FabledScribe/src/scribe/services/snippets.py

"""Snippet service — reusable functions/components recorded for recall.

A *snippet* is a Note with ``note_type='snippet'``: a named, reusable function or
component recorded once so a later session can recall it before writing a
one-off. It carries a name, language, signature, canonical location
(repo · path · symbol), a one-line "when to reach for it", and the code itself.

Structured fields are stored as a **body-convention** — no dedicated column, so
a snippet inherits everything a note has (embeddings, ACL, project/System
association, dedup) and, crucially, becomes eligible for semantic recall the
moment it's embedded:

  - ``title`` = ``"{name} — {when_to_use}"``. The title is exactly what the
    title-first auto-inject surfaces, so this one line self-describes the snippet
    in a recall menu.
  - ``tags``  = ``[language, "snippet", *caller_tags]``.
  - ``body``  = templated markdown (When to use / Signature / Location, then a
    fenced code block).

The public field API (name/language/signature/location/when_to_use/code) lives
here, so the underlying storage could later move to a structured column without
changing any caller (MCP tool, REST route, UI).
"""
from __future__ import annotations

import asyncio
import re

from scribe.services import knowledge as knowledge_svc
from scribe.services import notes as notes_svc

SNIPPET_NOTE_TYPE = "snippet"
SNIPPET_TAG = "snippet"


def _embed_snippet(note) -> None:
    """Fire-and-forget embedding refresh for a snippet.

    A snippet's whole value is *immediate* recall — it must join the semantic /
    auto-inject pool the moment it's recorded, not wait for the startup backfill.
    Unlike a plain note (embedded at the REST-route boundary only, so its MCP
    create path defers to restart-backfill), a snippet is recorded primarily via
    MCP, so we embed here in the service — covering BOTH the MCP tool and the
    REST route by construction. Mirrors the route pattern: fire-and-forget,
    text = title + body. Import lazily so the pure serialize/parse helpers can be
    imported without pulling in the embedding model.
    """
    text = f"{note.title}\n{note.body}".strip() if note.body else (note.title or "")
    if not text:
        return
    from scribe.services.embeddings import upsert_note_embedding
    asyncio.create_task(upsert_note_embedding(note.id, note.user_id, text))


# --- serialize: structured fields -> note (title/body/tags) ------------------

def compose_title(name: str, when_to_use: str = "") -> str:
    """`name — when to use` (or just `name` when no usage note is given)."""
    name = (name or "").strip()
    when = (when_to_use or "").strip()
    return f"{name} — {when}" if when else name


def compose_tags(language: str = "", tags: list[str] | None = None) -> list[str]:
    """Language (lowercased) first, then the `snippet` marker, then caller tags —
    de-duplicated, order preserved."""
    out: list[str] = []
    lang = (language or "").strip().lower()
    if lang:
        out.append(lang)
    out.append(SNIPPET_TAG)
    for t in tags or []:
        t = (t or "").strip()
        if t and t not in out:
            out.append(t)
    return out


def _normalize_locations(locations: list[dict] | None) -> list[dict]:
    """Clean a list of {repo,path,symbol} locations: strip fields, drop wholly
    empty entries, de-duplicate identical ones (order preserved). A merged
    snippet carries several locations (one per call site); a fresh one carries
    at most one."""
    out: list[dict] = []
    seen: set[tuple[str, str, str]] = set()
    for loc in locations or []:
        repo = (loc.get("repo") or "").strip()
        path = (loc.get("path") or "").strip()
        symbol = (loc.get("symbol") or "").strip()
        if not (repo or path or symbol):
            continue
        key = (repo, path, symbol)
        if key in seen:
            continue
        seen.add(key)
        out.append({"repo": repo, "path": path, "symbol": symbol})
    return out


def _location_str(loc: dict) -> str:
    """`repo` · `path` · `symbol` — only the non-empty parts."""
    parts = [(loc.get(k) or "").strip() for k in ("repo", "path", "symbol")]
    return " · ".join(f"`{p}`" for p in parts if p)


def _render_location_block(locations: list[dict]) -> str | None:
    """One `**Location:**` line for a single location; a `**Locations:**` bullet
    list for several. None when there are none."""
    locs = [loc for loc in locations if _location_str(loc)]
    if not locs:
        return None
    if len(locs) == 1:
        return f"**Location:** {_location_str(locs[0])}"
    lines = "\n".join(f"- {_location_str(loc)}" for loc in locs)
    return f"**Locations:**\n{lines}"


def compose_body(
    *,
    code: str,
    language: str = "",
    signature: str = "",
    when_to_use: str = "",
    repo: str = "",
    path: str = "",
    symbol: str = "",
    locations: list[dict] | None = None,
) -> str:
    """Render structured fields into the snippet body markdown. Empty fields are
    omitted so the body stays clean.

