Search

Using it

curl -G 'https://api.cmskite.com/v1/blog/search' \
  -H "authorization: Bearer $KEY" \
  --data-urlencode 'q=port congestion' \
  --data-urlencode 'limit=20'

Results come back in the same shape as a post listing, so the code that renders a listing renders a search result without changes. They are ranked by relevance, which is not a stable order to page through, so a search answers with one page and no cursor: ask for as many results as you want to show with limit.

What it does and does not do

• Searches titles, excerpts and bodies, with a match in the title counting most, then the excerpt.
• Matches whole words, not parts of words: port finds “port” but not “export” or “portal”. Case does not matter, English words match their basic plurals (ports finds “port”), and common words such as “the” and “and” are ignored. Other languages, Hindi included, match words only as they are spelled.
• Published content only with a project API key. A draft is not findable by searching for it with the key your site uses.
• Handles the punctuation people type. Quotes, apostrophes and stray brackets come back as results rather than as a parse error.
• Ranks by relevance.
• Scoped to one project, because the credential is.

In a site

// app/api/search/route.ts -- runs on your server, so the key stays there.
export async function GET(request: Request) {
  const q = new URL(request.url).searchParams.get('q')?.trim()
  if (!q) return Response.json({ data: [] })

  const url = new URL('https://api.cmskite.com/v1/blog/search')
  url.searchParams.set('q', q)
  url.searchParams.set('limit', '20')

  const res = await fetch(url, {
    headers: { authorization: `Bearer ${process.env.CMSKITE_API_KEY}` },
    // Search results change when content does, not on a timer you pick.
    next: { revalidate: 300 },
  })
  if (!res.ok) return Response.json({ data: [] }, { status: 502 })
  return Response.json(await res.json())
}