Files
brain-of-reese/app/models.py
T
ducoterra 8c706259e9
Build and Push Containers / build-and-push-app (push) Successful in 1m44s
Build and Push Containers / build-and-push-db (push) Successful in 13s
phase: 89_source_ignore_paths
All verification complete — TODO.md was already cleared in the roadmap commit; the two extra unit-test diffs are necessary fake-signature adaptations for the new keywords. Everything is green, no fixes were needed.

## Phase 89 — final verification pass: ALL GREEN

**Verified (all 6 task files present in `complete/`):**
- `git_sources.ignore_paths` JSONB column + migration 0013; `alembic downgrade -1 && upgrade head` round-trips (head `0013`)
- Importer: `normalize_ignore_path`/`is_ignored`/`_ignore_for_root`, `ignore` in walk + progress pre-walk, `ignore_by_root` in `import_sources`
- API: GET/POST carry list; admin-only `PATCH` (replace, 404/422 fixed details, anonymous 403)
- Pipelines wired: `_run_sync`, `_run_upload` re-upload, `scripts/import_docs.py`
- Sources-page box: dialog, §7.4 save lifecycle, `N ignored` tag, a11y; env rows get no box

**Test/lint results:**
- `uv run pytest` → 1808 passed
- `uv run pytest --cov=app --cov-report=term-missing` → TOTAL **99%** (>90%)
- `uv run pytest tests/e2e/test_source_ignore_paths.py -v --no-cov` → 6 passed (isolated, DB up)
- Regressions in isolation: `test_git_sources_admin` 6, `test_archive_upload_sources` 5, `test_sync_button` 3, `test_smoke` 3 — all passed
- `uv run ruff check . && uv run pyright` → clean (0 errors)

**Completion criteria:** box→PATCH 200→count+GET round-trip ✅ · sync excludes `ignore/` (no docs/chunks/embeddings/summaries) + prunes newly-ignored (pruned==2) ✅ · no-mid-path rule E2E ✅ · PATCH 404/422/replace/clear/403 ✅ · full gate green ✅ · commit + phase move left to harness per rules.

**Deviations:** none blocking — E2E pins `files == 4` (overview's "5" was an off-by-one vs its own 6-file tree, documented in-test); `tests/unit/test_importer.py` + `test_sync_button.py` test-double fakes extended for the new keywords (needed for the suite to stay green).

**Next pending phase:** none — `todo/` holds only this phase.
2026-09-09 01:45:42 -04:00

387 lines
19 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
"""SQLAlchemy models (PostgreSQL 17 + pgvector).
Data model — see ``.agents/PLAN.md`` §Data Model:
* ``documents`` — one row per imported A9 file (full content, path, sha256 hash).
* ``chunks`` — retrieval units; each chunk points at its parent document
via ``document_id``. This is how an embedding maps back to
a document path (the "feed the whole document" requirement).
* ``query_log`` — observability: every question, its retrieval score,
the deflection decision, and latency.
* ``steering_notes`` — owner tuning notes injected into the system prompt
of every chat turn (phase 15, ``<tuning>`` section).
* ``kb_overview`` — single-row lite-generated outline of the KB's basic
categories, injected as the ``<knowledge_base>``
section of every chat turn (phase 31).
* ``git_sources`` — admin-managed source registry (git URLs + local
directories) the Sync button and import_docs
import (phase 35; ``kind`` discriminator added in
phase 38; ``ignore_paths`` (phase 89 — JSONB list
of normalized path prefixes, server default
``'[]'``)).
* ``saved_chats`` — owner-saved chat conversations: one row per
explicitly Saved conversation (auto-``title`` +
the ``bor.chat.v1`` message list as JSONB, phase
14 shape) — phase 50; ``/api/chat`` stays
stateless; ``share_token`` (NULL = private,
``uuid4`` = publicly readable at
``/shared/<token>``) — phase 51; ``sources_version``
(the KB generation the conversation was saved
against; 0 = the pre-counter KB, stale on the
first bump) — phase 53.
