)
+        any_cost = True
+    if not any_cost:
+        return 0.0
+    return total / max(heads, 1)
+
+
+def score_triplet_shares(
+    triplet: tuple[Component, Component, Component],
+    shares: tuple[float, float, float],
+    *,
+    nutrient_cache: dict[str, dict[str, float]],
+    norms: dict[str, dict[str, float | None]],
+    optimize_keys: list[str],
+    herd_scale: float,
+    heads: int,
+    cost_weight: float,
+    profile_mass_kg: float | None,
+) -> tuple[float, float, float, list[dict[str, Any]]]:
+    """Return (violation, cost_head, score, lines)."""
+    model = build_triplet_score_model(
+        triplet,
+        nutrient_cache=nutrient_cache,
+        norms=norms,
+        optimize_keys=optimize_keys,
+        herd_scale=herd_scale,
+        heads=heads,
+        cost_weight=cost_weight,
+        profile_mass_kg=profile_mass_kg,
+    )
+    violation, cost_head, score = score_model_shares(model, shares)
+    lines = build_calc_lines(triplet, shares, herd_scale, nutrient_cache)
+    return violation, cost_head, score, lines

WESP_REL/app/lab/calc/formulate_validate.py

@@ -0,0 +1,86 @@
+from __future__ import annotations
+
+from typing import Any
+
+from app.lab.calc.feed_groups import classify_feed_group
+from app.lab.calc.gfe_policies import default_om_digestibility_pct
+from app.lab.nutrient_schema import read_from_mapping
+from app.lab.services.component_nutrients import derive_context_for_component, nutrients_full_dict, nutrients_is_empty
+from app.models import Component
+
+_REQUIRED_EAV_KEYS = ("Сыр. Протеин", "Сырая клетч", "Сырой жир")
+_MAIN_FEED_KEYS = ("Осн.Корм", "СВ Основной корм")
+_OMD_KEYS = ("ВРХ Орг Вещ", "КРС Орг Вещ")
+
+
+def main_feed_dm_g_per_kg(component_id: str | None) -> float | None:
+    if not component_id:
+        return None
+    full = nutrients_full_dict(component_id)
+    val = read_from_mapping(full, _MAIN_FEED_KEYS)
+    return float(val) if val is not None else None
+
+
+def validate_component(component_id: str) -> dict[str, Any]:
+    comp = Component.query.filter_by(id=component_id, is_deleted=False).first()
+    if comp is None:
+        return {
+            "id": component_id,
+            "name": None,
+            "eligible": False,
+            "missing": ["component_not_found"],
+            "warnings": [],
+            "dryMatterPct": None,
+            "hasPrice": False,
+            "price": None,
+        }
+
+    missing: list[str] = []
+    warnings: list[str] = []
+    dry_matter_pct = comp.dry_matter
+    if dry_matter_pct is None or float(dry_matter_pct) <= 0:
+        missing.append("dry_matter")
+
+    full = nutrients_full_dict(component_id)
+    if nutrients_is_empty(component_id):
+        missing.append("nutrients_empty")
+    else:
+        for key in _REQUIRED_EAV_KEYS:
+            if read_from_mapping(full, (key,)) is None:
+                missing.append(key)
+
+    has_price = comp.price is not None and float(comp.price) >= 0
+    if not has_price:
+        warnings.append("price_missing")
+
+    main_feed_dm = read_from_mapping(full, _MAIN_FEED_KEYS)
+    if main_feed_dm is None:
+        warnings.append("main_feed_unset")
+
+    feed_group = classify_feed_group(comp)
+    omd = read_from_mapping(full, _OMD_KEYS)
+    if feed_group in ("rough", "succulent") and omd is None:
+        warnings.append("omd_missing")
+    elif omd is None and comp.dry_matter and float(comp.dry_matter) > 0:
+        ctx = derive_context_for_component(component_id, full)
+        warnings.append(
+            f"omd_defaulted:{default_om_digestibility_pct(ctx):.0f}"
+        )
+
+    return {
+        "id": comp.id,
+        "name": comp.name,
+        "eligible": len(missing) == 0,
+        "missing": missing,
+        "warnings": warnings,
+        "dryMatterPct": dry_matter_pct,
+        "hasPrice": has_price,
+        "price": comp.price,
+        "mainFeedDmGPerKg": main_feed_dm,
+        "isMainFeed": main_feed_dm is not None and float(main_feed_dm) > 0,
+        "feedGroup": feed_group,
+    }
+
+
+def validate_components(component_ids: list[str]) -> list[dict[str, Any]]:
+    return [validate_component(cid) for cid in component_ids]

WESP_REL/app/lab/calc/gfe_norms.py

@@ -0,0 +1,94 @@
+"""Динамические нормы по уравнениям GfE (Германия)."""
+
+from __future__ import annotations
+
+from typing import Any
+
+USP_DYNAMIC_KEY = "usp"
+NEL_DYNAMIC_KEY = "nel"
+
+# GfE 2001 Milchkühe — Erhaltung + Milch (Standardmilch / FCM)
+NEL_MAINTENANCE_COEFF = 0.293  # MJ NEL / (kg LM)^0,75 / Tag
+NEL_PER_KG_MILK_MJ = 3.3  # MJ NEL / kg Milch (FCM)
+
+
+def gfe_usp_min_g(
+    mass_kg: float | None,
+    milk_yield_kg: float | None = None,
+) -> float | None:
+    """
+    Минимальная суточная потребность в усвояемом протеине (уСП / nXP), г/сут.
+
+    GfE: 0,09 × (масса^0,75) × 6,25 + удой × 85 г.
+    """
+    if mass_kg is None or mass_kg <= 0:
+        return None
+    maintenance = 0.09 * (mass_kg**0.75) * 6.25
+    milk = max(float(milk_yield_kg or 0), 0.0) * 85.0
+    return maintenance + milk
+
+
+def gfe_nel_min_mj(
+    mass_kg: float | None,
+    milk_yield_kg: float | None = None,
+) -> float | None:
+    """
+    Минимальная суточная потребность в ЧЭЛ (NEL), МДж/сут.
+
+    GfE 2001: 0,293 × LM^0,75 + удой × 3,3 (MJ NEL на кг молока).
+    """
+    if mass_kg is None or mass_kg <= 0:
+        return None
+    maintenance = NEL_MAINTENANCE_COEFF * (mass_kg**0.75)
+    milk = max(float(milk_yield_kg or 0), 0.0) * NEL_PER_KG_MILK_MJ
+    return maintenance + milk
+
+
+def preview_dynamic_norms(
+    mass_kg: float | None,
+    milk_yield_kg: float | None = None,
+) -> dict[str, Any]:
+    """Расчётные min по GfE для UI (без учёта сохранённых норм)."""
+    _, dynamic = apply_dynamic_norms(
+        {},
+        mass_kg=mass_kg,
+        milk_yield_kg=milk_yield_kg,
+    )
+    return dynamic
+
+
+def apply_dynamic_norms(
+    stored: dict[str, dict[str, float | None]],
+    *,
+    mass_kg: float | None,
+    milk_yield_kg: float | None,
+    ration_type: str | None = None,
+    force_dynamic: bool = False,
+) -> tuple[dict[str, dict[str, float | None]], dict[str, Any]]:
+    """
+    Заполняет нормы по GfE, если в БД min не задан.
+    force_dynamic=True — пересчитать min уСП/ЧЭЛ по массе и удою даже при нормах в БД
+    (автоготовка с явными mass_kg / milk_yield_kg).
+    Возвращает (resolved_norms, dynamic_meta).
+    """
+    del ration_type
+    resolved: dict[str, dict[str, float | None]] = {
+        key: {"min": bounds.get("min"), "max": bounds.get("max")}
+        for key, bounds in stored.items()
+    }
+    dynamic: dict[str, Any] = {}
+
+    for key, compute, formula in (
+        (USP_DYNAMIC_KEY, gfe_usp_min_g, "0.09×масса^0.75×6.25 + удой×85"),
+        (NEL_DYNAMIC_KEY, gfe_nel_min_mj, "0.293×масса^0.75 + удой×3.3"),
+    ):
+        bounds = resolved.get(key, {"min": None, "max": None})
+        if force_dynamic or bounds.get("min") is None:
+            computed = compute(mass_kg, milk_yield_kg)
+            if computed is not None:
+                entry = dict(bounds)
+                entry["min"] = computed
+                resolved[key] = entry
+                dynamic[key] = {"min": computed, "formula": formula}
+
+    return resolved, dynamic

