Files
sure/app/models/user.rb
Guillem Arias Fauste 8251b7e4d6 feat(ai): add Anthropic provider with chat parity (1/5) (#1983)
* feat(ai): add Anthropic provider with chat parity (1/5)

Introduces Provider::Anthropic alongside Provider::Openai, implementing
the LlmConcept chat_response contract over the official anthropic Ruby
SDK. Batch ops, PDF, and RAG land in follow-up PRs.

- Provider::Anthropic uses Messages API for sync and streaming responses
- ChatConfig builds requests with ephemeral prompt-cache markers on the
  system prompt and the last tool definition
- MessageFormatter reconstructs multi-turn history (text + tool_use +
  tool_result blocks) from raw Message records, including the paired
  user-role tool_result turn Anthropic requires after every tool_use
- ChatParser maps Anthropic Message into the shared ChatResponse Data
- Registry, Setting, User, Chat default model wired for ANTHROPIC_*
  envs and Setting.anthropic_*; LLM_PROVIDER selects between providers
- Responder forwards raw conversation_history (Array<Message>) so
  providers without hosted conversation state can rebuild context
- OpenAI provider accepts and ignores the new kwarg (no behavior change)

Tests cover provider init, model gating, MessageFormatter for all turn
shapes, ChatConfig request building (max_tokens, system cache, tool
conversion), ChatParser for text / tool_use / mixed blocks, Registry
discovery, and mocked chat_response success / error / function_request
paths. Live VCR cassettes recorded in a follow-up with a real key.

Stacked PRs: 2/5 batch ops + cost ledger, 3/5 PDF, 4/5 pgvector RAG,
5/5 settings UI + disclosure.

* fix(ai): address PR review on Anthropic provider foundation

Surface fixes raised by Codex + CodeRabbit on PR 1/5:

- Provider::Anthropic#chat_response now accepts (and ignores) a
  `messages:` kwarg. Assistant::Responder passes both `messages:`
  (OpenAI-shape) and `conversation_history:` (raw Message records) for
  cross-provider parity, so the previous signature raised
  ArgumentError on the first chat turn through the Anthropic provider.
- Provider::Anthropic#supports_model? bypasses the `claude` prefix
  gate when a custom base_url is configured, mirroring the OpenAI
  provider. Bedrock-shaped IDs like
  `anthropic.claude-sonnet-4-5-20250929-v1:0` and
  `claude-opus-4@20250514` are otherwise rejected by
  Assistant::Provided#get_model_provider and the chat dies.
- Setting.anthropic_access_token is now in
  EncryptedSettingFields::ENCRYPTED_FIELDS so the Anthropic API key
  is encrypted at rest like every other provider secret. Previously
  plaintext while siblings (openai_access_token, twelve_data_api_key,
  external_assistant_token) were ciphertext.
- Chat.default_model falls back to whichever provider is actually
  configured. Previously, with LLM_PROVIDER=anthropic but no
  Anthropic credentials, the default model resolved to a Claude ID
  that no registered provider supported, so chats failed even when
  OpenAI was fully configured. Adds Provider::{Anthropic,Openai}#configured?
  class methods for the readable callsite.
- Provider::Anthropic.effective_model uses
  `ENV["ANTHROPIC_MODEL"].presence || Setting.anthropic_model` so the
  Setting lookup is only performed when the env var is absent — the
  previous `ENV.fetch(KEY, default)` evaluated the default arg
  eagerly on every call.
- Provider::Anthropic::ChatConfig#anthropic_input_schema strips both
  `:strict` and `"strict"` keys so JSON-decoded schemas with string
  keys cannot leak the OpenAI-only flag through to Anthropic.

Test coverage added: supports_model? bypass on custom endpoints,
chat_response messages: kwarg compatibility, default_model fallback
in the three credential combinations, configured? against ENV +
Setting, strict-flag stripping for both key types, and a
`Setting.expects(:anthropic_model).never` assertion proving the
ENV-precedence test now exercises the lazy path.

All 4365 tests pass (1 pre-existing libvips env error unrelated).