* ``sources_meta`` — single-row sources-version counter: which
generation of the knowledge base is current,
bumped exactly once per KB-changing sync so saved
chats can be marked stale (phase 53).
* ``doc_drafts`` — server-side drafts of chat answers saved as
documentation: one row per "Save as doc" action
(the long answer body lives here, never in a URL),
keyed by an unguessable ``uuid4`` ``token`` (the
edit screen's URL credential — the share-token
trust model, phase 51); ``status`` moves
``draft`` → ``pushed`` (``branch`` +
``commit_sha`` recorded) when the push endpoint
commits + pushes the file to the
``BOR_DOCS_REPO`` branch (phase 59).
* ``api_tokens`` — admin-issued access tokens: one row per
generated token, so a person handed a token can
sign in to use the app (chat, suggestion chips,
cited documents) — the ONLY content that stays
anonymous is the shared chats (phase 79).
``token_hash`` is the SHA-256 hex digest of the
full ``bor_…`` token string (the stored
credential — the plaintext exists only in the 201
create response, returned exactly once);
``revoked_at`` set = dead (live-checked on the
holder's next request), ``last_used_at`` bumped
on ``POST /api/token-auth`` (task 03 — the only
request that presents the token; the in-app gate
re-sends the cached token on every page load).
"""
from __future__ import annotations
import uuid
from datetime import datetime
from pgvector.sqlalchemy import Vector
from sqlalchemy import (
Boolean,
DateTime,
Float,
ForeignKey,
Integer,
String,
Text,
UniqueConstraint,
func,
text,
)
from sqlalchemy.dialects.postgresql import JSONB, UUID
from sqlalchemy.orm import Mapped, mapped_column, relationship
from app.config import get_settings
from app.db import Base
# Single source of truth for the vector column size (see .agents/PLAN.md A6).
EMBEDDING_DIM: int = get_settings().embedding_dim
class Document(Base):
__tablename__ = "documents"
__table_args__ = (UniqueConstraint("source", "path", name="uq_documents_source_path"),)
id: Mapped[uuid.UUID] = mapped_column(UUID(as_uuid=True), primary_key=True, default=uuid.uuid4)
source: Mapped[str] = mapped_column(String(120), index=True) # e.g. "Homelab"
path: Mapped[str] = mapped_column(String(1000), index=True) # relative to source dir
full_path: Mapped[str] = mapped_column(String(2000)) # absolute path at import time
title: Mapped[str] = mapped_column(String(500))
content: Mapped[str] = mapped_column(Text) # full markdown — the RAG context
content_hash: Mapped[str] = mapped_column(String(64), index=True) # sha256 for change detection
indexed_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now())
#: Lite-model summary, phase 30. Natural-language summary of the
#: document (non-markdown A9 docs only, generated at import time by the
#: aipi ``lite`` model). NULL for markdown docs, pre-phase-30 rows, and
#: the fail-soft path where summary generation failed but the document
#: was still indexed.
summary: Mapped[str | None] = mapped_column(Text, default=None)
chunks: Mapped[list[Chunk]] = relationship(
back_populates="document", cascade="all, delete-orphan"
)
class Chunk(Base):
__tablename__ = "chunks"
id: Mapped[uuid.UUID] = mapped_column(UUID(as_uuid=True), primary_key=True, default=uuid.uuid4)
document_id: Mapped[uuid.UUID] = mapped_column(
UUID(as_uuid=True), ForeignKey("documents.id", ondelete="CASCADE"), index=True
)
position: Mapped[int] = mapped_column(Integer)
content: Mapped[str] = mapped_column(Text)
embedding: Mapped[list[float] | None] = mapped_column(Vector(EMBEDDING_DIM))
#: Summary chunk, position −1, phase 30. Marks the single extra embedded
#: chunk mirroring ``Document.summary``; default False keeps every
#: pre-phase-30 row (and ordinary content chunks) valid.