WESP_REL/app/lab/calc/gfe_policies.py

@@ -0,0 +1,84 @@
+"""WESP-политики расчёта на базе GfE 2001 (отличия от zootech Excel — осознанные)."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Literal
+
+FeedGroup = Literal["rough", "succulent", "concentrate", "other", "unknown"]
+
+# --- GfE 2001 константы (формулы не меняем) ---
+NEL_Q_COEFF = 0.004
+NEL_Q_REF_PCT = 57.0
+NEL_BASE = 0.6
+
+GE_CP = 0.0239
+GE_FAT = 0.0398
+GE_FIBER = 0.0201
+GE_NFE = 0.0175
+
+ME_FAT = 0.0312
+ME_FIBER = 0.0136
+ME_OR_RESIDUE = 0.0147
+ME_CP = 0.00234
+
+USP_FAT_THRESHOLD_G_PER_KG_DM = 70.0  # 7% СЖ/кг СВ
+
+# DCAB (Na+K)-(Cl+S), мэкв при минералах в г/кг СВ
+DCAB_NA = 43.5
+DCAB_K = 25.6
+DCAB_CL = 28.2
+DCAB_S = 62.4
+
+# Дефолты переваримости при пустых коэфф. (Excel legacy, кроме ОВ)
+DEFAULT_CP_DIGEST_PCT = 86.0
+DEFAULT_FAT_DIGEST_PCT = 75.0
+DEFAULT_FIBER_DIGEST_PCT = 86.0
+DEFAULT_NFE_DIGEST_PCT = 94.0
+DEFAULT_INSOLUBLE_PROTEIN_PCT = 15.0
+DEFAULT_PROTEIN_FRACTION_PCT = 18.9
+
+# WESP: дефолт ВРХ орг. вещ. при пустом поле — по классу корма
+DEFAULT_OMD_ROUGH_PCT = 65.0
+DEFAULT_OMD_SUCCULENT_PCT = 72.0
+DEFAULT_OMD_CONCENTRATE_PCT = 91.0
+
+
+@dataclass(frozen=True)
+class DeriveContext:
+    feed_group: FeedGroup = "unknown"
+    is_main_feed: bool = False
+
+    @classmethod
+    def from_cells(cls, *, main_feed_g: float = 0.0, feed_group: FeedGroup = "unknown") -> DeriveContext:
+        return cls(
+            feed_group=feed_group,
+            is_main_feed=main_feed_g > 0,
+        )
+
+    @classmethod
+    def infer_from_cells(cls, cells: dict[str, float], feed_group: FeedGroup = "unknown") -> DeriveContext:
+        main_feed = cells.get("E", 0.0)
+        return cls.from_cells(main_feed_g=main_feed, feed_group=feed_group)
+
+
+def default_om_digestibility_pct(ctx: DeriveContext | None) -> float:
+    """Дефолт переваримости ОВ (%) при отсутствии ВРХ Орг Вещ."""
+    if ctx is None:
+        return DEFAULT_OMD_CONCENTRATE_PCT
+    if ctx.is_main_feed or ctx.feed_group == "rough":
+        return DEFAULT_OMD_ROUGH_PCT
+    if ctx.feed_group == "succulent":
+        return DEFAULT_OMD_SUCCULENT_PCT
+    return DEFAULT_OMD_CONCENTRATE_PCT
+
+
+def default_digestibility_coefficients() -> dict[str, float]:
+    return {
+        "cp": DEFAULT_CP_DIGEST_PCT,
+        "fat": DEFAULT_FAT_DIGEST_PCT,
+        "fiber": DEFAULT_FIBER_DIGEST_PCT,
+        "nfe": DEFAULT_NFE_DIGEST_PCT,
+        "insoluble_protein": DEFAULT_INSOLUBLE_PROTEIN_PCT,
+        "protein_fraction": DEFAULT_PROTEIN_FRACTION_PCT,
+    }

WESP_REL/app/lab/calc/ingredient_catalog.py

@@ -0,0 +1,187 @@
+"""Каталог показателей zootech «База сырья» (109 колонок)."""
+
+from __future__ import annotations
+
+from openpyxl.utils import column_index_from_string, get_column_letter
+
+# Заголовки row 3 в xlsx_extracted/data_csv_cleaned/База сырья.csv
+INGREDIENT_HEADERS: tuple[str, ...] = (
+    "№",
+    "Наименование",
+    "Цена 1 кг",
+    "СВ",
+    "Осн.Корм",
+    "Сыр. Протеин",
+    "уСП",
+    "БРА",
+    "ЧЭЛ- КРС",
+    "ОЭ-КРС",
+    "Сырая клетч",
+    "Структур клетч",
+    "Сырой жир",
+    "НДК",
+    "КДК",
+    "NFC",
+    "Ca",
+    "P",
+    "Mg",
+    "Fe",
+    "Zn",
+    "Cu",
+    "Co",
+    "Mn",
+    "Se",
+    "J",
+    "Na",
+    "K",
+    "CL",
+    "S",
+    "DCAB Форм",
+    "Сахар и Крохм",
+    "Нераств  Крохм",
+    "Сахар",
+    "Крахмал",
+    "Нераств крахмал",
+    "Вит А",
+    "Вит D",
+    "Вит Е",
+    "Вит В1",
+    "Вит В2",
+    "Вит В6",
+    "Вит В12",
+    "Пант Кальц",
+    "Никот Ки-та",
+    "Фол ки-та",
+    "Холин",
+    "Биотин",
+    "Сырая зола",
+    "БЕР",
+    "Лизин",
+    "Метионин",
+    "Треонин",
+    "Триптофан",
+    "Изолейцин",
+    "Лейцин",
+    "Валин",
+    "Каротин",
+    "b -Каротин",
+    "Линолевая к-та",
+    "Линоленовая ки-та",
+    "Масляная ки-та",
+    "Арахидоновая ки-та",
+    "Полиэновая ки-та",
+    "Мочевина",
+    "СВ Основной корм",
+    "ВРХ Орг Вещ",
+    "Перевар Орг Вещ",
+    "КРС Протеин",
+    "Переварим Протеин",
+    "КРС Сырой жир",
+    "Переварим Сырой жир",
+    "КРС Сырая клетч",
+    "Переварим сырая клетч",
+    "КРС БЭВ",
+    "Переварим БЭВ",
+    "ВЕ",
+    "OЭ КРС форм",
+    "ЧЭЛ - КРС Форм",
+    "НДК Общ",
+    "НДК Осн. Корм",
+    "КДК общ",
+    "%  нераствор протеин",
+    "Нерастворим прот",
+    "СЖ/кг СВ",
+    "НСП/кг СВ",
+    "СП/кг СВ",
+    "пОВ/кг СВ",
+    "пСЖ/кг СВ",
+    "уСП<7%CЖ",
+    "уСП>7%CЖ",
+    "уСП/кг СВ формул",
+    "уСП в ОР",
+    "уСП формул",
+    "БРА",
+    "Нераств крохмал",
+    "доля крохмала",
+    "Доля белка",
+    "% перев в кишках",
+    "OEB",
+    "Синтез Мдж",
+    "Промеж рез 1",
+    "Промеж рез 2",
+    "Метаб ОЕТ",
+    "Метаб лизин",
+    "Метабол Треон",
+    "Метабол Лейцин",
+    "Метабол  Изолейц",
+    "Метабол Валин",
+)
+
+HEADER_TO_LETTER: dict[str, str] = {
+    header: get_column_letter(i + 1) for i, header in enumerate(INGREDIENT_HEADERS)
+}
+
+LETTER_TO_HEADER: dict[str, str] = {
+    get_column_letter(i + 1): header for i, header in enumerate(INGREDIENT_HEADERS)
+}
+
+# Колонки с формулами в шаблонной строке 6 (native port, не Excel runtime).
+DERIVED_LETTERS: frozenset[str] = frozenset(
+    {
+        "AE",
+        "AF",
+        "AG",
+        "AX",
+        "BN",
+        "BP",
+        "BR",
+        "BT",
+        "BV",
+        "BX",
+        "BY",
+        "BZ",
+        "CA",
+        "CF",
+        "CG",
+        "CH",
+        "CI",
+        "CJ",
+        "CK",
+        "CL",
+        "CM",
+        "CN",
+        "CP",
+        "CQ",
+        "CX",
+        "CY",
+        "CZ",
+        "DA",
+        "DB",
+        "DC",
+        "DD",
+        "DE",
+    }
+)
+
+# Отображаемые поля G–J синхронизируются с расчётными CP/CQ/CA/BZ.
+DISPLAY_SYNC: tuple[tuple[str, str], ...] = (
+    ("G", "CP"),  # уСП
+    ("H", "CQ"),  # RNB (legacy заголовок «БРА», дубль CQ)
+    ("I", "CA"),  # ЧЭЛ- КРС
+    ("J", "BZ"),  # ОЭ-КРС
+)
+
+DERIVED_HEADERS: frozenset[str] = frozenset(
+    LETTER_TO_HEADER[letter] for letter in DERIVED_LETTERS
+) | frozenset(LETTER_TO_HEADER[g] for g, _ in DISPLAY_SYNC)
+
+INPUT_HEADERS: frozenset[str] = frozenset(INGREDIENT_HEADERS) - DERIVED_HEADERS - frozenset(
+    ("№", "Наименование", "Цена 1 кг")
+)
+
+# Дублирующий заголовок «БРА»/RNB (H и CQ) — в derive используем CQ.
+DUPLICATE_HEADERS: frozenset[str] = frozenset({"БРА"})
+
+
+def letter_index(letter: str) -> int:
+    return column_index_from_string(letter) - 1

