index/index1c/parse_meta.py

"""Разбор XML метаданных объектов 1С (выгрузка EDT)."""

from __future__ import annotations

import re
import xml.etree.ElementTree as ET
from dataclasses import asdict, dataclass, field
from pathlib import Path
from typing import Any, Iterable

from .config import FIELD_RICH_TYPES, OBJECT_TYPE_FOLDERS

MD = "{http://v8.1c.ru/8.3/MDClasses}"
V8 = "{http://v8.1c.ru/8.1/data/core}"

REF_TYPE_RE = re.compile(
    r"^cfg:(CatalogRef|DocumentRef|EnumRef|ChartOfCharacteristicTypesRef|"
    r"ChartOfAccountsRef|ChartOfCalculationTypesRef|ExchangePlanRef|"
    r"BusinessProcessRef|TaskRef|DefinedType)\.(.+)$"
)

FIELD_TAGS = frozenset(
    {
        "Attribute",
        "Dimension",
        "Resource",
        "TabularSection",
        "Column",
        "EnumValue",
        "AddressingAttribute",
    }
)


def local(tag: str) -> str:
    return tag.split("}")[-1] if "}" in tag else tag


def text_of(el: ET.Element | None) -> str:
    if el is None:
        return ""
    return (el.text or "").strip()


def child_text(props: ET.Element, tag: str) -> str:
    node = props.find(f"{MD}{tag}")
    if node is None:
        # без namespace — на всякий случай
        for ch in props:
            if local(ch.tag) == tag:
                return (ch.text or "").strip()
        return ""
    return (node.text or "").strip()


def collect_localized(props: ET.Element, tag: str) -> dict[str, str]:
    """Синоним / подсказка: lang → content."""
    node = props.find(f"{MD}{tag}")
    if node is None:
        return {}
    out: dict[str, str] = {}
    for item in node.findall(f"{V8}item"):
        lang = text_of(item.find(f"{V8}lang")) or "ru"
        content = text_of(item.find(f"{V8}content"))
        if content:
            out[lang] = content
    return out


def synonym_str(loc: dict[str, str]) -> str:
    if not loc:
        return ""
    if "ru" in loc:
        parts = [loc["ru"]] + [f"{k}:{v}" for k, v in loc.items() if k != "ru"]
        return " | ".join(parts)
    return " | ".join(f"{k}:{v}" for k, v in loc.items())


def collect_types(props: ET.Element) -> list[str]:
    typ = props.find(f"{MD}Type")
    if typ is None:
        return []
    types: list[str] = []
    for t in typ.findall(f".//{V8}Type"):
        raw = (t.text or "").strip()
        if raw:
            types.append(raw)
    for t in typ.findall(f".//{V8}TypeSet"):
        raw = (t.text or "").strip()
        if raw:
            types.append(raw)
    return types


def parse_ref_targets(types: Iterable[str]) -> list[dict[str, str]]:
    refs: list[dict[str, str]] = []
    for t in types:
        m = REF_TYPE_RE.match(t)
        if m:
            refs.append({"kind": m.group(1), "name": m.group(2), "raw": t})
        elif t.startswith("cfg:") and "Ref." in t:
            # прочие cfg:*Ref.*
            body = t[4:]
            if "." in body:
                kind, name = body.split(".", 1)
                refs.append({"kind": kind, "name": name, "raw": t})
    return refs


@dataclass
class FieldRec:
    kind: str  # Attribute | Dimension | Resource | Column | EnumValue | ...
    name: str
    synonym: str = ""
    comment: str = ""
    tooltip: str = ""
    types: list[str] = field(default_factory=list)
    refs: list[dict[str, str]] = field(default_factory=list)
    tabular_section: str = ""
    fill_checking: str = ""  # ShowError | DontCheck | …


@dataclass
class ObjectRec:
    config: str
    object_type: str
    name: str
    synonym: str = ""
    comment: str = ""
    tooltip: str = ""
    explanation: str = ""
    uuid: str = ""
    path: str = ""
    fields: list[FieldRec] = field(default_factory=list)
    owners: list[str] = field(default_factory=list)
    register_records: list[str] = field(default_factory=list)
    based_on: list[str] = field(default_factory=list)
    input_by_string: list[str] = field(default_factory=list)

    @property
    def full_name(self) -> str:
        return f"{self.object_type}.{self.name}"

    def search_text(self) -> str:
        parts = [
            self.object_type,
            self.name,
            self.synonym,
            self.comment,
            self.tooltip,
            self.explanation,
        ]
        for f in self.fields:
            parts.extend([f.kind, f.name, f.synonym, f.comment, f.tooltip, f.tabular_section])
            parts.extend(f.types)
        parts.extend(self.owners)
        parts.extend(self.register_records)
        parts.extend(self.based_on)
        return "\n".join(p for p in parts if p)

    def to_dict(self) -> dict[str, Any]:
        d = asdict(self)
        d["full_name"] = self.full_name
        return d


def _parse_field_props(props: ET.Element, kind: str, tabular: str = "") -> FieldRec | None:
    name = child_text(props, "Name")
    if not name:
        return None
    types = collect_types(props)
    return FieldRec(
        kind=kind,
        name=name,
        synonym=synonym_str(collect_localized(props, "Synonym")),
        comment=child_text(props, "Comment"),
        tooltip=synonym_str(collect_localized(props, "ToolTip")),
        types=types,
        refs=parse_ref_targets(types),
        tabular_section=tabular,
        fill_checking=child_text(props, "FillChecking"),
    )