is_summary: Mapped[bool] = mapped_column(Boolean, default=False)
document: Mapped[Document] = relationship(back_populates="chunks")
class QueryLog(Base):
__tablename__ = "query_log"
id: Mapped[uuid.UUID] = mapped_column(UUID(as_uuid=True), primary_key=True, default=uuid.uuid4)
question: Mapped[str] = mapped_column(Text)
top_score: Mapped[float] = mapped_column(Float, default=0.0) # best cosine similarity
#: Lexical (FTS) candidates matched — the OR-tsquery hit count (A8). NULL
#: for pre-hybrid rows (migration 0002).
fts_hits: Mapped[int | None] = mapped_column(Integer)
chunk_hits: Mapped[int] = mapped_column(Integer, default=0)
deflected: Mapped[bool] = mapped_column(Boolean, default=False) # True = honest "no idea"
sources: Mapped[str] = mapped_column(Text, default="") # comma-joined source paths
latency_ms: Mapped[int] = mapped_column(Integer, default=0)
created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now())
class SteeringNote(Base):
"""One owner tuning instruction (phase 15).
Notes are read into the system prompt of **every** chat turn as the
``<tuning>`` section (oldest first, char-budgeted — see
:func:`app.rag.prompts.build_steering_section`).
"""
__tablename__ = "steering_notes"
id: Mapped[uuid.UUID] = mapped_column(UUID(as_uuid=True), primary_key=True, default=uuid.uuid4)
note: Mapped[str] = mapped_column(Text) # trimmed, 1–2000 chars (API-enforced)
created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now())
class KbOverview(Base):
"""Single-row, lite-generated outline of the knowledge base (phase 31).
Exactly one row (``id = 1``, enforced by the migration 0005 server
defaults) holds a plain-text outline of the KB's basic categories,
generated by the aipi ``lite`` model whenever an import changes the KB.
Chat turns only read this row (one indexed PK lookup) and inject it into
the system prompt of every turn as the ``<knowledge_base>`` section — an
empty row means the section is absent and the prompt stays
byte-identical to the pre-phase text (phase 15 convention).
"""
__tablename__ = "kb_overview"
id: Mapped[int] = mapped_column(Integer, primary_key=True, server_default="1")
content: Mapped[str] = mapped_column(Text, server_default="") # the outline text
updated_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), server_default=func.now()
)
class SourcesMeta(Base):
"""Single-row sources-version counter (phase 53).
Exactly one row (``id = 1``, seeded by migration 0010 — the
``kb_overview`` id=1 precedent) holds the current **generation** of
the knowledge base. ``version`` is bumped exactly once per sync that
actually changed the KB (phase 53, task 02 — change-gated on
``added + updated + pruned > 0``), so it doubles as the invalidation
marker for saved chats: a ``saved_chats`` row stamped with an older
generation is *stale* — its answers predate the current index and
may be Regenerated against it (phase 53, tasks 03/05). The row is
seeded by the migration (not lazily on first bump), so
:func:`app.rag.sources_meta.current_sources_version` is a plain PK
read.
"""
__tablename__ = "sources_meta"
id: Mapped[int] = mapped_column(Integer, primary_key=True, server_default="1")
#: The KB generation. 0 = the pre-counter KB (everything indexed
#: before phase 53); incremented by one per KB-changing sync.
version: Mapped[int] = mapped_column(Integer, server_default="0")
updated_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), server_default=func.now(), onupdate=func.now()
)
class GitSource(Base):
"""One admin-managed source (phase 35; kind discriminator, phase 38).
The UI-maintained list the Sync button (phase 32) and import_docs
(phase 28) import from. ``kind`` discriminates: ``git`` rows carry a
repo ``url`` (cloned/pulled), ``local`` rows carry an existing
directory ``path`` (walked directly). DB rows win over the
BOR_GIT_SOURCES env var (git-only fallback), which is a fallback
while this table is empty (see
app.rag.git_sources.effective_git_sources).