WESP_REL/app/lab/calc/ingredient_derive.py

@@ -0,0 +1,216 @@
+"""Native derive формул zootech «База сырья» — WESP GfE 2001 engine."""
+
+from __future__ import annotations
+
+from typing import Any
+
+from app.lab.calc.gfe_policies import (
+    DCAB_CL,
+    DCAB_K,
+    DCAB_NA,
+    DCAB_S,
+    DEFAULT_CP_DIGEST_PCT,
+    DEFAULT_FAT_DIGEST_PCT,
+    DEFAULT_FIBER_DIGEST_PCT,
+    DEFAULT_INSOLUBLE_PROTEIN_PCT,
+    DEFAULT_NFE_DIGEST_PCT,
+    DEFAULT_PROTEIN_FRACTION_PCT,
+    DeriveContext,
+    GE_CP,
+    GE_FAT,
+    GE_FIBER,
+    GE_NFE,
+    ME_CP,
+    ME_FAT,
+    ME_FIBER,
+    ME_OR_RESIDUE,
+    NEL_BASE,
+    NEL_Q_COEFF,
+    NEL_Q_REF_PCT,
+    USP_FAT_THRESHOLD_G_PER_KG_DM,
+    default_om_digestibility_pct,
+)
+from app.lab.calc.ingredient_catalog import (
+    DISPLAY_SYNC,
+    HEADER_TO_LETTER,
+    INGREDIENT_HEADERS,
+    LETTER_TO_HEADER,
+)
+
+
+def _parse_num(value: Any) -> float | None:
+    if value is None or value == "":
+        return None
+    try:
+        n = float(value)
+    except (TypeError, ValueError):
+        return None
+    return None if n != n else n
+
+
+def _normalize_key(value: str) -> str:
+    return " ".join((value or "").split()).strip().lower()
+
+
+def _v(cells: dict[str, float], letter: str, default: float = 0.0) -> float:
+    return cells.get(letter, default)
+
+
+def _if_pos(test: float, when_true, when_false: float = 0.0) -> float:
+    """Excel IF(test>0, …) — ветка when_true не вычисляется при test<=0."""
+    if test > 0:
+        return when_true() if callable(when_true) else when_true
+    return when_false
+
+
+def dict_to_cells(data: dict[str, Any] | None) -> dict[str, float]:
+    """Словарь {заголовок: значение} → {буква колонки: значение}."""
+    cells: dict[str, float] = {}
+    if not data:
+        return cells
+    norm_index = {_normalize_key(h): h for h in INGREDIENT_HEADERS}
+    for raw_key, raw_val in data.items():
+        n = _parse_num(raw_val)
+        if n is None:
+            continue
+        nk = _normalize_key(str(raw_key))
+        header = norm_index.get(nk)
+        if header is None:
+            continue
+        letter = HEADER_TO_LETTER.get(header)
+        if letter:
+            cells[letter] = n
+    return cells
+
+
+def cells_to_dict(cells: dict[str, float]) -> dict[str, float]:
+    out: dict[str, float] = {}
+    for letter, value in cells.items():
+        header = LETTER_TO_HEADER.get(letter)
+        if header and header not in ("№", "Наименование", "Цена 1 кг"):
+            out[header] = value
+    return out
+
+
+def derive_cells(
+    cells: dict[str, float],
+    *,
+    context: DeriveContext | None = None,
+) -> dict[str, float]:
+    """Пересчёт derived-колонок по цепочке формул row 6 «База сырья»."""
+    c = dict(cells)
+    ctx = context or DeriveContext.infer_from_cells(c)
+    omd_default = default_om_digestibility_pct(ctx)
+
+    c["AE"] = DCAB_NA * _v(c, "AA") + DCAB_K * _v(c, "AB") - DCAB_CL * _v(c, "AC") - DCAB_S * _v(c, "AD")
+    c["AG"] = _v(c, "AJ") * _v(c, "AI") / 100.0
+    c["AF"] = _v(c, "AI") - c["AG"] + _v(c, "AH")
+    c["AX"] = _v(c, "D") - _v(c, "F") - _v(c, "K") - _v(c, "M") - _v(c, "AW")
+    c["BN"] = _if_pos(_v(c, "E"), lambda: _v(c, "D") / _v(c, "E") * 1000.0)
+    c["BP"] = _if_pos(
+        _v(c, "BO"),
+        lambda: (_v(c, "D") - _v(c, "AW")) * _v(c, "BO") / 100.0,
+        (_v(c, "D") - _v(c, "AW")) * omd_default / 100.0,
+    )
+    c["BR"] = _if_pos(
+        _v(c, "BQ"), lambda: _v(c, "F") * _v(c, "BQ") / 100.0, _v(c, "F") * DEFAULT_CP_DIGEST_PCT / 100.0
+    )
+    c["BT"] = _if_pos(
+        _v(c, "BS"), lambda: _v(c, "M") * _v(c, "BS") / 100.0, _v(c, "M") * DEFAULT_FAT_DIGEST_PCT / 100.0
+    )
+    c["BV"] = _if_pos(
+        _v(c, "BU"), lambda: _v(c, "K") * _v(c, "BU") / 100.0, _v(c, "K") * DEFAULT_FIBER_DIGEST_PCT / 100.0
+    )
+    c["BX"] = _if_pos(
+        _v(c, "BW"), lambda: c["AX"] * _v(c, "BW") / 100.0, c["AX"] * DEFAULT_NFE_DIGEST_PCT / 100.0
+    )
+    c["BY"] = GE_CP * _v(c, "F") + GE_FAT * _v(c, "M") + GE_FIBER * _v(c, "K") + GE_NFE * c["AX"]
+    c["BZ"] = (
+        ME_FAT * c["BT"]
+        + ME_FIBER * c["BV"]
+        + ME_OR_RESIDUE * (c["BP"] - c["BT"] - c["BV"])
+        + ME_CP * _v(c, "F")
+    )
+    c["CA"] = _if_pos(
+        c["BY"],
+        lambda: (NEL_BASE * (1.0 + NEL_Q_COEFF * (c["BZ"] / c["BY"] * 100.0 - NEL_Q_REF_PCT)) * c["BZ"]),
+    )
+    c["CF"] = _if_pos(
+        _v(c, "CE"),
+        lambda: _v(c, "F") * _v(c, "CE") / 100.0,
+        _v(c, "F") * DEFAULT_INSOLUBLE_PROTEIN_PCT / 100.0,
+    )
+    c["CG"] = _if_pos(_v(c, "D"), lambda: _v(c, "M") * 1000.0 / _v(c, "D"))
+    c["CH"] = _if_pos(_v(c, "D"), lambda: c["CF"] * 1000.0 / _v(c, "D"))
+    c["CI"] = _if_pos(_v(c, "D"), lambda: _v(c, "F") * 1000.0 / _v(c, "D"))
+    c["CJ"] = _if_pos(_v(c, "D"), lambda: c["BP"] / _v(c, "D"))
+    c["CK"] = _if_pos(_v(c, "D"), lambda: c["BT"] / _v(c, "D"))
+    c["CL"] = _if_pos(c["CI"], lambda: (187.7 - 115.4 * c["CH"] / c["CI"]) * c["CJ"] + 1.03 * c["CH"])
+    c["CM"] = _if_pos(c["CI"], lambda: (196.1 - 127.5 * c["CH"] / c["CI"]) * (c["CJ"] - c["CK"]) + 1.03 * c["CH"])
+    co = _v(c, "CO")
+    if c["CG"] < USP_FAT_THRESHOLD_G_PER_KG_DM + 0.01:
+        c["CN"] = c["CL"]
+    elif c["CG"] > USP_FAT_THRESHOLD_G_PER_KG_DM:
+        c["CN"] = c["CM"]
+    else:
+        c["CN"] = 0.0
+    if co < 1.01:
+        c["CP"] = c["CN"] * _v(c, "D") / 1000.0
+    elif co > 1.0:
+        c["CP"] = co
+    else:
+        c["CP"] = 0.0
+    c["CQ"] = (_v(c, "F") - c["CP"]) / 6.25
+    c["CX"] = _if_pos(
+        _v(c, "CT"),
+        lambda: _v(c, "F") * _v(c, "CT") / 100.0,
+        _v(c, "F") * DEFAULT_PROTEIN_FRACTION_PCT / 100.0,
+    )
+    nel = c["CA"]
+    c["CY"] = nel * _v(c, "CW")
+    f_val = _v(c, "F")
+    bo = _v(c, "BO")
+    if f_val == 0.0:
+        c["CZ"] = 0.0
+        c["DA"] = 0.0
+        c["DB"] = 0.0
+        c["DC"] = 0.0
+        c["DD"] = 0.0
+        c["DE"] = 0.0
+    else:
+        c["DA"] = c["CX"] * (_v(c, "AY") / 10.0) / (f_val / 10.0) * (bo / 100.0) + c["CY"] * 0.071 * 0.8
+        c["CZ"] = _if_pos(
+            bo,
+            lambda: c["CX"] * (_v(c, "AZ") / 10.0) / (f_val / 10.0) * (bo / 100.0) + c["CY"] * 0.018 * 0.8,
+        )
+        c["DB"] = _if_pos(
+            bo,
+            lambda: c["CX"] * (_v(c, "BA") / 10.0) / (f_val / 10.0) * (bo / 100.0) + c["CY"] * 0.044 * 0.8,
+        )
+        c["DC"] = _if_pos(
+            _v(c, "BD"),
+            lambda: c["CX"] * (_v(c, "BD") / 10.0) / (f_val / 10.0) * (bo / 100.0) + c["CY"] * 0.063 * 0.8,
+        )
+        c["DD"] = _if_pos(
+            _v(c, "BC"),
+            lambda: c["CX"] * (_v(c, "BC") / 10.0) / (f_val / 10.0) * (bo / 100.0) + c["CY"] * 0.049 * 0.8,
+        )
+        c["DE"] = _if_pos(
+            _v(c, "BE"),
+            lambda: c["CX"] * (_v(c, "BE") / 10.0) / (f_val / 10.0) * (bo / 100.0) + c["CY"] * 0.048 * 0.8,
+        )
+
+    for display, source in DISPLAY_SYNC:
+        c[display] = c[source]
+
+    return c
+
+
+def derive_ingredient_nutrients(
+    data: dict[str, Any] | None,
+    *,
+    context: DeriveContext | None = None,
+) -> dict[str, float]:
+    """Полный набор показателей: входные + пересчитанные derived."""
+    cells = dict_to_cells(data)
+    return cells_to_dict(derive_cells(cells, context=context))