    Locations: pass ``locations`` (a list of {repo,path,symbol}) for the general
    multi-location case; the single ``repo``/``path``/``symbol`` params remain as
    a back-compat shorthand for one location and are used only when ``locations``
    is not given.
    """
    if locations is None:
        locations = [{"repo": repo, "path": path, "symbol": symbol}]
    locs = _normalize_locations(locations)

    header: list[str] = []
    if (when_to_use or "").strip():
        header.append(f"**When to use:** {when_to_use.strip()}")
    if (signature or "").strip():
        header.append(f"**Signature:** `{signature.strip()}`")
    loc_block = _render_location_block(locs)
    if loc_block:
        header.append(loc_block)
    fence_lang = (language or "").strip().lower()
    code_block = f"```{fence_lang}\n{(code or '').rstrip()}\n```"
    if header:
        return "\n\n".join(header) + "\n\n" + code_block + "\n"
    return code_block + "\n"


# --- parse: note -> structured fields (best-effort, never raises) ------------

_WHEN_RE = re.compile(r"^\*\*When to use:\*\*\s*(.+?)\s*$", re.MULTILINE)
_SIG_RE = re.compile(r"^\*\*Signature:\*\*\s*(.+?)\s*$", re.MULTILINE)
_LOC_RE = re.compile(r"^\*\*Location:\*\*\s*(.+?)\s*$", re.MULTILINE)
_LOCS_RE = re.compile(
    r"^\*\*Locations:\*\*[ \t]*\n((?:[ \t]*-[ \t]*.+\n?)+)", re.MULTILINE
)
_CODE_RE = re.compile(r"```([\w+.#-]*)\n(.*?)\n```", re.DOTALL)


def _parse_location_str(s: str) -> dict | None:
    """Parse a `repo` · `path` · `symbol` fragment back into a location dict, or
    None if it's empty."""
    raw = [p.strip().strip("`").strip() for p in (s or "").split("·")]
    repo = raw[0] if len(raw) > 0 else ""
    path = raw[1] if len(raw) > 1 else ""
    symbol = raw[2] if len(raw) > 2 else ""
    if not (repo or path or symbol):
        return None
    return {"repo": repo, "path": path, "symbol": symbol}


def parse_snippet_fields(
    title: str, body: str, tags: list[str] | None = None
) -> dict:
    """Recover structured fields from a snippet note. Tolerant by design: a field
    that isn't present comes back empty and this never raises, so a hand-edited
    body can't break the edit form.

    ``locations`` is a list of {repo,path,symbol}; ``repo``/``path``/``symbol``
    mirror the FIRST location for back-compat with the single-location callers."""
    title = title or ""
    body = body or ""
    name, _, when_from_title = title.partition(" — ")
    fields = {
        "name": name.strip(),
        "when_to_use": when_from_title.strip(),
        "signature": "",
        "language": "",
        "repo": "",
        "path": "",
        "symbol": "",
        "locations": [],
        "code": "",
    }

    m = _WHEN_RE.search(body)
    if m:
        fields["when_to_use"] = m.group(1).strip()
    m = _SIG_RE.search(body)
    if m:
        fields["signature"] = m.group(1).strip().strip("`").strip()