"""
__tablename__ = "git_sources"
id: Mapped[uuid.UUID] = mapped_column(UUID(as_uuid=True), primary_key=True, default=uuid.uuid4)
url: Mapped[str] = mapped_column(Text, unique=True, nullable=False)
#: Source-kind discriminator (phase 38): "git" (default) or "local"
#: — enforced by the ``ck_git_sources_kind`` CHECK constraint.
kind: Mapped[str] = mapped_column(Text, default="git", server_default="'git'")
#: Absolute directory of a ``local`` source; NULL for git rows.
#: Unique — Postgres treats NULLs as distinct under a unique index.
path: Mapped[str | None] = mapped_column(Text, unique=True)
#: Per-source ignore paths (phase 89, A1/A4): the normalized,
#: non-empty source-relative path prefixes the owner types into the
#: box on the Sources page. A file is ignored when its
#: source-relative POSIX path starts with any entry (raw prefix —
#: no mid-path matching, no globs). Server default '[]' — every
#: pre-phase-89 row imports exactly as before.
ignore_paths: Mapped[list] = mapped_column(
JSONB, default=list, server_default=text("'[]'"), nullable=False
)
added_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now())
class DocDraft(Base):
"""One server-side draft of a chat answer saved as documentation
(phase 59, task 01).
A long answer body must live on the **server**, never in a URL: the
"Save as doc" action POSTs the answer's raw markdown to
``POST /api/doc-drafts`` (task 02), which stores it here and hands
back an unguessable 128-bit ``uuid4`` ``token`` — the edit
screen's URL credential (``/doc-edit.html?draft=<token>``, the
share-token trust model, phase 51). ``status`` stays ``draft``
until the push endpoint (task 04) commits + pushes the file to the
``BOR_DOCS_REPO`` branch — then it is ``pushed``, with ``branch``
and ``commit_sha`` recorded (the UI's branch + sha feedback; D3:
no PR tooling — the owner opens the PR themselves).
"""
__tablename__ = "doc_drafts"
id: Mapped[uuid.UUID] = mapped_column(UUID(as_uuid=True), primary_key=True, default=uuid.uuid4)
#: The URL credential (``/doc-edit.html?draft=<token>``): an
#: unguessable 128-bit ``uuid4`` — never the row id, never
#: sequential/guessable. Unique NOT NULL: unlike the NULLable
#: ``saved_chats.share_token`` there is no "un-drafted" state, so
#: NULLs never occur (always set on create).
token: Mapped[uuid.UUID] = mapped_column(
UUID(as_uuid=True), unique=True, nullable=False, default=uuid.uuid4
)
#: The document's title. Defaults client-side to the last user
#: question (whitespace-collapsed, ≤120 chars — the chat auto-title
#: convention, phase 50); the edit screen changes anything.
title: Mapped[str] = mapped_column(Text)
#: The in-repo file path (default ``docs/<slug>.md``). Guard-railled
#: by the API layer (task 02 — repo-relative, no ``..``); the
#: column itself is plain TEXT (the ``documents.path`` precedent).
path: Mapped[str] = mapped_column(Text)
#: The markdown body — the answer's raw text (never HTML — the
#: ``bor.chat.v1`` record's ``text``), edited on the edit screen.
body: Mapped[str] = mapped_column(Text)
#: "draft" until the push endpoint commits + pushes the file, then
#: "pushed" — the domain is enforced by the API layer (the
#: ``git_sources.kind`` phase-38 precedent: plain TEXT + server
#: default, no CHECK constraint).
status: Mapped[str] = mapped_column(Text, default="draft", server_default="'draft'")
#: Set on push (task 04): the branch the commit landed on (the
#: ``BOR_DOCS_BRANCH`` name); NULL while still a draft.
branch: Mapped[str | None] = mapped_column(Text)
#: ... and the pushed branch's new HEAD sha (must equal
#: ``git rev-parse <branch>`` in the repo); NULL while still a
#: draft.
commit_sha: Mapped[str | None] = mapped_column(Text)
created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now())
updated_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), server_default=func.now(), onupdate=func.now()
)
class SavedChat(Base):
"""One owner-saved chat conversation (phase 50).