WESP_REL/app/lab/calc/norms.py

@@ -0,0 +1,38 @@
+from __future__ import annotations
+
+from typing import Any
+
+from app.lab.calc.nutrients import parse_num
+from app.lab.constants import NORM_COLUMN_ALIASES, RATION_QUALITY_INDICATORS
+
+
+def merge_norms_from_profile(profile_data: Any, ration_type: str) -> dict[str, dict[str, float | None]]:
+    del ration_type
+    if not profile_data or not isinstance(profile_data, dict):
+        return {}
+    indicators = profile_data.get("indicators")
+    if isinstance(indicators, dict):
+        out: dict[str, dict[str, float | None]] = {}
+        for key, bounds in indicators.items():
+            if not isinstance(bounds, dict):
+                continue
+            out[str(key)] = {
+                "min": parse_num(bounds.get("min")),
+                "max": parse_num(bounds.get("max")),
+            }
+        return out
+    out = {}
+    for defn in RATION_QUALITY_INDICATORS:
+        key = defn["key"]
+        aliases = NORM_COLUMN_ALIASES.get(key)
+        if not aliases:
+            continue
+        for alias in aliases:
+            entry = profile_data.get(alias)
+            if isinstance(entry, dict):
+                out[key] = {
+                    "min": parse_num(entry.get("min")),
+                    "max": parse_num(entry.get("max")),
+                }
+                break
+    return out

WESP_REL/app/lab/calc/norms_derived.py

@@ -0,0 +1,86 @@
+"""Производные min/max норм из базовых показателей RACION."""
+
+from __future__ import annotations
+
+from typing import Any
+
+from app.lab.indicators import RATION_ALL_INDICATORS, indicator_by_key
+
+
+def _norm_min(bounds: dict[str, float | None] | None) -> float | None:
+    if not bounds:
+        return None
+    v = bounds.get("min")
+    return float(v) if v is not None else None
+
+
+def _apply_derived_min(
+    norms: dict[str, dict[str, float | None]],
+    defn: dict[str, Any],
+    *,
+    mass_kg: float | None = None,
+) -> float | None:
+    derived = defn.get("derived")
+    key = defn["key"]
+    if derived == "alias":
+        src = norms.get(defn.get("alias_of") or "")
+        if src and src.get("min") is not None:
+            return src["min"]
+        return None
+    if derived == "pct_of_dm":
+        src = _norm_min(norms.get(defn.get("from_key") or ""))
+        dm = _norm_min(norms.get("dry_matter"))
+        if src is None or not dm or dm <= 0:
+            return None
+        return src / dm * 100.0
+    if derived == "g_per_kg_dm":
+        src = _norm_min(norms.get(defn.get("from_key") or ""))
+        dm = _norm_min(norms.get("dry_matter"))
+        if src is None or not dm or dm <= 0:
+            return None
+        return src / (dm / 1000.0)
+    if derived == "nel_per_kg_dm":
+        dm = _norm_min(norms.get("dry_matter"))
+        nel = _norm_min(norms.get("nel"))
+        if not dm or dm <= 0 or nel is None:
+            return None
+        return nel / (dm / 1000.0)
+    if derived == "ratio":
+        num = _norm_min(norms.get(defn.get("ratio_num") or ""))
+        den = _norm_min(norms.get(defn.get("ratio_den") or ""))
+        if num is None or den is None or den == 0:
+            return None
+        return num / den
+    if derived == "dm_pct_bw":
+        dm = _norm_min(norms.get("dry_matter"))
+        if dm is None or not mass_kg or mass_kg <= 0:
+            return None
+        return (dm / 1000.0 / mass_kg) * 100.0
+    if derived == "ration_pct_bw":
+        return None
+    if derived in ("rnb", "bra_rnb"):
+        return _norm_min(norms.get(key))
+    return None
+
+
+def apply_derived_norms(
+    norms: dict[str, dict[str, float | None]],
+    *,
+    mass_kg: float | None = None,
+) -> tuple[dict[str, dict[str, float | None]], dict[str, Any]]:
+    """Дополняет norms производными min; max не трогает."""
+    out = {k: dict(v) for k, v in norms.items()}
+    dynamic: dict[str, Any] = {}
+    for defn in RATION_ALL_INDICATORS:
+        if not defn.get("derived"):
+            continue
+        key = defn["key"]
+        if _norm_min(out.get(key)) is not None:
+            continue
+        val = _apply_derived_min(out, defn, mass_kg=mass_kg)
+        if val is None:
+            continue
+        rounded = round(val, 3)
+        out[key] = {"min": rounded, "max": out.get(key, {}).get("max")}
+        dynamic[key] = {"min": rounded, "derived": defn.get("derived")}
+    return out, dynamic