    # Locations: prefer the multi-location `**Locations:**` bullet list, else the
    # legacy single `**Location:**` line.
    locations: list[dict] = []
    m = _LOCS_RE.search(body)
    if m:
        for line in m.group(1).splitlines():
            line = line.strip()
            if line.startswith("-"):
                loc = _parse_location_str(line[1:])
                if loc:
                    locations.append(loc)
    else:
        m = _LOC_RE.search(body)
        if m:
            loc = _parse_location_str(m.group(1))
            if loc:
                locations.append(loc)
    fields["locations"] = locations
    if locations:
        fields["repo"] = locations[0]["repo"]
        fields["path"] = locations[0]["path"]
        fields["symbol"] = locations[0]["symbol"]

    m = _CODE_RE.search(body)
    if m:
        fields["language"] = m.group(1).strip()
        fields["code"] = m.group(2)

    if not fields["language"]:
        for t in tags or []:
            if t and t != SNIPPET_TAG:
                fields["language"] = t
                break
    return fields


def snippet_to_dict(note) -> dict:
    """Note serialization plus a parsed ``snippet`` sub-object of structured
    fields, so callers get both the raw record and the typed view."""
    data = note.to_dict()
    data["snippet"] = parse_snippet_fields(note.title, note.body, note.tags)
    return data


# --- service wrappers over notes_svc ----------------------------------------

async def create_snippet(
    user_id: int,
    *,
    name: str,
    code: str,
    language: str = "",
    signature: str = "",
    when_to_use: str = "",
    repo: str = "",
    path: str = "",
    symbol: str = "",
    locations: list[dict] | None = None,
    tags: list[str] | None = None,
    project_id: int | None = None,
):
    """Create a snippet note (embedded on create for immediate recall). Returns
    the created Note. Pass ``locations`` for the multi-location case; the single
    ``repo``/``path``/``symbol`` are the one-location shorthand."""
    note = await notes_svc.create_note(
        user_id,
        title=compose_title(name, when_to_use),
        body=compose_body(
            code=code, language=language, signature=signature,
            when_to_use=when_to_use, repo=repo, path=path, symbol=symbol,
            locations=locations,
        ),
        note_type=SNIPPET_NOTE_TYPE,
        tags=compose_tags(language, tags),
        project_id=project_id,
    )
    _embed_snippet(note)
    return note


async def get_snippet(user_id: int, snippet_id: int):
    """Fetch a snippet by id, or None if it doesn't exist / isn't a snippet."""
    note = await notes_svc.get_note(user_id, snippet_id)
    if note is None or note.note_type != SNIPPET_NOTE_TYPE:
        return None
    return note


async def list_snippets(
    user_id: int,
    *,
    q: str | None = None,
    tag: str = "",
    limit: int = 50,
    offset: int = 0,
) -> tuple[list[dict], int]:
    """List snippets (id/title/tags/preview dicts), most-recently-updated first."""
    return await knowledge_svc.query_knowledge(
        user_id=user_id,
        note_type=SNIPPET_NOTE_TYPE,
        tags=[tag] if tag else [],
        sort="modified",
        q=q,
        limit=max(1, min(limit, 100)),
        offset=max(0, offset),
    )


async def update_snippet(
    user_id: int,
    snippet_id: int,
    *,
    name: str | None = None,
    code: str | None = None,
    language: str | None = None,
    signature: str | None = None,
    when_to_use: str | None = None,
    repo: str | None = None,
    path: str | None = None,
    symbol: str | None = None,
    locations: list[dict] | None = None,
    tags: list[str] | None = None,
    project_id: int | None = None,
):
    """Partial update: only fields passed (not None) change. Re-serializes the
    merged field set back into title/body/tags. Returns the Note, or None if the
    id isn't a snippet.

    Locations: ``locations`` replaces the whole set; else a legacy single
    ``repo``/``path``/``symbol`` overlays onto the first existing location; else
    the existing locations are kept."""
    note = await notes_svc.get_note(user_id, snippet_id)
    if note is None or note.note_type != SNIPPET_NOTE_TYPE:
        return None

    cur = parse_snippet_fields(note.title, note.body, note.tags)
    overlay = {
        "name": name, "code": code, "language": language,
        "signature": signature, "when_to_use": when_to_use,
    }
    merged = {**cur, **{k: v for k, v in overlay.items() if v is not None}}