Only conversations the owner explicitly **Saves** are stored —
``/api/chat`` itself stays stateless (A10, owner-locked extension
2026-08-29) and nothing is stored about a conversation that was not
saved. ``messages`` holds the exact ``bor.chat.v1`` localStorage
record shape (phase 14 — raw text, never HTML), so a saved chat
restores pixel-identical through the existing
``renderStoredMessage`` path.
"""
__tablename__ = "saved_chats"
id: Mapped[uuid.UUID] = mapped_column(UUID(as_uuid=True), primary_key=True, default=uuid.uuid4)
#: Auto-title (the first question, API-side). Plain column on purpose:
#: a future rename needs no migration.
title: Mapped[str] = mapped_column(String(500))
#: ``list[dict]`` in the ``bor.chat.v1`` record shape
#: (``{who, text, sources?, deflected?, suggestions?, thinking?,
#: tools?, stopped?}``); the API always supplies a list, so no
#: default is needed.
messages: Mapped[list] = mapped_column(JSONB)
#: Anonymous share link (phase 51): a 128-bit ``uuid4`` token; when
#: set, the chat is publicly readable at ``/shared/<token>`` without
#: any admin session, and unsharing (token → NULL) revokes it.
#: Unique — Postgres treats NULLs as distinct under a unique index
#: (the ``git_sources.path`` precedent, phase 38), so any number of
#: unshared chats coexist while two identical tokens can never exist.
share_token: Mapped[uuid.UUID | None] = mapped_column(
UUID(as_uuid=True), unique=True, nullable=True
)
#: The sources version (KB generation) the conversation was saved
#: against (phase 53): stamped by the API at save time. Existing
#: rows (saved before the counter existed) stamp ``0`` — "the
#: pre-counter KB" — and become stale on the first KB-changing sync
#: (stale = ``sources_version < current``, computed server-side by
#: the chats API, phase 53 task 03).
sources_version: Mapped[int] = mapped_column(Integer, server_default="0")
created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now())
updated_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), server_default=func.now(), onupdate=func.now()
)
class ApiToken(Base):
"""One admin-issued access token (phase 79, task 01).
The admin generates named tokens and hands them out so people can
sign in to the app and use it (chat, suggestion chips, cited
documents) — the ONLY content that stays anonymous is the shared chats
(extending the phase-16 single-admin auth; the ``require_user``
live-check and the admin token API land in tasks 02/03).
Trust model — the plaintext token (``bor_`` + 32 hex chars) exists
only in the 201 response of the create call, returned **exactly
once**; the row never carries it. The stored credential is the
SHA-256 hex digest of the **full** token string (``token_hash``):
hashing the full string, not the suffix, so a stripped prefix can
never collide. The ``saved_chats.share_token`` /
``doc_drafts.token`` lineage — but HASHED: unlike those
unguessable ``uuid4`` link tokens these are long-lived hand-out
credentials, and a leaked database must not hand anyone working
tokens.
"""
__tablename__ = "api_tokens"
id: Mapped[uuid.UUID] = mapped_column(UUID(as_uuid=True), primary_key=True, default=uuid.uuid4)
#: The hand-out name (e.g. "alice") — display-only: no index, not
#: unique (two tokens may share a label).
label: Mapped[str] = mapped_column(String(120), nullable=False)
#: The stored credential: the SHA-256 hex digest of the full
#: ``bor_…`` token string (the ``documents.content_hash``
#: String(64) precedent). Unique — the lookup is a unique-index hit
#: (``ix_api_tokens_token_hash`` — the explicit unique-index shape
#: of ``ix_saved_chats_share_token``, phase 51).
token_hash: Mapped[str] = mapped_column(String(64), unique=True, nullable=False)
created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now())
#: Bumped to now() on ``POST /api/token-auth`` (task 03 calls the
#: service's ``mark_used`` and commits); NULL until the token is
#: first used.
last_used_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True)
#: Set on revocation (task 02) — the row is dead from that moment
#: (enforced immediately on the holder's next request); NULL while
#: active.
revoked_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True)