WESP_REL/app/lab/calc/norms_resolver.py

@@ -0,0 +1,192 @@
+"""Роутер методик суточных норм: WESP / Москва / Петербург."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+from typing import Any, Literal
+
+from app.lab.calc.gfe_norms import apply_dynamic_norms
+from app.lab.calc.racion.moscow import MoscowDairyParams, resolve_moscow_dairy_norms
+from app.lab.calc.racion.piter import PiterDairyParams, PiterPrepError, resolve_piter_dairy_norms
+from app.lab.indicators import RATION_ALL_INDICATORS
+
+NormsMethod = Literal["wesp", "racion_moscow", "racion_piter"]
+
+VALID_NORMS_METHODS: frozenset[str] = frozenset({"wesp", "racion_moscow", "racion_piter"})
+
+
+@dataclass
+class NormsParams:
+    milk_fat_pct: float | None = None
+    lactation_no: int | None = None
+    lactation_stage: int | None = None
+    body_condition: int | None = None
+    housing_system: int | None = None
+    konc_oe_sv: float | None = None
+
+    @classmethod
+    def from_dict(cls, raw: dict[str, Any] | None) -> NormsParams:
+        if not raw:
+            return cls()
+        return cls(
+            milk_fat_pct=_flt(raw.get("milkFatPct", raw.get("milk_fat_pct"))),
+            lactation_no=_int(raw.get("lactationNo", raw.get("lactation_no"))),
+            lactation_stage=_int(raw.get("lactationStage", raw.get("lactation_stage"))),
+            body_condition=_int(raw.get("bodyCondition", raw.get("body_condition"))),
+            housing_system=_int(raw.get("housingSystem", raw.get("housing_system"))),
+            konc_oe_sv=_flt(raw.get("koncOeSv", raw.get("konc_oe_sv"))),
+        )
+
+
+def _flt(v: Any) -> float | None:
+    if v is None or v == "":
+        return None
+    try:
+        return float(v)
+    except (TypeError, ValueError):
+        return None
+
+
+def _int(v: Any) -> int | None:
+    if v is None or v == "":
+        return None
+    try:
+        return int(v)
+    except (TypeError, ValueError):
+        return None
+
+
+def normalize_norms_method(method: str | None) -> NormsMethod:
+    m = (method or "wesp").strip().lower()
+    if m not in VALID_NORMS_METHODS:
+        return "wesp"
+    return m  # type: ignore[return-value]
+
+
+@dataclass
+class NormsResolveRequest:
+    method: NormsMethod = "wesp"
+    stored: dict[str, dict[str, float | None]] = field(default_factory=dict)
+    mass_kg: float | None = None
+    milk_yield_kg: float | None = None
+    ration_type: str | None = None
+    force_dynamic: bool = False
+    params: NormsParams = field(default_factory=NormsParams)
+
+
+def merge_hybrid_norms(
+    racion_resolved: dict[str, dict[str, float | None]],
+    stored: dict[str, dict[str, float | None]],
+    *,
+    dynamic_meta: dict[str, Any] | None = None,
+) -> tuple[dict[str, dict[str, float | None]], dict[str, Any]]:
+    """
+    RACION + derived имеют приоритет по min.
+    stored дополняет ключи без RACION-min; max всегда из stored, если задан.
+    """
+    out: dict[str, dict[str, float | None]] = {}
+    coverage: dict[str, list[str]] = {
+        "racion": [],
+        "derived": [],
+        "fallback": [],
+        "missing": [],
+    }
+    dynamic = dynamic_meta or {}
+    all_keys = {d["key"] for d in RATION_ALL_INDICATORS}
+    all_keys.update(racion_resolved.keys())
+    all_keys.update(stored.keys())
+
+    for key in sorted(all_keys):
+        rac = racion_resolved.get(key) or {}
+        st = stored.get(key) or {}
+        rac_min = rac.get("min")
+        st_min = st.get("min")
+        st_max = st.get("max")
+
+        source: str | None = None
+        min_v: float | None = None
+        if rac_min is not None:
+            min_v = rac_min
+            src = (dynamic.get(key) or {}).get("source")
+            source = "derived" if src == "derived" else "racion"
+        elif st_min is not None:
+            min_v = st_min
+            source = "fallback"
+
+        max_v = st_max if st_max is not None else rac.get("max")
+        if min_v is not None or max_v is not None:
+            out[key] = {"min": min_v, "max": max_v}
+        if source == "racion":
+            coverage["racion"].append(key)
+        elif source == "derived":
+            coverage["derived"].append(key)
+        elif source == "fallback":
+            coverage["fallback"].append(key)
+        elif key in {d["key"] for d in RATION_ALL_INDICATORS}:
+            coverage["missing"].append(key)
+
+    coverage["withMin"] = len([k for k, b in out.items() if b.get("min") is not None])
+    coverage["total"] = len(RATION_ALL_INDICATORS)
+    return out, coverage
+
+
+def _require_dairy_params(req: NormsResolveRequest) -> tuple[float, float]:
+    if req.mass_kg is None or req.mass_kg <= 0:
+        raise ValueError("Для методики Москва/Петербург укажите живую массу, кг")
+    if req.milk_yield_kg is None or req.milk_yield_kg <= 0:
+        raise ValueError("Для методики Москва/Петербург укажите суточный удой, кг")
+    return float(req.mass_kg), float(req.milk_yield_kg)
+
+
+def _moscow_params(req: NormsResolveRequest) -> MoscowDairyParams:
+    mass, milk = _require_dairy_params(req)
+    p = req.params
+    return MoscowDairyParams(
+        mass_kg=mass,
+        milk_yield_kg=milk,
+        milk_fat_pct=p.milk_fat_pct if p.milk_fat_pct is not None else 4.0,
+        lactation_no=p.lactation_no if p.lactation_no is not None else 2,
+        body_condition=p.body_condition if p.body_condition is not None else 1,
+        housing_system=p.housing_system if p.housing_system is not None else 1,
+    )
+
+
+def _piter_params(req: NormsResolveRequest) -> PiterDairyParams:
+    mass, milk = _require_dairy_params(req)
+    p = req.params
+    konc = p.konc_oe_sv
+    if konc is None or konc <= 0:
+        raise ValueError("Для методики Петербург укажите концентрацию ОЭ/СВ, МДж/кг СВ")
+    return PiterDairyParams(
+        mass_kg=mass,
+        milk_yield_kg=milk,
+        milk_fat_pct=p.milk_fat_pct if p.milk_fat_pct is not None else 4.0,
+        lactation_no=p.lactation_no if p.lactation_no is not None else 2,
+        body_condition=p.body_condition if p.body_condition is not None else 1,
+        housing_system=p.housing_system if p.housing_system is not None else 1,
+        konc_oe_sv=float(konc),
+    )
+
+
+def resolve_norms(req: NormsResolveRequest) -> tuple[dict[str, dict[str, float | None]], dict[str, Any]]:
+    method = normalize_norms_method(req.method)
+    if method == "wesp":
+        resolved, dynamic = apply_dynamic_norms(
+            req.stored,
+            mass_kg=req.mass_kg,
+            milk_yield_kg=req.milk_yield_kg,
+            ration_type=req.ration_type,
+            force_dynamic=req.force_dynamic,
+        )
+        return resolved, {"normsMethod": "wesp", "dynamicNorms": dynamic}
+
+    try:
+        if method == "racion_moscow":
+            racion, pack = resolve_moscow_dairy_norms(_moscow_params(req))
+        else:
+            racion, pack = resolve_piter_dairy_norms(_piter_params(req))
+    except PiterPrepError as exc:
+        raise ValueError(str(exc)) from exc
+
+    resolved, coverage = merge_hybrid_norms(racion, req.stored, dynamic_meta=pack.get("dynamic"))
+    return resolved, {"normsMethod": method, "coverage": coverage, **pack}

WESP_REL/app/lab/calc/nutrients.py

@@ -0,0 +1,135 @@
+from __future__ import annotations
+
+from typing import Any, Mapping
+
+from app.lab.nutrient_schema import resolve_sv_g_per_kg
+
+
+def normalize_key(value: str) -> str:
+    return " ".join((value or "").split()).strip().lower()
+
+
+def parse_num(value: Any) -> float | None:
+    if value is None or value == "":
+        return None
+    try:
+        n = float(value)
+    except (TypeError, ValueError):
+        return None
+    return None if n != n else n
+
+
+def get_nutrient_value(
+    dry_matter: float | None,
+    nutrients: Mapping[str, Any] | None,
+    nutrient_keys: list[str],
+    *,
+    indicator_key: str | None = None,
+) -> float | None:
+    """Только точное совпадение ключа (заголовок или indicator_key slug)."""
+    nutrients = nutrients or {}
+    if indicator_key:
+        for k, v in nutrients.items():
+            if normalize_key(str(k)) == normalize_key(indicator_key):
+                n = parse_num(v)
+                if n is not None:
+                    return n
+    for search in nutrient_keys:
+        target = normalize_key(search)
+        for k, v in nutrients.items():
+            if normalize_key(str(k)) != target:
+                continue
+            n = parse_num(v)
+            if n is not None:
+                return n
+    if any(normalize_key(k) == "св" for k in nutrient_keys):
+        return resolve_sv_g_per_kg(nutrients, dry_matter)
+    return None
+
+
+def weighted_average(
+    lines: list[dict[str, Any]],
+    total_kg: float,
+    nutrient_keys: list[str],
+    *,
+    indicator_key: str | None = None,
+) -> float | None:
+    if total_kg <= 0:
+        return None
+    total = 0.0
+    weight = 0.0
+    for line in lines:
+        kg = float(line.get("daily_kg") or 0)
+        if kg <= 0:
+            continue
+        v = get_nutrient_value(
+            line.get("dry_matter"),
+            line.get("nutrients"),
+            nutrient_keys,
+            indicator_key=indicator_key,
+        )
+        if v is None:
+            continue
+        total += kg * v
+        weight += kg
+    if weight <= 0:
+        return None
+    return total / weight
+
+
+def daily_intake_total(
+    lines: list[dict[str, Any]],
+    nutrient_keys: list[str],
+    *,
+    heads_per_trip: int = 1,
+    indicator_key: str | None = None,
+) -> float | None:
+    """Суточная доза на голову (г или МДж): Σ (кг/день/гол × г/кг). Как Excel Рацион КРС."""
+    heads = max(int(heads_per_trip or 1), 1)
+    total = 0.0
+    any_value = False
+    for line in lines:
+        herd_kg = float(line.get("daily_kg") or 0)
+        if herd_kg <= 0:
+            continue
+        kg = herd_kg / heads
+        v = get_nutrient_value(
+            line.get("dry_matter"),
+            line.get("nutrients"),
+            nutrient_keys,
+            indicator_key=indicator_key,
+        )
+        if v is None:
+            continue
+        total += kg * v
+        any_value = True
+    return total if any_value else None
+
+
+def rnb(
+    crude_protein: float | None,
+    usp: float | None,
+) -> float | None:
+    """RNB (ruminal nitrogen balance, GfE): (сырой протеин − уСП) / 6,25."""
+    if crude_protein is None or usp is None:
+        return None
+    return (crude_protein - usp) / 6.25
+
+
+def bra_rnb(crude_protein: float | None, usp: float | None) -> float | None:
+    """Deprecated alias for :func:`rnb`."""
+    return rnb(crude_protein, usp)
+
+
+def norm_diff(
+    content: float | None,
+    min_val: float | None,
+    max_val: float | None,
+) -> float | None:
+    if content is None:
+        return None
+    if min_val is not None and content < min_val:
+        return content - min_val
+    if max_val is not None and content > max_val:
+        return content - max_val
+    return 0.0

WESP_REL/app/lab/calc/racion/__init__.py

@@ -0,0 +1 @@
+"""Российские методики суточных норм (Москва / Петербург)."""

