Text Parsing Examples

Real-world usage of transformText, readNip94Tags, and nip94TagsToIMeta from the Snort app.

Cached text transformation for performance

Text parsing can be expensive in timelines with many notes. The app caches results by event ID:

// useTextTransformCache.tsx
export function transformTextCached(id: string, content: string, tags: Array<Array<string>>): Array<ParsedFragment> {
  const newCache = transformText(content, tags)
  return newCache
}

Extracting media from events using NIP-94 imeta

Combine transformText fragments with readNip94TagsFromIMeta to get full file metadata for attachments:

// getEventMedia.ts
const meta = event.tags.filter(a => a[0] === "imeta").map(readNip94TagsFromIMeta)

Building imeta tags from upload metadata

When creating notes with file attachments, convert NIP-94 metadata back to imeta tags:

// NoteCreator.tsx
const n94 = (at.nip94?.length ?? 0) > 0
  ? readNip94Tags(at.nip94!)
  : ({ url: at.url, hash: at.sha256, size: at.size, mimeType: at.type } as Nip94Tags)

n94.fallback ??= []
n94.fallback.push(...v.slice(1).filter(a => a.url).map(a => a.url!))
extraTags?.push(nip94TagsToIMeta(n94))