* test(chat): make default_model tests resilient to ENV model overrides

CodeRabbit flagged on PR review: the new default_model tests asserted
against Provider::*::DEFAULT_MODEL, but Chat.default_model actually
returns Provider::*.effective_model.presence (which reads
OPENAI_MODEL / ANTHROPIC_MODEL from the environment). With either env
var set, the tests would fail intermittently even though routing was
correct.

- New default_model tests now assert against the provider's
  effective_model directly, so they verify the routing decision
  (which provider's value wins) without coupling to the constant.
- Pre-existing "creates with default model" assertions had the same
  brittleness; switch them to compare against Chat.default_model so
  the chosen model is whatever the env / Setting cascade resolves to.

Verified by running `ANTHROPIC_MODEL=claude-haiku-4-5 OPENAI_MODEL=gpt-4o
bin/rails test test/models/chat_test.rb` — 16 runs, 0 failures
(previously 2 pre-existing failures + 0 from the new tests).

* fix(ai): address local review on Anthropic foundation

- Provider::Anthropic#supports_pdf_processing? bypasses prefix gate for
  custom endpoints, mirroring supports_model?
- Provider::Anthropic#initialize raises Error when custom_endpoint? AND
  model.blank?, parity with Provider::Openai
- stream_chat_response captures partial usage on mid-stream errors and
  records it via the new on_partial callback so chat_response can skip
  the duplicate error row in the outer rescue
- safe_accumulated_message swallows the secondary failure when the SDK
  cannot reconstruct a snapshot
- langfuse_client memoizes properly (||= instead of =) so repeated calls
  don't churn Langfuse instances
- MessageFormatter sorts tool_calls by created_at then id so the
  message array is deterministic across replays; skips tool_calls
  missing both provider_call_id and provider_id rather than sending
  `id: nil` and getting rejected by Anthropic
- Setting.anthropic_access_token default falls back through
  ENV["ANTHROPIC_API_KEY"].presence (was missing .presence, so an
  empty-string env value bled through)
- User#openai_configured? / #anthropic_configured? delegate to the
  Provider::* class methods — single source of truth
- Assistant::Responder renames the OpenAI-shape history builder
  conversation_history → openai_messages_payload so the kwarg name
  matches the local method name (messages: openai_messages_payload,
  conversation_history: chat_message_records)
- Assistant::Builtin stale-history comment updated to reference both
  builders

Adds a streaming chat_response test using ad-hoc subclasses of the
SDK event types so the case/when dispatch matches via is_a? without
stubbing class-level === behavior.

* test(ai): add Anthropic tool_use round-trip + multi-tool turn coverage

Addresses @jjmata's "worth confirming" note on PR #1983: tool-use turns
from prior assistant messages must round-trip correctly when retrieved
from the database.

- New `ChatParser → ToolCall::Function → MessageFormatter` test walks
  the full path: Anthropic response with a tool_use block →
  ChatFunctionRequest → ToolCall::Function.from_function_request →
  persisted on the AssistantMessage → MessageFormatter rebuild on the
  next turn. Asserts the original `tool_use.id` is preserved end-to-end
  as both `tool_use.id` and the paired `tool_result.tool_use_id`, and
  that the original `input` hash and serialized result content survive.
- New multi-tool assistant turn test confirms two tool_use blocks on a
  single assistant message render as two tool_use blocks followed by
  two paired tool_result blocks in a single user-role follow-up,
  matching Anthropic's required alternation.

Both tests exercise the existing PR1 code without behavior changes.

* test(ai): require "ostruct" explicitly in Anthropic provider tests

OpenStruct is moving out of Ruby's default load path (warning in 3.4+,
removed in 3.5+). Tests work today because ActiveSupport transitively
loads it, but that's incidental. Match the existing convention in
test/controllers/settings/hostings_controller_test.rb which explicitly
requires ostruct for the same reason.

* fix(ai): sanitize Langfuse warn logs, normalize tool_use.input, dedup history fetch

Addresses three open CodeRabbit findings on PR #1983.

- Provider::Anthropic Langfuse rescue branches no longer include
  `e.full_message` in `Rails.logger.warn`. `full_message` bundles the
  backtrace + cause chain and on some SDK error types includes the
  serialized request/response payload (prompt, model output). Logs
  now report `#{e.class}: #{e.message}` only. Three sites:
  create_langfuse_trace, log_langfuse_generation, upsert_langfuse_trace.
  Note: Provider::Openai has the same pattern (copy-pasted source) —
  harmonization deferred to a follow-up cleanup PR; this commit fixes
  only the Anthropic provider to keep PR scope tight.

