refactor!: make addrkit a multi-city address normalizer

- remove the events/outages module, ArcGIS adapter, portal config and event statuses
- add a configurable city dictionary/parser and make City part of StreetKey
- seed: read city/city_name, namespace IDs by city, keep structures distinct
- fix reliability: schema v2 with fail-closed load, deterministic ordering,
  input limits, config validation, atomic DB publish, LLM response caps
- add read-only shadow mode comparing the old ES path with addrkit
- breaking: registry.db must be re-seeded; config has no portal section
This commit is contained in:
Maksim Totmin
2026-10-27 14:35:51 +07:00
parent a0b4cb2b14
commit 17ff5c591f
38 changed files with 2978 additions and 1164 deletions
+85 -5
View File
@@ -4,6 +4,8 @@ import (
"os"
"path/filepath"
"testing"
"git.totmin.ru/en2zmax/addrkit/internal/domain/match"
)
func writeFile(t *testing.T, dir, name, content string) string {
@@ -19,7 +21,6 @@ func baseConfig(promptFields string) string {
return `{
"addr": ":6171",
"data_path": "registry.db",
"portal": { "url": "http://portal" },
"llm": { "enabled": true, "url": "http://llm"` + promptFields + ` }
}`
}
@@ -60,11 +61,90 @@ func TestLoadMissingSystemPromptFileFailsClosed(t *testing.T) {
}
}
func TestPortalURLOptional(t *testing.T) {
func TestDefaultsMatchMatcher(t *testing.T) {
dir := t.TempDir()
// No portal.url: the events module is disabled, the core still loads.
path := writeFile(t, dir, "config.json", `{"addr": ":6171", "data_path": "x.db"}`)
if _, err := Load(path); err != nil {
t.Fatalf("portal.url must be optional: %v", err)
cfg, err := Load(path)
if err != nil {
t.Fatalf("Load: %v", err)
}
want := match.DefaultOptions()
if cfg.Resolve.FuzzyThreshold != want.FuzzyThreshold ||
cfg.Resolve.AmbiguityMargin != want.AmbiguityMargin ||
cfg.Resolve.MaxCandidates != want.MaxCandidates {
t.Fatalf("config defaults %+v do not match matcher defaults %+v", cfg.Resolve, want)
}
}
func TestRejectsBadResolveRanges(t *testing.T) {
dir := t.TempDir()
cases := map[string]string{
"threshold": `{"addr":":6171","data_path":"x.db","resolve":{"fuzzy_threshold":0}}`,
"margin": `{"addr":":6171","data_path":"x.db","resolve":{"ambiguity_margin":2}}`,
"candidates": `{"addr":":6171","data_path":"x.db","resolve":{"max_candidates":0}}`,
}
for name, body := range cases {
path := writeFile(t, dir, name+".json", body)
if _, err := Load(path); err == nil {
t.Errorf("%s: expected validation error", name)
}
}
}
func TestCitiesAndDefaultCity(t *testing.T) {
dir := t.TempDir()
path := writeFile(t, dir, "config.json", `{
"addr": ":6171", "data_path": "x.db",
"cities": [
{"name": "Новосибирск", "aliases": ["нск", "новосибирская область"]},
{"name": "Бердск"}
],
"default_city": "Новосибирск"
}`)
cfg, err := Load(path)
if err != nil {
t.Fatalf("Load: %v", err)
}
if len(cfg.Cities) != 2 || cfg.DefaultCity != "Новосибирск" {
t.Fatalf("cities = %+v, default = %q", cfg.Cities, cfg.DefaultCity)
}
}
func TestRejectsUnknownDefaultCity(t *testing.T) {
dir := t.TempDir()
path := writeFile(t, dir, "config.json", `{
"addr": ":6171", "data_path": "x.db",
"cities": [{"name": "Бердск"}],
"default_city": "Новосибирск"
}`)
if _, err := Load(path); err == nil {
t.Fatal("expected error for default_city absent from cities")
}
}
func TestLoadCitiesFile(t *testing.T) {
dir := t.TempDir()
path := writeFile(t, dir, "cities.json", `[
{"name": "Новосибирск", "aliases": ["нск"]},
{"name": "Бердск"}
]`)
cities, err := LoadCitiesFile(path)
if err != nil {
t.Fatalf("LoadCitiesFile: %v", err)
}
if len(cities) != 2 || cities[0].Name != "Новосибирск" {
t.Fatalf("cities = %+v", cities)
}
}
func TestRejectsUnknownConfigFields(t *testing.T) {
dir := t.TempDir()
// "portal" was removed with the events module; silently ignoring it would
// hide a stale deployment config.
path := writeFile(t, dir, "config.json",
`{"addr":":6171","data_path":"x.db","portal":{"url":"http://example"}}`)
if _, err := Load(path); err == nil {
t.Fatal("expected error for unknown config field")
}
}