WESP_REL/app/lab/calc/racion/interp.py

@@ -0,0 +1,25 @@
+"""Линейная интерполяция (аналог FRAC в методичке)."""
+
+from __future__ import annotations
+
+
+def frac(numerator: float, denominator: float) -> float:
+    if denominator == 0:
+        return 0.0
+    return numerator / denominator
+
+
+def lerp(x: float, x1: float, n1: float, x2: float, n2: float) -> float:
+    """Norma = n1 + frac(n2 - n1, x2 - x1) * (x - x1)."""
+    if x2 == x1:
+        return n1
+    return n1 + frac(n2 - n1, x2 - x1) * (x - x1)
+
+
+def popr_index(udoy: float, boundaries: list[float]) -> int:
+    """1-based индекс столбца POPR_K по суточному удою."""
+    idx = 1
+    for i, bound in enumerate(boundaries, start=1):
+        if udoy >= bound:
+            idx = i + 1
+    return min(idx, len(boundaries) + 1)

WESP_REL/app/lab/calc/racion/moscow.py

@@ -0,0 +1,107 @@
+"""Методика Москва — лактирующие коровы (NORM_1_1_CALC)."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any
+
+from app.lab.calc.racion.interp import popr_index
+from app.lab.calc.norms_derived import apply_derived_norms
+from app.lab.calc.racion.npitv_map import DAIRY_LACTIR_NPITV, NPITV_TO_INDICATOR
+from app.lab.calc.racion.tables import load_moskwa_lactir
+
+
+@dataclass(frozen=True)
+class MoscowDairyParams:
+    mass_kg: float
+    milk_yield_kg: float
+    milk_fat_pct: float = 4.0
+    lactation_no: int = 2
+    body_condition: int = 1
+    housing_system: int = 1
+
+
+def _row_lookup(rows: list[dict], npitv: int) -> dict | None:
+    for row in rows:
+        if row["npitv"] == npitv and row["pom"] == 1:
+            return row
+    return None
+
+
+def _popr_value(row: dict, udoy: float, boundaries: list[float]) -> tuple[float | None, float | None]:
+    popr = row.get("popr_k") or []
+    if not popr:
+        return None, None
+    idx = popr_index(udoy, boundaries) - 1
+    idx = max(0, min(idx, len(popr) - 1))
+    p_k = popr[idx]
+    koef = float(row.get("koef") or 0)
+    return p_k, koef
+
+
+def compute_moscow_norm(npitv: int, params: MoscowDairyParams) -> float | None:
+    data = load_moskwa_lactir()
+    boundaries = data.get("udoy_boundaries") or []
+    row = _row_lookup(data.get("rows") or [], npitv)
+    if row is None:
+        return None
+
+    mass = params.mass_kg
+    udoy = params.milk_yield_kg
+    jir = params.milk_fat_pct
+    wmassa = mass * 1.02 if params.body_condition > 1 else mass
+
+    p_k, koef = _popr_value(row, udoy, boundaries)
+    if p_k is None:
+        return None
+
+    norma: float | None = None
+    if npitv == 1:
+        temp = 0.005 if udoy <= 22 else 0.0025
+        norma = temp * (wmassa - 500) + p_k * udoy - ((4 - jir) * udoy) / 148
+    elif npitv == 2:
+        temp = 0.09 if udoy <= 22 else 0.065
+        norma = temp * (wmassa - 500) + p_k * udoy - ((4 - jir) * udoy) / 15
+    elif npitv == 3:
+        temp = 0.017 if udoy <= 22 else 0.015
+        norma = temp * (wmassa - 500) + p_k * udoy
+    elif 4 <= npitv <= 24:
+        norma = p_k * udoy + (wmassa - 500) * koef
+        if npitv == 10:
+            norma *= 0.393
+    else:
+        return None
+
+    if norma is None:
+        return None
+    if params.lactation_no == 1:
+        norma *= 0.95
+    elif params.lactation_no == 3:
+        norma *= 1.05
+    if params.housing_system == 2:
+        norma *= 1.1
+    return round(norma, 3)
+
+
+def resolve_moscow_dairy_norms(params: MoscowDairyParams) -> tuple[dict[str, dict[str, float | None]], dict[str, Any]]:
+    resolved: dict[str, dict[str, float | None]] = {}
+    dynamic: dict[str, Any] = {}
+    for npitv in DAIRY_LACTIR_NPITV:
+        value = compute_moscow_norm(npitv, params)
+        if value is None:
+            continue
+        key = NPITV_TO_INDICATOR.get(npitv)
+        if not key:
+            continue
+        resolved[key] = {"min": value, "max": None}
+        dynamic[key] = {"min": value, "npitv": npitv, "method": "racion_moscow", "source": "racion"}
+    resolved, derived_dyn = apply_derived_norms(resolved, mass_kg=params.mass_kg)
+    for k, v in derived_dyn.items():
+        dynamic[k] = {**v, "method": "racion_moscow", "source": "derived"}
+    meta = {
+        "method": "racion_moscow",
+        "massKg": params.mass_kg,
+        "milkYieldKg": params.milk_yield_kg,
+        "milkFatPct": params.milk_fat_pct,
+    }
+    return resolved, {"meta": meta, "dynamic": dynamic}

WESP_REL/app/lab/calc/racion/npitv_map.py

@@ -0,0 +1,42 @@
+"""NPitV (справочник питательных веществ) → ключи показателей WESP."""
+
+from __future__ import annotations
+
+# NORMY_MOSKWA_LACTIR / NORM_1_1_CALC: NPitV 1–24 (pom=1)
+NPITV_TO_INDICATOR: dict[int, str] = {
+    1: "feed_units",
+    2: "oe",
+    3: "dry_matter",
+    4: "crude_protein",
+    5: "digestible_protein",
+    6: "crude_fat",
+    7: "nel",
+    8: "rnb",
+    9: "usp",
+    10: "sodium",
+    11: "magnesium",
+    12: "starch",
+    13: "potassium",
+    14: "calcium",
+    15: "phosphorus",
+    16: "iron",
+    17: "copper",
+    18: "zinc",
+    19: "manganese",
+    20: "cobalt",
+    21: "iodine",
+    22: "carotene",
+    23: "vitamin_d",
+    24: "vitamin_e",
+}
+
+DAIRY_LACTIR_NPITV: tuple[int, ...] = tuple(range(1, 25))
+
+# Обратная совместимость
+DAIRY_CORE_NPITV: tuple[int, ...] = (2, 3, 4, 5, 7, 8, 9, 10, 12, 14, 15)
+
+INDICATOR_TO_NPITV: dict[str, int] = {v: k for k, v in NPITV_TO_INDICATOR.items()}
+
+
+def indicator_for_npitv(npitv: int) -> str | None:
+    return NPITV_TO_INDICATOR.get(npitv)