- MessageFormatter#parse_arguments now coerces any non-Hash parsed
  result to `{}`. Anthropic's Messages API requires `tool_use.input`
  to be a JSON object (map); a stored ToolCall::Function record whose
  arguments parse to a scalar, bool, or array (corrupt row, legacy
  data, cross-provider bleed) would otherwise produce a payload the
  API rejects. Normal flow stores Hash arguments end-to-end so the
  fix is defensive — adds 2 tests covering scalar/array JSON strings
  and non-String non-Hash inputs.

- Assistant::Responder dedups the chat-history fetch. The previous
  layout fired two near-identical `chat.messages.where(...).includes(
  :tool_calls).ordered` queries per LLM turn (one for the OpenAI-shape
  payload, one for the raw-records kwarg). A new memoized
  `complete_chat_messages` fetches once; `chat_message_records` filters
  out the current message via `Array#reject`, `openai_messages_payload`
  iterates the cached array unchanged. One SQL query per turn instead
  of two. Memoization scope = single Responder instance (per LLM call),
  so cache invalidation is not a concern.

All 4370 tests pass (1 pre-existing libvips env error unrelated).
Rubocop + brakeman clean.

* fix(ci): replace sk-ant- prefixed test placeholders

Pipelock secret scanner pattern-matches `sk-ant-*` as a real Anthropic
API key and fails the PR security-scan check. Test stubs and
ClimateControl env values used `sk-ant-test`, `sk-ant-from-setting`,
`sk-ant-x`, `sk-ant-y` as obvious placeholders, but the scanner does
not care about value entropy.

Switched to `fake-anthropic-key-*` / `fake-token-*` strings so the
scanner stops flagging them. No production code touched, no behavior
change — Provider::Anthropic still accepts any non-blank token.
2026-05-31 16:11:28 +02:00