    if locations is not None:
        merged_locations = _normalize_locations(locations)
    elif repo is not None or path is not None or symbol is not None:
        base = cur["locations"][0] if cur["locations"] else {"repo": "", "path": "", "symbol": ""}
        merged_locations = _normalize_locations([{
            "repo": repo if repo is not None else base["repo"],
            "path": path if path is not None else base["path"],
            "symbol": symbol if symbol is not None else base["symbol"],
        }])
    else:
        merged_locations = cur["locations"]

    fields: dict = {
        "title": compose_title(merged["name"], merged["when_to_use"]),
        "body": compose_body(
            code=merged["code"], language=merged["language"],
            signature=merged["signature"], when_to_use=merged["when_to_use"],
            locations=merged_locations,
        ),
    }
    # Recompute tags: keep any non-language, non-marker tags the note already had
    # (or the caller's replacement set), then re-derive language + marker.
    existing_extra = [
        t for t in (note.tags or []) if t not in (SNIPPET_TAG, cur.get("language", ""))
    ]
    fields["tags"] = compose_tags(
        merged["language"], tags if tags is not None else existing_extra
    )
    if project_id is not None:
        fields["project_id"] = project_id

    updated = await notes_svc.update_note(user_id, snippet_id, **fields)
    if updated is not None:
        # Title/body changed → refresh the embedding so recall reflects the edit.
        _embed_snippet(updated)
    return updated


# --- merge: unify found one-offs into one canonical snippet ------------------

def _extra_tags(tags: list[str] | None, language: str = "") -> list[str]:
    """A snippet's caller tags — everything except the language + `snippet`
    markers that compose_tags re-derives."""
    lang = (language or "").strip().lower()
    return [t for t in (tags or []) if t and t != SNIPPET_TAG and t != lang]


def merge_snippet_fields(
    target_fields: dict, target_tags: list[str] | None, sources: list[tuple[dict, list]]
) -> tuple[list[dict], list[str]]:
    """Pure merge: union locations (target's first, then each source in order)
    and union extra tags. The target's scalar fields (name/when_to_use/signature/
    language/code) win — only locations and tags accumulate. ``sources`` is a
    list of (parsed_fields, tags). Returns (locations, extra_tags)."""
    locations = list(target_fields.get("locations") or [])
    extra = _extra_tags(target_tags, target_fields.get("language", ""))
    for sfields, stags in sources:
        locations.extend(sfields.get("locations") or [])
        for t in _extra_tags(stags, sfields.get("language", "")):
            if t not in extra:
                extra.append(t)
    return _normalize_locations(locations), extra


async def merge_snippets(user_id: int, target_id: int, source_ids: list[int]):
    """Unify source snippets INTO the target (all owned by user_id): union their
    locations + tags onto the canonical target, keep the target's scalar fields,
    trash the sources (recoverable), re-embed the survivor.

    Returns (merged_target_note, merged_source_ids), or None if the target isn't
    the user's snippet. Source ids that aren't the user's snippets are skipped."""
    target = await notes_svc.get_note(user_id, target_id)
    if target is None or target.note_type != SNIPPET_NOTE_TYPE:
        return None

    sources = []
    for sid in source_ids:
        if sid == target_id:
            continue
        s = await notes_svc.get_note(user_id, sid)
        if s is not None and s.note_type == SNIPPET_NOTE_TYPE:
            sources.append(s)

    tgt_fields = parse_snippet_fields(target.title, target.body, target.tags)
    parsed_sources = [
        (parse_snippet_fields(s.title, s.body, s.tags), s.tags) for s in sources
    ]
    locations, extra_tags = merge_snippet_fields(tgt_fields, target.tags, parsed_sources)

    updated = await notes_svc.update_note(
        user_id, target_id,
        body=compose_body(
            code=tgt_fields["code"], language=tgt_fields["language"],
            signature=tgt_fields["signature"], when_to_use=tgt_fields["when_to_use"],
            locations=locations,
        ),
        tags=compose_tags(tgt_fields["language"], extra_tags),
    )
    if updated is None:
        return None

    # Retire the merged-in sources to the trash (recoverable).
    from scribe.services.trash import delete as trash_delete
    merged_ids: list[int] = []
    for s in sources:
        batch = await trash_delete(user_id, "note", s.id)
        if batch is not None:
            merged_ids.append(s.id)

    _embed_snippet(updated)
    return updated, merged_ids