WESP_REL/app/lab/calc/racion/piter.py

@@ -0,0 +1,193 @@
+"""Методика Петербург — лактирующие коровы (NORM_1_2_CALC)."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any
+
+from app.lab.calc.norms_derived import apply_derived_norms
+from app.lab.calc.racion.interp import lerp
+from app.lab.calc.racion.moscow import MoscowDairyParams
+from app.lab.calc.racion.npitv_map import DAIRY_LACTIR_NPITV, NPITV_TO_INDICATOR
+from app.lab.calc.racion.piter_prep import PiterPrepError, PiterPrepResult, prepare_piter_calc
+from app.lab.calc.racion.tables import load_piter_lactir
+
+
+@dataclass(frozen=True)
+class PiterDairyParams(MoscowDairyParams):
+    konc_oe_sv: float = 10.3
+
+
+def _konc_bracket(konc: float, konc_values: list[float]) -> tuple[float, float]:
+    sorted_k = sorted(set(konc_values))
+    positive = [k for k in sorted_k if k > 0]
+    if not positive:
+        return konc, konc
+    if konc <= positive[0]:
+        return positive[0], positive[min(1, len(positive) - 1)]
+    for i in range(len(positive) - 1):
+        if positive[i] <= konc <= positive[i + 1]:
+            return positive[i], positive[i + 1]
+    return positive[-2], positive[-1]
+
+
+def _udoy_bracket(udoy_jir: float, udoys: list[float]) -> tuple[float, float]:
+    sorted_u = sorted(set(udoys))
+    if len(sorted_u) < 2:
+        return sorted_u[0], sorted_u[0]
+    if udoy_jir <= sorted_u[0]:
+        return sorted_u[0], sorted_u[1]
+    for i in range(len(sorted_u) - 1):
+        if sorted_u[i] <= udoy_jir < sorted_u[i + 1]:
+            return sorted_u[i], sorted_u[i + 1]
+    return sorted_u[-2], sorted_u[-1]
+
+
+def _norm_at_mass(
+    entry_normy: list[float],
+    mass_ind: int,
+    wmassa: float,
+    m_a: float,
+    m_b: float,
+) -> float | None:
+    i = mass_ind - 1
+    if i < 0 or i + 1 >= len(entry_normy):
+        return None
+    n_a, n_b = entry_normy[i], entry_normy[i + 1]
+    if n_a <= 0 or n_b <= 0:
+        return None
+    return lerp(wmassa, m_a, n_a, m_b, n_b)
+
+
+def _entries_for(data: dict, npitv: int) -> list[dict]:
+    npv = 4 if npitv == 5 else npitv
+    return [e for e in data.get("entries") or [] if e["npitv"] == npv]
+
+
+def _compute_with_konc(
+    entries: list[dict],
+    npitv: int,
+    params: PiterDairyParams,
+    prep: PiterPrepResult,
+    konc_pred: float,
+    konc_sled: float,
+) -> float | None:
+    by_konc = {k: [e for e in entries if abs(e["konc"] - k) < 1e-6] for k in (konc_pred, konc_sled)}
+
+    def _at_konc(konc: float) -> float | None:
+        rows = by_konc.get(konc) or []
+        if not rows:
+            return None
+        udoys = [e["udoy"] for e in rows]
+        ud_pred, ud_sled = _udoy_bracket(prep.udoy_jir, udoys)
+        if ud_pred == ud_sled:
+            return None
+        e_pred = next((e for e in rows if e["udoy"] == ud_pred), None)
+        e_sled = next((e for e in rows if e["udoy"] == ud_sled), None)
+        if not e_pred or not e_sled:
+            return None
+        n01 = _norm_at_mass(e_pred["normy"], prep.mass_ind, prep.wmassa, prep.m_a, prep.m_b)
+        n02 = _norm_at_mass(e_sled["normy"], prep.mass_ind, prep.wmassa, prep.m_a, prep.m_b)
+        if n01 is None or n02 is None:
+            return None
+        return lerp(prep.udoy_jir, ud_pred, n01, ud_sled, n02)
+
+    norma1 = _at_konc(konc_pred)
+    norma2 = _at_konc(konc_sled)
+    if norma1 is None or norma2 is None or konc_pred == konc_sled:
+        return None
+    norma = lerp(params.konc_oe_sv, konc_pred, norma1, konc_sled, norma2)
+    if npitv == 5:
+        norma *= 0.65
+    return round(norma, 3)
+
+
+def _compute_konc_independent(
+    entries: list[dict],
+    params: PiterDairyParams,
+    prep: PiterPrepResult,
+) -> float | None:
+    rows = [e for e in entries if e["konc"] == -10]
+    if not rows:
+        return None
+    udoys = [e["udoy"] for e in rows]
+    ud_pred, ud_sled = _udoy_bracket(prep.udoy_jir, udoys)
+    if ud_pred == ud_sled:
+        return None
+    e_pred = next((e for e in rows if e["udoy"] == ud_pred), None)
+    e_sled = next((e for e in rows if e["udoy"] == ud_sled), None)
+    if not e_pred or not e_sled:
+        return None
+    n01 = _norm_at_mass(e_pred["normy"], prep.mass_ind, prep.wmassa, prep.m_a, prep.m_b)
+    n02 = _norm_at_mass(e_sled["normy"], prep.mass_ind, prep.wmassa, prep.m_a, prep.m_b)
+    if n01 is None or n02 is None:
+        return None
+    norma = lerp(prep.udoy_jir, ud_pred, n01, ud_sled, n02)
+    if params.housing_system == 2:
+        norma *= 1.1
+    return round(norma, 3)
+
+
+def compute_piter_norm(npitv: int, params: PiterDairyParams, prep: PiterPrepResult | None = None) -> float | None:
+    data = load_piter_lactir()
+    entries = _entries_for(data, npitv)
+    if not entries:
+        return None
+
+    if prep is None:
+        prep = prepare_piter_calc(
+            mass_kg=params.mass_kg,
+            milk_yield_kg=params.milk_yield_kg,
+            milk_fat_pct=params.milk_fat_pct,
+            konc_oe_sv=params.konc_oe_sv,
+            body_condition=params.body_condition,
+        )
+
+    konc_vals = sorted({e["konc"] for e in entries if e["konc"] > 0})
+    if konc_vals:
+        min_k = min(konc_vals)
+        if min_k > 0:
+            konc_pred, konc_sled = _konc_bracket(params.konc_oe_sv, konc_vals)
+            return _compute_with_konc(entries, npitv, params, prep, konc_pred, konc_sled)
+    return _compute_konc_independent(entries, params, prep)
+
+
+def resolve_piter_dairy_norms(params: PiterDairyParams) -> tuple[dict[str, dict[str, float | None]], dict[str, Any]]:
+    prep = prepare_piter_calc(
+        mass_kg=params.mass_kg,
+        milk_yield_kg=params.milk_yield_kg,
+        milk_fat_pct=params.milk_fat_pct,
+        konc_oe_sv=params.konc_oe_sv,
+        body_condition=params.body_condition,
+    )
+    resolved: dict[str, dict[str, float | None]] = {}
+    dynamic: dict[str, Any] = {}
+    for npitv in DAIRY_LACTIR_NPITV:
+        value = compute_piter_norm(npitv, params, prep=prep)
+        if value is None:
+            continue
+        key = NPITV_TO_INDICATOR.get(npitv)
+        if not key:
+            continue
+        resolved[key] = {"min": value, "max": None}
+        dynamic[key] = {"min": value, "npitv": npitv, "method": "racion_piter", "source": "racion"}
+    resolved, derived_dyn = apply_derived_norms(resolved, mass_kg=params.mass_kg)
+    for k, v in derived_dyn.items():
+        dynamic[k] = {**v, "method": "racion_piter", "source": "derived"}
+    meta = {
+        "method": "racion_piter",
+        "massKg": params.mass_kg,
+        "milkYieldKg": params.milk_yield_kg,
+        "koncOeSv": params.konc_oe_sv,
+        "prep": {
+            "massInd": prep.mass_ind,
+            "mA": prep.m_a,
+            "mB": prep.m_b,
+            "koncPred": prep.konc_pred,
+            "koncSled": prep.konc_sled,
+        },
+    }
+    return resolved, {"meta": meta, "dynamic": dynamic}
+
+
+__all__ = ["PiterDairyParams", "PiterPrepError", "compute_piter_norm", "resolve_piter_dairy_norms"]