560 lines
16 KiB
Ruby

class User < ApplicationRecord
include Encryptable
# Allow nil password for SSO-only users (JIT provisioning).
# Custom validation ensures password is present for non-SSO registration.
has_secure_password validations: false
# Encrypt sensitive fields if ActiveRecord encryption is configured
if encryption_ready?
# MFA secrets
encrypts :otp_secret, deterministic: true
# PII - emails (deterministic for lookups, downcase for case-insensitive)
encrypts :email, deterministic: true, downcase: true
encrypts :unconfirmed_email, deterministic: true, downcase: true
# PII - names (non-deterministic for maximum security)
encrypts :first_name
encrypts :last_name
end
belongs_to :family
belongs_to :last_viewed_chat, class_name: "Chat", optional: true
belongs_to :default_account, class_name: "Account", optional: true
has_many :sessions, dependent: :destroy
has_many :chats, dependent: :destroy
has_many :api_keys, dependent: :destroy
has_many :webauthn_credentials, dependent: :destroy
has_many :mobile_devices, dependent: :destroy
has_many :invitations, foreign_key: :inviter_id, dependent: :destroy
has_many :impersonator_support_sessions, class_name: "ImpersonationSession", foreign_key: :impersonator_id, dependent: :destroy
has_many :impersonated_support_sessions, class_name: "ImpersonationSession", foreign_key: :impersonated_id, dependent: :destroy
has_many :oidc_identities, dependent: :destroy
has_many :sso_audit_logs, dependent: :nullify
has_many :owned_accounts, class_name: "Account", foreign_key: :owner_id
has_many :account_shares, dependent: :destroy
has_many :shared_accounts, through: :account_shares, source: :account
accepts_nested_attributes_for :family, update_only: true
MFA_BACKUP_CODE_COUNT = 8
validates :email, presence: true, uniqueness: true, format: { with: URI::MailTo::EMAIL_REGEXP }
validate :ensure_valid_profile_image
validates :default_period, inclusion: { in: Period::PERIODS.keys }
validates :default_account_order, inclusion: { in: AccountOrder::ORDERS.keys }
validates :locale, inclusion: { in: I18n.available_locales.map(&:to_s) }, allow_nil: true
# Password is required on create unless the user is being created via SSO JIT.
# SSO JIT users have password_digest = nil and authenticate via OIDC only.
validates :password, presence: true, on: :create, unless: :skip_password_validation?
validates :password, length: { minimum: 8 }, allow_nil: true
normalizes :email, with: ->(email) { email.strip.downcase }
normalizes :unconfirmed_email, with: ->(email) { email&.strip&.downcase }
normalizes :locale, with: ->(locale) { locale.presence }
normalizes :first_name, :last_name, with: ->(value) { value.strip.presence }
enum :role, { guest: "guest", member: "member", admin: "admin", super_admin: "super_admin" }, validate: true
enum :ui_layout, { dashboard: "dashboard", intro: "intro" }, validate: true, prefix: true
before_validation :apply_ui_layout_defaults
before_validation :apply_role_based_ui_defaults
# Returns the appropriate role for a new user creating a family.
# The very first user of an instance becomes super_admin; subsequent users
# get the specified fallback role (typically :admin for family creators).
def self.role_for_new_family_creator(fallback_role: :admin)
User.exists? ? fallback_role : :super_admin
end
has_one_attached :profile_image, dependent: :purge_later do |attachable|
attachable.variant :thumbnail, resize_to_fill: [ 300, 300 ], convert: :webp, saver: { quality: 80 }
attachable.variant :small, resize_to_fill: [ 72, 72 ], convert: :webp, saver: { quality: 80 }, preprocessed: true
end
validate :profile_image_size
generates_token_for :password_reset, expires_in: 15.minutes do
password_salt&.last(10)
end
generates_token_for :email_confirmation, expires_in: 1.day do
unconfirmed_email
end
def pending_email_change?
unconfirmed_email.present?
end
def initiate_email_change(new_email)
return false if new_email == email
if Rails.application.config.app_mode.self_hosted? && !Setting.require_email_confirmation
update(email: new_email)
else
if update(unconfirmed_email: new_email)
EmailConfirmationMailer.with(user: self).confirmation_email.deliver_later
true
else
false
end
end
end
def resend_confirmation_email
if pending_email_change?
EmailConfirmationMailer.with(user: self).confirmation_email.deliver_later
true
else
false
end
end
def request_impersonation_for(user_id)
impersonated = User.find(user_id)
impersonator_support_sessions.create!(impersonated: impersonated)
end
def admin?
super_admin? || role == "admin"
end
def accessible_accounts
family.accounts.accessible_by(self)
end
def finance_accounts
family.accounts.included_in_finances_for(self)
end
def display_name
