SEAD JSON API Server (JAS)

This server provides a REST API like /site/1 which outputs all the information related to that site in an hierarchical JSON format.

This server also acts as the backend for a number of result section charts.

Since a MongoDB is used as a backend storage/cache for the JSON documents, the server can also accept simple queries like below.

To find which site has the analysis_entity 131350: /search/datasets.analysis_entities.analysis_entity_id/value/131350

To find all sites which contain the species 'spelta dicoccum': /search/lookup_tables.taxa.species/value/spelta%2Fdicoccum

Webclient Search API (New)

The old combined endpoint has been replaced with category-specific endpoints:

/search/:category/:search

Where :category must be one of:

• sites
• sample_groups
• datasets
• methods

Pagination params:

• limit (1-50, default 20)
• page (default 1)

Example:

• /search/sites/Glastonbury?limit=20&page=1
• /search/sample_groups/phosphate?limit=20&page=1
• /search/datasets/phosphate?limit=20&page=1
• /search/methods/phosphate?limit=20&page=1

Frontend Integration Change

The client should now perform 4 requests per search term (one per tab/category), instead of one combined request.

Recommended pattern:

1. User types query q
2. Fire 4 requests in parallel for page 1:
   • /search/sites/${q}?limit=20&page=1
   • /search/sample_groups/${q}?limit=20&page=1
   • /search/datasets/${q}?limit=20&page=1
   • /search/methods/${q}?limit=20&page=1
3. Render each tab independently from its own response
4. On scroll in one tab, only request the next page for that tab/category

Response Shape (All Categories)

• query
• category (key, label)
• pagination (page, limit, total, total_pages, has_more)
• items (paged list)

Item fields:

• site_id
• site_name
• matched_value
• score
• is_exact
• is_prefix
• is_contains
• is_fts_only
• sample_group_id (for sample_groups, otherwise null)
• dataset_id (for datasets, otherwise null)
• method_id (for methods, otherwise null)