WESP_REL/app/lab/calc/racion/piter_prep.py

@@ -0,0 +1,202 @@
+"""Подготовка параметров NORM_1_2_PREP (интервалы массы и концентрации)."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+
+from app.lab.calc.racion.interp import lerp
+from app.lab.calc.racion.tables import load_normy_info
+
+
+@dataclass(frozen=True)
+class PiterPrepResult:
+    mass_ind: int
+    konc_pred: float
+    konc_sled: float
+    m_a: float
+    m_b: float
+    udoy_jir: float
+    wmassa: float
+
+
+class PiterPrepError(ValueError):
+    def __init__(self, code: int, message: str, *, info: str | None = None) -> None:
+        super().__init__(message)
+        self.code = code
+        self.info = info
+
+
+def _info_values(nperem: int) -> list[float]:
+    data = load_normy_info()
+    for row in data.get("rows") or []:
+        if row.get("nperem") == nperem:
+            return list(row.get("znachenie") or [])
+    return []
+
+
+def _fl_from_str(values: list[float], index: int) -> float | None:
+    """1-based index как FlFromStr в RACION."""
+    i = index - 1
+    if i < 0 or i >= len(values):
+        return None
+    return float(values[i])
+
+
+def _wmassa(mass_kg: float, body_condition: int) -> float:
+    if body_condition > 1:
+        return mass_kg * 1.02
+    return mass_kg
+
+
+def prepare_piter_calc(
+    *,
+    mass_kg: float,
+    milk_yield_kg: float,
+    milk_fat_pct: float,
+    konc_oe_sv: float,
+    body_condition: int = 1,
+) -> PiterPrepResult:
+    """Порт NORM_1_2_PREP: интервалы для NORM_1_2_CALC."""
+    if milk_yield_kg <= 0:
+        raise PiterPrepError(-11, "Укажите суточный удой, кг")
+    if milk_fat_pct <= 0:
+        raise PiterPrepError(-12, "Укажите жирность молока, %")
+    if mass_kg <= 0:
+        raise PiterPrepError(-14, "Укажите живую массу, кг")
+    if konc_oe_sv <= 0:
+        raise PiterPrepError(-15, "Укажите концентрацию ОЭ/СВ, МДж/кг СВ")
+
+    wmassa = _wmassa(mass_kg, body_condition)
+    udoy_jir = milk_yield_kg * milk_fat_pct * 0.25
+
+    kol_konc_vals = _info_values(7)
+    kol_mass_vals = _info_values(7)
+    if len(kol_konc_vals) < 2 or len(kol_mass_vals) < 2:
+        raise PiterPrepError(-1, "Справочник NORMY_INFO (NPerem=7) не задан")
+    kol_konc = int(kol_konc_vals[0])
+    kol_mass = int(kol_mass_vals[1])
+    if kol_konc < 2 or kol_mass < 2:
+        raise PiterPrepError(-1, "Некорректные размеры таблицы концентраций/масс")
+
+    masses = _info_values(6) or _info_values(14)
+    koncss = _info_values(3)
+    if not masses or not koncss or len(masses) < 2 or len(koncss) < 2:
+        raise PiterPrepError(-1, "Справочник NORMY_INFO: массы или концентрации не заданы")
+
+    # Интервал концентрации
+    konc_ind = 1
+    for i in range(2, kol_konc):
+        v = _fl_from_str(koncss, i)
+        if v is not None and konc_oe_sv >= v:
+            konc_ind = i
+    konc_pred = _fl_from_str(koncss, konc_ind)
+    konc_sled = _fl_from_str(koncss, konc_ind + 1)
+    if konc_pred is None or konc_sled is None or konc_pred < 0 or konc_sled < 0 or konc_pred == konc_sled:
+        raise PiterPrepError(-1, "Не удалось определить интервал концентрации")
+
+    # Интервал массы
+    mass_ind = 1
+    for i in range(2, kol_mass):
+        v = _fl_from_str(masses, i)
+        if v is not None and wmassa >= v:
+            mass_ind = i
+    m_a = _fl_from_str(masses, mass_ind)
+    m_b = _fl_from_str(masses, mass_ind + 1)
+    if m_a is None or m_b is None or m_a < 0 or m_b < 0 or m_a == m_b:
+        raise PiterPrepError(-1, "Не удалось определить интервал массы")
+
+    # Проверка удоя (NPerem=4,5)
+    udoy_str_4 = _info_values(4)
+    udoy_str_5 = _info_values(5)
+    udoy_min = _fl_from_str(udoy_str_4, 1) if udoy_str_4 else None
+    udoy_max = _fl_from_str(udoy_str_5, kol_konc) if udoy_str_5 else None
+
+    if udoy_min is not None and udoy_max is not None:
+        if udoy_jir < udoy_min or udoy_jir > udoy_max:
+            jir_str = _info_values(2)
+            gr_udoy1 = max(udoy_min * 4 / milk_fat_pct, udoy_min)
+            gr_udoy2 = min(udoy_max * 4 / milk_fat_pct, udoy_max)
+            gr_jir1 = udoy_min * 4 / milk_yield_kg
+            gr_jir2 = udoy_max * 4 / milk_yield_kg
+            if jir_str:
+                if len(jir_str) >= 1:
+                    gr_jir1 = max(gr_jir1, jir_str[0])
+                if len(jir_str) >= 2:
+                    gr_jir2 = min(gr_jir2, jir_str[1])
+            code = -4 if udoy_jir < udoy_min else -5
+            info = f"{gr_udoy1:.4f};{gr_udoy2:.4f};{gr_jir1:.3f};{gr_jir2:.3f};"
+            raise PiterPrepError(code, "Удой вне допустимого диапазона для жирности", info=info)
+
+        # Допустимый диапазон концентрации для udoy_jir
+        i = 1
+        if udoy_jir >= (_fl_from_str(udoy_str_4, 1) or 0):
+            while i < kol_konc - 1:
+                nxt = _fl_from_str(udoy_str_4, i + 1)
+                if nxt is None or nxt > udoy_jir:
+                    break
+                i += 1
+        else:
+            while i < kol_konc - 1:
+                nxt = _fl_from_str(udoy_str_4, i + 1)
+                cur = _fl_from_str(udoy_str_4, i)
+                if nxt is None or cur is None or nxt != cur:
+                    break
+                i += 1
+        ud_a = _fl_from_str(udoy_str_4, i)
+        ud_b = _fl_from_str(udoy_str_4, i + 1)
+        konc_a = _fl_from_str(koncss, i)
+        konc_b = _fl_from_str(koncss, i + 1)
+        if ud_a is not None and ud_b is not None and konc_a is not None and konc_b is not None:
+            if ud_a == ud_b:
+                konc_max = konc_b
+            else:
+                konc_max = lerp(udoy_jir, ud_a, konc_a, ud_b, konc_b)
+
+            i = kol_konc
+            if udoy_jir <= (_fl_from_str(udoy_str_5, kol_konc) or udoy_jir):
+                while i > 2:
+                    prev = _fl_from_str(udoy_str_5, i - 1)
+                    if prev is None or prev < udoy_jir:
+                        break
+                    i -= 1
+            else:
+                while i > 2:
+                    prev = _fl_from_str(udoy_str_5, i - 1)
+                    last = _fl_from_str(udoy_str_5, kol_konc)
+                    if prev is None or last is None or prev != last:
+                        break
+                    i -= 1
+            ud_a2 = _fl_from_str(udoy_str_5, i - 1)
+            ud_b2 = _fl_from_str(udoy_str_5, i)
+            konc_a2 = _fl_from_str(koncss, i - 1)
+            konc_b2 = _fl_from_str(koncss, i)
+            if ud_a2 is not None and ud_b2 is not None and konc_a2 is not None and konc_b2 is not None:
+                if ud_a2 == ud_b2:
+                    konc_min = konc_b2
+                else:
+                    konc_min = lerp(udoy_jir, ud_a2, konc_a2, ud_b2, konc_b2)
+
+                konc_lo = _fl_from_str(koncss, 1) or konc_min
+                konc_hi = _fl_from_str(koncss, kol_konc) or konc_max
+                if konc_min < konc_lo:
+                    konc_min = konc_lo
+                if konc_max > konc_hi:
+                    konc_max = konc_hi
+                if round(konc_oe_sv, 1) < round(konc_min, 1) or round(konc_oe_sv, 1) > round(konc_max, 1):
+                    code = -2 if round(konc_oe_sv, 1) < round(konc_min, 1) else -3
+                    info = f"{round(konc_min, 1)};{round(konc_max, 1)};"
+                    raise PiterPrepError(
+                        code,
+                        f"Концентрация ОЭ/СВ вне допустимого диапазона ({round(konc_min, 1)}–{round(konc_max, 1)})",
+                        info=info,
+                    )
+
+    return PiterPrepResult(
+        mass_ind=mass_ind,
+        konc_pred=float(konc_pred),
+        konc_sled=float(konc_sled),
+        m_a=float(m_a),
+        m_b=float(m_b),
+        udoy_jir=udoy_jir,
+        wmassa=wmassa,
+    )

WESP_REL/app/lab/calc/racion/tables.py

@@ -0,0 +1,75 @@
+"""Загрузка справочников норм RACION (БД → JSON fallback)."""
+
+from __future__ import annotations
+
+import json
+from functools import lru_cache
+from pathlib import Path
+
+_SEED_DIR = Path(__file__).resolve().parents[4] / "data" / "seed" / "racion"
+
+
+def _read_json(name: str) -> dict:
+    path = _SEED_DIR / name
+    if not path.exists():
+        raise FileNotFoundError(f"Справочник не найден: {path}")
+    return json.loads(path.read_text(encoding="utf-8"))
+
+
+def _load_from_db(loader_name: str) -> dict | None:
+    try:
+        from flask import has_app_context
+
+        if not has_app_context():
+            return None
+        from app.lab.services.racion_reference import (
+            load_moskwa_lactir_from_db,
+            load_normy_info_from_db,
+            load_piter_lactir_from_db,
+        )
+
+        loaders = {
+            "moskwa": load_moskwa_lactir_from_db,
+            "piter": load_piter_lactir_from_db,
+            "info": load_normy_info_from_db,
+        }
+        fn = loaders.get(loader_name)
+        if fn is None:
+            return None
+        data = fn()
+        return data if data else None
+    except Exception:
+        return None
+
+
+@lru_cache(maxsize=1)
+def load_moskwa_lactir() -> dict:
+    data = _load_from_db("moskwa")
+    if data:
+        return data
+    return _read_json("moskwa_lactir.json")
+
+
+@lru_cache(maxsize=1)
+def load_piter_lactir() -> dict:
+    data = _load_from_db("piter")
+    if data:
+        return data
+    return _read_json("piter_lactir.json")
+
+
+@lru_cache(maxsize=1)
+def load_normy_info() -> dict:
+    data = _load_from_db("info")
+    if data:
+        return data
+    try:
+        return _read_json("normy_info.json")
+    except FileNotFoundError:
+        return {"rows": [], "mass_kg_values": [400, 450, 500, 550, 600, 650, 700, 750]}
+
+
+def clear_tables_cache() -> None:
+    load_moskwa_lactir.cache_clear()
+    load_piter_lactir.cache_clear()
+    load_normy_info.cache_clear()