def _walk_child_objects(co: ET.Element | None, object_type: str) -> list[FieldRec]:
    if co is None or object_type not in FIELD_RICH_TYPES:
        return []
    fields: list[FieldRec] = []

    for child in co:
        tag = local(child.tag)
        if tag == "TabularSection":
            ts_props = child.find(f"{MD}Properties")
            ts_name = child_text(ts_props, "Name") if ts_props is not None else ""
            if ts_props is not None and ts_name:
                fields.append(
                    FieldRec(
                        kind="TabularSection",
                        name=ts_name,
                        synonym=synonym_str(collect_localized(ts_props, "Synonym")),
                        comment=child_text(ts_props, "Comment"),
                    )
                )
            inner = child.find(f"{MD}ChildObjects")
            if inner is not None:
                for col in inner:
                    if local(col.tag) != "Attribute":
                        continue
                    cprops = col.find(f"{MD}Properties")
                    if cprops is None:
                        continue
                    rec = _parse_field_props(cprops, "Column", tabular=ts_name)
                    if rec:
                        fields.append(rec)
            continue

        if tag not in FIELD_TAGS:
            continue
        props = child.find(f"{MD}Properties")
        if props is None:
            continue
        rec = _parse_field_props(props, tag)
        if rec:
            fields.append(rec)

    return fields


_META_PREFIXES = (
    "Catalog.",
    "Document.",
    "InformationRegister.",
    "AccumulationRegister.",
    "Enum.",
    "Constant.",
    "ChartOfCharacteristicTypes.",
    "ExchangePlan.",
    "BusinessProcess.",
    "Task.",
    "DocumentJournal.",
)


def _list_from_props(props: ET.Element, tag: str) -> list[str]:
    node = props.find(f"{MD}{tag}")
    if node is None:
        return []
    items: list[str] = []
    for el in node.iter():
        t = (el.text or "").strip()
        if not t or "." not in t:
            continue
        if any(t.startswith(p) for p in _META_PREFIXES):
            items.append(t)
    seen: set[str] = set()
    out: list[str] = []
    for x in items:
        if x not in seen:
            seen.add(x)
            out.append(x)
    return out


def parse_metadata_xml(
    path: Path,
    *,
    config: str,
    object_type: str,
) -> ObjectRec | None:
    try:
        tree = ET.parse(path)
    except ET.ParseError:
        return None
    root = tree.getroot()

    # Корневой элемент объекта: <Catalog>, <Document>, ...
    obj_el: ET.Element | None = None
    for ch in root:
        if local(ch.tag) == object_type:
            obj_el = ch
            break
    if obj_el is None:
        for ch in root:
            if ch.find(f"{MD}Properties") is not None or ch.find("Properties") is not None:
                obj_el = ch
                break
    if obj_el is None:
        return None

    props = obj_el.find(f"{MD}Properties")
    if props is None:
        props = obj_el.find("Properties")
    if props is None:
        return None

    name = child_text(props, "Name")
    if not name:
        return None

    uuid = obj_el.attrib.get("uuid", "")

    fields = _walk_child_objects(obj_el.find(f"{MD}ChildObjects"), object_type)

    # Для констант / определяемых типов тип лежит в Properties
    if object_type in {"Constant", "DefinedType", "CommonAttribute", "SessionParameter"} and not fields:
        types = collect_types(props)
        if types:
            fields.append(
                FieldRec(
                    kind="ValueType",
                    name="Type",
                    types=types,
                    refs=parse_ref_targets(types),
                )
            )

    return ObjectRec(
        config=config,
        object_type=object_type,
        name=name,
        synonym=synonym_str(collect_localized(props, "Synonym")),
        comment=child_text(props, "Comment"),
        tooltip=synonym_str(collect_localized(props, "ToolTip")),
        explanation=synonym_str(collect_localized(props, "Explanation")),
        uuid=uuid,
        path=str(path),
        fields=fields,
        owners=_list_from_props(props, "Owners"),
        register_records=_list_from_props(props, "RegisterRecords"),
        based_on=_list_from_props(props, "BasedOn"),
        input_by_string=_list_from_props(props, "InputByString"),
    )


def iter_object_xml_files(
    src: Path,
    types: set[str] | None = None,
    skip_types: set[str] | None = None,
    *,
    source_kind: str = "src",
    external_name: str = "",
) -> list[tuple[str, Path]]:
    """Список (object_type, xml_path) для обычной выгрузки или external processing."""
    skip = skip_types or set()
    out: list[tuple[str, Path]] = []
    if not src.is_dir():
        return out
    if source_kind == "external_processor":
        otype = "ExternalDataProcessor"
        if otype in skip:
            return out
        if types is not None and otype not in types:
            return out
        if not external_name:
            return out
        xml = src / f"{external_name}.xml"
        if xml.is_file():
            out.append((otype, xml))
        return out
    for folder in sorted(src.iterdir()):
        if not folder.is_dir():
            continue
        otype = OBJECT_TYPE_FOLDERS.get(folder.name)
        if otype is None:
            continue
        if otype in skip:
            continue
        if types is not None and otype not in types:
            continue
        for xml in sorted(folder.glob("*.xml")):
            out.append((otype, xml))
    return out