[ first_name, last_name ].compact.join(" ").presence || email
end
def initial
(display_name&.first || email.first).upcase
end
def initials
if first_name.present? && last_name.present?
"#{first_name.first}#{last_name.first}".upcase
else
initial
end
end
def show_ai_sidebar?
show_ai_sidebar
end
def ai_available?
return true unless Rails.application.config.app_mode.self_hosted?
effective_type = ENV["ASSISTANT_TYPE"].presence || family&.assistant_type.presence || "builtin"
case effective_type
when "external"
Assistant::External.available_for?(self)
else
openai_configured? || anthropic_configured?
end
end
def openai_configured?
Provider::Openai.configured?
end
def anthropic_configured?
Provider::Anthropic.configured?
end
def ai_enabled?
ai_enabled && ai_available?
end
def self.default_ui_layout
layout = Rails.application.config.x.ui&.default_layout || "dashboard"
layout.in?(%w[intro dashboard]) ? layout : "dashboard"
end
# SSO-only users have OIDC identities but no local password.
# They cannot use password reset or local login.
def sso_only?
password_digest.nil? && oidc_identities.exists?
end
# Check if user has a local password set (can authenticate locally)
def has_local_password?
password_digest.present?
end
# Attribute to skip password validation during SSO JIT provisioning
attr_accessor :skip_password_validation
# Deactivation
validate :can_deactivate, if: -> { active_changed? && !active }
after_update_commit :purge_later, if: -> { saved_change_to_active?(from: true, to: false) }
def deactivate
update active: false, email: deactivated_email
end
def can_deactivate
if admin? && family.users.count > 1
errors.add(:base, :cannot_deactivate_admin_with_other_users)
end
end
def purge_later
UserPurgeJob.perform_later(self)
end
def purge
if last_user_in_family?
family.destroy
else
reassign_owned_accounts!
destroy
end
end
# MFA
def setup_mfa!
update!(
otp_secret: ROTP::Base32.random(32),
otp_required: false,
otp_backup_codes: []
)
end
def enable_mfa!
raise ArgumentError, "OTP secret must be set before enabling MFA" if otp_secret.blank?
backup_codes = generate_backup_codes
# Store bcrypt digests only; this Postgres array cannot use AR encryption.
update!(
otp_required: true,
otp_backup_codes: backup_codes.map { |code| digest_backup_code(code) }
)
backup_codes
end
def disable_mfa!
transaction do
update!(
otp_secret: nil,
otp_required: false,
otp_backup_codes: []
)
webauthn_credentials.destroy_all
end
end
def verify_otp?(code)
return false if otp_secret.blank?
normalized_code = normalize_mfa_code(code)
return false if normalized_code.blank?
return true if totp.verify(normalized_code, drift_behind: 15)
return false unless backup_code_input?(normalized_code)
consume_backup_code!(normalized_code)
end
def provisioning_uri
return nil unless otp_secret.present?
totp.provisioning_uri(email)
end
def ensure_webauthn_id!
return webauthn_id if webauthn_id.present?
with_lock do
update!(webauthn_id: WebAuthn.generate_user_id) unless webauthn_id.present?
end
webauthn_id
end
def webauthn_enabled?
otp_required? && webauthn_credentials.exists?
end
def onboarded?
onboarded_at.present?
end
def needs_onboarding?
!onboarded?
end
def account_order
AccountOrder.find(default_account_order) || AccountOrder.default
end
def default_account_for_transactions
return nil unless default_account_id.present?
account = default_account
return nil unless account&.eligible_for_transaction_default? && account.family_id == family_id
account
end
# Dashboard preferences management
def dashboard_section_collapsed?(section_key)
preferences&.dig("collapsed_sections", section_key) == true
end
def dashboard_section_order
preferences&.[]("section_order") || default_dashboard_section_order
end
def update_dashboard_preferences(prefs)
# Use pessimistic locking to ensure atomic read-modify-write
# This prevents race conditions when multiple sections are collapsed quickly
transaction do
lock! # Acquire row-level lock (SELECT FOR UPDATE)
updated_prefs = (preferences || {}).deep_dup
prefs.each do |key, value|
if value.is_a?(Hash)
updated_prefs[key] ||= {}
updated_prefs[key] = updated_prefs[key].merge(value)
else
updated_prefs[key] = value
end
end
update!(preferences: updated_prefs)
end
end
# Reports preferences management
def reports_section_collapsed?(section_key)
preferences&.dig("reports_collapsed_sections", section_key) == true
end
def reports_section_order
preferences&.[]("reports_section_order") || default_reports_section_order
end
def update_reports_preferences(prefs)
# Use pessimistic locking to ensure atomic read-modify-write
transaction do
lock!
updated_prefs = (preferences || {}).deep_dup
prefs.each do |key, value|
if value.is_a?(Hash)
