feat(search+docs): trigram fuzzy search, brand/country filters, developer docs
CI / Python (ingestion) (pull_request) Successful in 12s
CI / Migrations (postgres) (pull_request) Successful in 22s
CI / Go (api) (pull_request) Successful in 47s

Search:
- migration 0009: trigram GIN index on brand.name + btree on country_of_origin
- SearchProducts: typo-tolerant word_similarity matching (>=0.42) on top of
  ILIKE substring + barcode; new brand/country filters; rank by
  similarity * (0.5 + quality_score). Response gains country_of_origin,
  quality_score and per-result relevance score.
- public search UI: brand/country filter inputs; show country in results

Docs:
- serve embedded OpenAPI 3 spec at GET /api/v1/openapi.json (not rate limited)
- ApiDocs page: auth + rate-limit section, updated search params/response
- docs/api.md developer guide

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
This commit is contained in:
novaalphastrikeomegaz663
2026-06-20 09:38:27 +00:00
parent cf255e2380
commit 69a0149bbe
11 changed files with 613 additions and 48 deletions
+67 -22
View File
@@ -82,11 +82,22 @@ func (s *Store) ProductBarcodes(ctx context.Context, productID string) ([]Barcod
// ProductSummary is a lightweight row used in search/listing responses.
type ProductSummary struct {
ID string `json:"id"`
GTIN *string `json:"gtin"`
Name string `json:"name"`
Brand *string `json:"brand"`
CategoryPath *string `json:"category_path"`
ID string `json:"id"`
GTIN *string `json:"gtin"`
Name string `json:"name"`
Brand *string `json:"brand"`
CategoryPath *string `json:"category_path"`
Country *string `json:"country_of_origin"`
QualityScore float64 `json:"quality_score"`
Score *float64 `json:"score,omitempty"`
}
// SearchFilters bundles the optional filters accepted by SearchProducts.
type SearchFilters struct {
Query string // fuzzy name / barcode query
Category string // ltree path; matches the subtree
Brand string // fuzzy brand name
Country string // country_of_origin prefix (case-insensitive)
}
const productSelect = `
@@ -148,34 +159,67 @@ func (s *Store) ProductByID(ctx context.Context, id string) (*Product, error) {
return p, nil
}
// SearchProducts performs a fuzzy name search with optional category subtree filter.
func (s *Store) SearchProducts(ctx context.Context, q, category string, limit, offset int) ([]ProductSummary, int, error) {
// fuzzyThreshold is the minimum word_similarity for a name to be considered a
// fuzzy match. ~0.42 tolerates common typos (e.g. "choclate"→"Chocolate")
// without returning unrelated products.
const fuzzyThreshold = "0.42"
// SearchProducts runs a trigram-fuzzy name search with optional category /
// brand / country filters. When a query is present, matching is inclusive
// (substring OR trigram-similar OR barcode), and results are ranked by name
// similarity blended with quality_score so the best, most-complete records
// surface first. Without a query, results are ordered by quality_score.
func (s *Store) SearchProducts(ctx context.Context, f SearchFilters, limit, offset int) ([]ProductSummary, int, error) {
args := []any{}
where := "WHERE p.status = 'active'"
if q != "" {
args = append(args, q)
where += ` AND (p.name ILIKE '%' || $1 || '%'
qIdx := 0
if f.Query != "" {
args = append(args, f.Query)
qIdx = len(args)
q := "$" + strconv.Itoa(qIdx)
where += ` AND (p.name ILIKE '%' || ` + q + ` || '%'
OR word_similarity(` + q + `, p.name) >= ` + fuzzyThreshold + `
OR EXISTS (SELECT 1 FROM product_barcode pb
WHERE pb.product_id = p.id AND pb.gtin ILIKE '%' || $1 || '%'))`
WHERE pb.product_id = p.id AND pb.gtin ILIKE '%' || ` + q + ` || '%'))`
}
if category != "" {
args = append(args, category)
if f.Category != "" {
args = append(args, f.Category)
where += " AND c.path <@ $" + strconv.Itoa(len(args)) + "::ltree"
}
if f.Brand != "" {
args = append(args, f.Brand)
where += " AND b.name ILIKE '%' || $" + strconv.Itoa(len(args)) + " || '%'"
}
if f.Country != "" {
args = append(args, f.Country)
where += " AND p.country_of_origin ILIKE $" + strconv.Itoa(len(args)) + " || '%'"
}
from := `FROM product p
LEFT JOIN brand b ON b.id = p.brand_id
LEFT JOIN category c ON c.id = p.category_id `
countSQL := "SELECT count(*) FROM product p LEFT JOIN category c ON c.id = p.category_id " + where
var total int
if err := s.pool.QueryRow(ctx, countSQL, args...).Scan(&total); err != nil {
if err := s.pool.QueryRow(ctx, "SELECT count(*) "+from+where, args...).Scan(&total); err != nil {
return nil, 0, err
}
// Ranking: when querying, similarity drives order, multiplied by a
// quality factor floored at 0.5 so low-quality records aren't zeroed out.
scoreExpr := "NULL::real"
orderBy := "p.quality_score DESC, p.name"
if f.Query != "" {
q := "$" + strconv.Itoa(qIdx)
scoreExpr = "word_similarity(" + q + ", p.name)"
orderBy = scoreExpr + " * (0.5 + p.quality_score) DESC, p.quality_score DESC, p.name"
}
args = append(args, limit, offset)
listSQL := `
SELECT p.id, p.gtin, p.name, b.name, c.path::text
FROM product p
LEFT JOIN brand b ON b.id = p.brand_id
LEFT JOIN category c ON c.id = p.category_id ` + where +
" ORDER BY p.name LIMIT $" + strconv.Itoa(len(args)-1) + " OFFSET $" + strconv.Itoa(len(args))
listSQL := "SELECT p.id, p.gtin, p.name, b.name, c.path::text, p.country_of_origin, p.quality_score, " +
scoreExpr + " AS score " + from + where +
" ORDER BY " + orderBy +
" LIMIT $" + strconv.Itoa(len(args)-1) + " OFFSET $" + strconv.Itoa(len(args))
rows, err := s.pool.Query(ctx, listSQL, args...)
if err != nil {
@@ -186,7 +230,8 @@ LEFT JOIN category c ON c.id = p.category_id ` + where +
out := []ProductSummary{}
for rows.Next() {
var ps ProductSummary
if err := rows.Scan(&ps.ID, &ps.GTIN, &ps.Name, &ps.Brand, &ps.CategoryPath); err != nil {
if err := rows.Scan(&ps.ID, &ps.GTIN, &ps.Name, &ps.Brand, &ps.CategoryPath,
&ps.Country, &ps.QualityScore, &ps.Score); err != nil {
return nil, 0, err
}
out = append(out, ps)