Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/devRepro/reprodisseny/llms.txt

Use this file to discover all available pages before exploring further.

Returns ranked autocomplete suggestions matching a search term across published products and categories. Results are cached in Nitro server storage for 5 minutes per unique query. Method: GET
Path: /api/search/suggest
Caching: 5 minutes in-memory (Nitro storage)
Search is powered by a lightweight scoring function that ranks results by prefix match (score 100) and substring match (score 50). Diacritics are normalised so that "gran" matches "Gran Formato" and "Gràfic".

​
Query parameters

​
q
string
required
Search term. Whitespace is trimmed. Returns an empty result set when blank.
​
limit
number
default:"8"
Maximum number of results to return. Clamped between 1 and 30.
​
force
string
Set to "1" to bypass the Nitro cache and force a fresh search. Useful for testing.

​
Response

​
200 OK

​
q
string
required
The raw search term from the request.
​
items
object[]
required
Array of matching items, ordered by score (products before categories at equal score).

​
Scoring and ranking

The scoring function normalises both the search term and candidate titles to lowercase NFD form before comparing: 
function score(title: string, term: string): number {
  const t = normText(title)   // lowercase + remove diacritics
  if (!t) return 0
  if (t.startsWith(term)) return 100   // prefix match
  if (t.includes(term)) return 50      // substring match
  return 0                             // no match
}
Results with score 0 are excluded. Within the same score tier, results are sorted alphabetically by title. Products appear before categories in the final merged list.

​
Filtering

Only items that pass both conditions are included in the search pool:
  • isPublished !== false
  • hidden !== true
Categories must additionally have showInNav !== false.

​
Caching

Results are stored in Nitro’s cache storage layer with a 5-minute TTL: 
const cacheKey = `search:suggest:${term}:${limit}`
await storage.setItem(cacheKey, payload, { ttl: 60 * 5 })
Use ?force=1 to bypass the cache during development.

​
Example

curl "https://reprodisseny.com/api/search/suggest?q=gran&limit=5"

{
  "q": "gran",
  "items": [
    {
      "id": "gran-formato",
      "kind": "categoria",
      "title": "Gran Formato",
      "href": "/categorias/gran-formato",
      "image": "https://webcms.blob.core.windows.net/media/categories/gran-formato.jpg"
    },
    {
      "id": "lonas-gran-formato",
      "kind": "producto",
      "title": "Lonas Gran Formato",
      "href": "/productos/lonas-gran-formato",
      "image": null
    }
  ]
}

​
Triggering search from Vue

const results = await $fetch("/api/search/suggest", {
  query: { q: searchQuery.value, limit: 8 },
})
// results.items -> SuggestItem[]

Build docs developers (and LLMs) love

Get started for freeTalk to us