updated_prefs[key] ||= {}
updated_prefs[key] = updated_prefs[key].merge(value)
else
updated_prefs[key] = value
end
end
update!(preferences: updated_prefs)
end
end
# Transactions preferences management
def transactions_section_collapsed?(section_key)
preferences&.dig("transactions_collapsed_sections", section_key) == true
end
def show_split_grouped?
preferences&.dig("show_split_grouped") != false
end
def dashboard_two_column?
preferences&.dig("dashboard_two_column") == true
end
def preview_features_enabled?
preferences&.dig("preview_features_enabled") == true
end
def update_transactions_preferences(prefs)
transaction do
lock!
updated_prefs = (preferences || {}).deep_dup
prefs.each do |key, value|
if value.is_a?(Hash)
updated_prefs["transactions_#{key}"] ||= {}
updated_prefs["transactions_#{key}"] = updated_prefs["transactions_#{key}"].merge(value)
else
updated_prefs["transactions_#{key}"] = value
end
end
update!(preferences: updated_prefs)
end
end
private
def apply_ui_layout_defaults
self.ui_layout = (ui_layout.presence || self.class.default_ui_layout)
end
def apply_role_based_ui_defaults
if ui_layout_intro?
if guest?
self.show_sidebar = false
self.show_ai_sidebar = false
self.ai_enabled = true
else
self.ui_layout = "dashboard"
end
elsif guest?
self.ui_layout = "intro"
self.show_sidebar = false
self.show_ai_sidebar = false
self.ai_enabled = true
end
if leaving_guest_role?
self.show_sidebar = true unless show_sidebar
self.show_ai_sidebar = true unless show_ai_sidebar
end
if new_record? && member? && !ai_available?
self.show_ai_sidebar = false
end
end
def leaving_guest_role?
return false unless will_save_change_to_role?
previous_role, new_role = role_change_to_be_saved
previous_role == "guest" && new_role != "guest"
end
def skip_password_validation?
skip_password_validation == true
end
def default_dashboard_section_order
%w[cashflow_sankey outflows_donut net_worth_chart balance_sheet]
end
def default_reports_section_order
%w[trends_insights transactions_breakdown]
end
def ensure_valid_profile_image
return unless profile_image.attached?
unless profile_image.content_type.in?(%w[image/jpeg image/png])
errors.add(:profile_image, "must be a JPEG or PNG")
profile_image.purge
end
end
def last_user_in_family?
family.users.count == 1
end
def reassign_owned_accounts!
account_ids = owned_accounts.pluck(:id)
return if account_ids.empty?
new_owner = family.users.where.not(id: id)
.find_by(role: %w[admin super_admin]) ||
family.users.where.not(id: id)
.order(:created_at).first
return unless new_owner
Account.where(id: account_ids).update_all(owner_id: new_owner.id)
# Remove shares the new owner had for these accounts (they now own them)
AccountShare.where(account_id: account_ids, user_id: new_owner.id).delete_all
end
def deactivated_email
email.gsub(/@/, "-deactivated-#{SecureRandom.uuid}@")
end
def profile_image_size
if profile_image.attached? && profile_image.byte_size > 10.megabytes
errors.add(:profile_image, :invalid_file_size, max_megabytes: 10)
end
end
def totp
ROTP::TOTP.new(otp_secret, issuer: "Sure Finances")
end
def consume_backup_code!(normalized_code)
consumed = false
transaction do
lock!
if otp_backup_codes.present?
matching_index = otp_backup_codes.index do |stored_code|
backup_code_matches?(stored_code, normalized_code)
end
if matching_index
remaining_codes = otp_backup_codes.dup
remaining_codes.delete_at(matching_index)
update!(otp_backup_codes: remaining_codes)
consumed = true
end
end
end
consumed
end
def generate_backup_codes
MFA_BACKUP_CODE_COUNT.times.map { SecureRandom.hex(8) }
end
def digest_backup_code(code)
BCrypt::Password.create(normalize_mfa_code(code), cost: backup_code_digest_cost).to_s
end
def backup_code_matches?(stored_code, normalized_code)
if backup_code_digest?(stored_code)
return false unless backup_code_input?(normalized_code)
BCrypt::Password.new(stored_code).is_password?(normalized_code)
else
# Legacy plaintext codes are accepted once so existing MFA users are
# not locked out after backup-code hashing ships.
ActiveSupport::SecurityUtils.secure_compare(stored_code.to_s, normalized_code)
end
rescue BCrypt::Errors::InvalidHash
false
end
def backup_code_digest?(stored_code)
stored_code.to_s.start_with?("$2a$", "$2b$", "$2y$")
end
def normalize_mfa_code(code)
code.to_s.strip.downcase
end
def backup_code_input?(code)
backup_code_candidate?(code) || legacy_plaintext_backup_code_candidate?(code)
end
def backup_code_candidate?(code)
code.to_s.match?(/\A[0-9a-f]{16}\z/)
end
def legacy_plaintext_backup_code_candidate?(code)
code.to_s.match?(/\A[0-9a-f]{8}\z/)
end
def backup_code_digest_cost
ActiveModel::SecurePassword.min_cost ? BCrypt::Engine::MIN_COST : BCrypt::Engine.cost
end
end