system_prompt.py

"""Assemble the system prompt for one turn.

Per v4.1 §2.1 the prompt injects: (1) the agent's role/rules, (2) the effective
semantic layer for the current scope, (3) recalled facts, (4) DB schema. V1
keeps each section simple; later versions enrich them without changing the
loop. Sections are suppressed when empty.
"""

from __future__ import annotations

import json

from .context import HarnessContext

_BASE = """\
You are Lang2SQL, a read-only data analytics agent.

Rules:
- Only ever read data. Never modify the database.
- When you need data, call the run_sql tool with a single SELECT/WITH query.
- Discover schema with explore_schema before guessing table or column names.
- Prefer definitions from the semantic layer below over your own assumptions.
- Answer concisely. Show only the final successful SQL you ran, not intermediate attempts.

Clarification rule:
- If the user's query contains business terms (e.g. '활성고객', '월매출', '신규유저') that are NOT defined in the semantic layer and whose meaning is ambiguous, ask ONE concise clarifying question BEFORE running any SQL. Do not guess.
- After answering with SQL results, if you had to infer a term's meaning yourself, suggest the user save it: "이 정의를 저장하려면 `/term_custom`으로 등록해두세요. 다음 질문부터 자동 적용됩니다."
"""


async def build_system_prompt(ctx: HarnessContext) -> str:
    parts: list[str] = [_BASE]

    if ctx.explorer is not None:
        tables = await ctx.explorer.list_tables()
        if tables:
            scope = ctx.identity.kv_scope if ctx.store else None
            has_enrichment = bool(
                scope and ctx.store and ctx.store.kv_get(scope, "schema_relationships")
            )

            if has_enrichment and scope and ctx.store:
                schema_lines: list[str] = []
                for tbl in tables:
                    try:
                        described = await ctx.explorer.describe_table(tbl.name)
                    except Exception:
                        schema_lines.append(f"- {tbl.qualified}")
                        continue
                    col_lines = []
                    for col in described.columns:
                        desc = (
                            col.description
                            or ctx.store.kv_get(
                                scope, f"enriched_desc:{tbl.name}:{col.name}"
                            )
                            or ""
                        )
                        col_lines.append(f"  - {col.name}{': ' + desc if desc else ''}")
                    schema_lines.append(f"- {tbl.qualified}\n" + "\n".join(col_lines))
                parts.append(
                    "## Known tables (with column descriptions)\n"
                    + "\n".join(schema_lines)
                )
            else:
                names = ", ".join(t.qualified for t in tables)
                parts.append("## Known tables\n" + names)

    if ctx.store is not None:
        scope = ctx.identity.kv_scope
        raw = ctx.store.kv_get(scope, "schema_relationships")
        if raw:
            try:
                rels = json.loads(raw)
                if rels:
                    rel_text = "\n".join(f"- {r}" for r in rels)
                    parts.append(
                        "## Table relationships (use these for JOINs)\n" + rel_text
                    )
            except (ValueError, TypeError):
                pass

        from ..tools.semantic_federation import build_prompt_section

        user_id = ctx.identity.user_id or "unknown"
        channel_id = ctx.identity.effective_channel_id
        semfed_section = build_prompt_section(ctx.store, scope, channel_id, user_id)
        if semfed_section:
            parts.append(semfed_section)

    return "\n\n".join(parts)