Files
compass_teams_nhs/internal/config/config.go
Sam Rolfe d7d529eb0f Phase 3: parse stored Compass pages into structured items
Turn the raw pages stored by Phase 2 into rows in the SQLite items table
(AGENTS.md §11 Phase 3, FR-3/FR-4/FR-5/FR-12).

- internal/parse: dependency-free HTML reader plus extractors for the ExtJS
  learning-task grid (subject, title, category, kind, due date, status), the
  home news feed (author, posted time, title, body) and the calendar week grid
  (day column + grid offset give the time; label gives title/room/teacher).
- internal/parse: fail loudly. A challenge page, login page, changed layout or
  unknown due-date format is an explicit error, never an empty day.
- FR-4: structure is parsed deterministically; importance is classified by an
  OmniRoute model (default openrouter/google/gemini-3.5-flash-lite) with a
  deterministic fallback. The classifier used is printed and recorded.
- internal/llm: small OpenAI-compatible OmniRoute client (handles SSE replies).
- internal/store: migration 002 adds source_page, importance and external_ref
  with a unique (child_id, source, type, external_ref) index; UpsertItems,
  CountItemsByType, SampleItems and ChildByName.
- cmd/school parse: reads var/raw/<child>/<run>, upserts items, records the scan
  row (ok=0 plus the error on failure) and prints counts and samples.
- scripts/parse-compass.sh and make parse run it in a one-off container against
  the service's own volume without restarting the service.
- docs/phase-3-parse-notes.md records the approach, the FR-4 decision and the
  known gaps.

Evidence on the live database: 56 rows for Harry (20 task, 1 message, 35 event).
Re-running the same raw pages inserts 0 and updates 56.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-15 17:27:25 +11:00

150 lines
4.3 KiB
Go

// Package config loads the Compass/NHS service configuration from the process
// environment. Every secret comes from the environment (environment.d); nothing
// is compiled into the binary and nothing secret is read from a committed file.
package config
import (
"errors"
"fmt"
"log/slog"
"os"
"strconv"
"strings"
"time"
// Embed the IANA time zone database so SCHOOL_TIMEZONE resolves even in a
// container without tzdata.
_ "time/tzdata"
)
// Config is the fully resolved runtime configuration.
type Config struct {
// Addr is the listen address, for example ":8792".
Addr string
// DBPath is the SQLite database file, for example "/data/school.db".
DBPath string
// SessionSecret signs the session cookie. Required, >= 16 bytes.
SessionSecret []byte
// AdminUsername is the single dashboard login name (Sam).
AdminUsername string
// AdminPassword is the plaintext password. Optional if AdminPasswordHash is set.
AdminPassword string
// AdminPasswordHash is a bcrypt hash. Optional if AdminPassword is set.
AdminPasswordHash string
// CookieSecure sets the Secure attribute on cookies. Keep true behind HTTPS.
CookieSecure bool
// SessionTTL is how long a session cookie stays valid.
SessionTTL time.Duration
// LogLevel is the slog level.
LogLevel slog.Level
// AppEnv is a label such as "production" or "development".
AppEnv string
// Version is injected at build time.
Version string
// Location is the family timezone. Due dates and briefs are wall-clock, so
// this must be the family's zone, not the container default of UTC.
Location *time.Location
}
// DefaultTimezone is the family timezone used when SCHOOL_TIMEZONE is unset.
const DefaultTimezone = "Australia/Sydney"
// Load reads and validates the configuration.
func Load() (*Config, error) {
cfg := &Config{
Addr: env("SCHOOL_ADDR", ":8792"),
DBPath: env("SCHOOL_DB_PATH", "/data/school.db"),
AdminUsername: env("SCHOOL_ADMIN_USERNAME", "sam"),
AdminPassword: os.Getenv("SCHOOL_ADMIN_PASSWORD"),
AdminPasswordHash: os.Getenv("SCHOOL_ADMIN_PASSWORD_HASH"),
CookieSecure: envBool("SCHOOL_COOKIE_SECURE", true),
AppEnv: env("SCHOOL_APP_ENV", "production"),
}
secret := os.Getenv("SCHOOL_SESSION_SECRET")
if len(secret) < 16 {
return nil, errors.New("SCHOOL_SESSION_SECRET must be set and at least 16 characters")
}
cfg.SessionSecret = []byte(secret)
if cfg.AdminPassword == "" && cfg.AdminPasswordHash == "" {
return nil, errors.New("set SCHOOL_ADMIN_PASSWORD or SCHOOL_ADMIN_PASSWORD_HASH")
}
if cfg.AdminUsername == "" {
return nil, errors.New("SCHOOL_ADMIN_USERNAME must not be empty")
}
ttl, err := time.ParseDuration(env("SCHOOL_SESSION_TTL", "12h"))
if err != nil {
return nil, fmt.Errorf("SCHOOL_SESSION_TTL: %w", err)
}
cfg.SessionTTL = ttl
lvl, err := parseLevel(env("SCHOOL_LOG_LEVEL", "info"))
if err != nil {
return nil, err
}
cfg.LogLevel = lvl
loc, err := time.LoadLocation(env("SCHOOL_TIMEZONE", DefaultTimezone))
if err != nil {
return nil, fmt.Errorf("SCHOOL_TIMEZONE: %w", err)
}
cfg.Location = loc
return cfg, nil
}
// DBConfig is the minimum a database-only command needs. The parse subcommand
// opens the same SQLite file as the service without requiring the service's
// session secret or login.
type DBConfig struct {
DBPath string
Location *time.Location
}
// LoadDB reads the database path and the family timezone only.
func LoadDB() (*DBConfig, error) {
loc, err := time.LoadLocation(env("SCHOOL_TIMEZONE", DefaultTimezone))
if err != nil {
return nil, fmt.Errorf("SCHOOL_TIMEZONE: %w", err)
}
return &DBConfig{
DBPath: env("SCHOOL_DB_PATH", "/data/school.db"),
Location: loc,
}, nil
}
func env(key, fallback string) string {
if v := strings.TrimSpace(os.Getenv(key)); v != "" {
return v
}
return fallback
}
func envBool(key string, fallback bool) bool {
v := strings.TrimSpace(os.Getenv(key))
if v == "" {
return fallback
}
b, err := strconv.ParseBool(v)
if err != nil {
return fallback
}
return b
}
func parseLevel(s string) (slog.Level, error) {
switch strings.ToLower(strings.TrimSpace(s)) {
case "debug":
return slog.LevelDebug, nil
case "info", "":
return slog.LevelInfo, nil
case "warn", "warning":
return slog.LevelWarn, nil
case "error":
return slog.LevelError, nil
default:
return 0, fmt.Errorf("SCHOOL_LOG_LEVEL: unknown level %q", s)
}
}