REST

Query Rest

The parameter pretty=false can be added to create compressed output on many of the calls

Simple Query

http://localhost:32192/query?&index=pubs&q=title:gene&rows=10

Simple Query in csv

CSV output is served from the /query/csv path (there is no format parameter):

http://localhost:32192/query/csv?index=pubs&q=title:gene&rows=10&fl=id&fl=title

Multiple and Wildcard Indexes

Repeat the index parameter to query several indexes at once, and use * / ? wildcards to match index names by pattern:

http://localhost:32192/query?index=pubs&index=news&q=title:gene&rows=10
http://localhost:32192/query?index=logs-*&q=error&rows=10

Return Fields

http://localhost:32192/query?&index=pubs&q=title:gene&rows=10&fl=title

Filter Fields Returned

http://localhost:32192/query?&q=title:gene&rows=10&fl=-doi&fl=-abstract

http://localhost:32192/query?&index=pubs&facet=issn

http://localhost:32192/query?&index=pubs&facet=issn:100

http://localhost:32192/query?&index=pubs&facet=issn:100&drillDown=issn:1234-1234

Query Fields

http://localhost:32192/query?&index=pubs&q=cancer&qf=title&qf=abstract&rows=10

Minimum should match / Default Query Operator Example

http://localhost:32192/query?&index=pubs&q=gene biology genotype&qf=title&qf=abstract&mm=2&defaultOp=OR&rows=10

Sort

Can use -1/DESC or 1/ASC and multiple sort parameters to sort on for example title and then date

http://localhost:32192/query?&index=pubs&sort=title:-1&rows=10

Filter Query

http://localhost:32192/query?index=pubs&q=*:*&fq=title:cancer&rows=10

Advanced Filter Query

A filter clause with its own query fields, default operator, and minimum-should-match is expressed with qJson using "queryType":"FILTER" (there is no fqJson parameter):

http://localhost:32192/query?index=pubs&q=*:*&qJson={"queryType":"FILTER","q":"cancer diabetes genes","defaultOp":"OR","mm":2,"qf":["title","abstract"]}&rows=10

Score Function Query

Score functions are passed via the qJson parameter using the protobuf JSON format:

http://localhost:32192/query?index=pubs&rows=10&qJson={"queryType":"SCORE_MUST","q":"title:gene","scoreFunction":"zuliaScore * popularity"}

Multiple query clauses can be combined using multiple qJson parameters:

http://localhost:32192/query?index=pubs&rows=10&qJson={"queryType":"SCORE_MUST","q":"title:gene","qf":["title"],"scoreFunction":"zuliaScore * (1 + ln(pageRank))"}&qJson={"queryType":"FILTER","q":"year:[2020 TO *]"}

See Query-Syntax for the full expression reference.

More Like This Query

A MORE_LIKE_THIS clause finds documents similar to one or more sources (existing document IDs and/or raw text). It is expressed with qJson (see the Java Client for modes and tuning parameters):

# similar to an existing document, judged on the title and abstract fields
http://localhost:32192/query?index=pubs&rows=10&qJson={"queryType":"MORE_LIKE_THIS","moreLikeThisParams":{"documentId":["someDocId"],"field":["title","abstract"]}}

# similar to free text
http://localhost:32192/query?index=pubs&rows=10&qJson={"queryType":"MORE_LIKE_THIS","moreLikeThisParams":{"likeText":["CRISPR gene editing in agriculture"],"field":["title","abstract"]}}

By default the source documents are excluded from the results; add "includeSourceDocs":true inside moreLikeThisParams to keep them.

Realtime Query

Triggers an NRT (Near Real-Time) refresh to include recently stored but uncommitted documents in the results.

http://localhost:32192/query?index=pubs&q=title:gene&rows=10&realtime=true

Debug

Logs the parsed and rewritten Lucene query for each shard in the server logs. Useful for troubleshooting query behavior.

http://localhost:32192/query?index=pubs&q=title:gene&rows=10&debug=true

Concurrency

Sets the number of virtual threads used for parallel segment search within each shard. Defaults to the index or node-level defaultConcurrency setting if not specified.

http://localhost:32192/query?index=pubs&q=title:gene&rows=10&concurrency=4

Cursor Pagination

Cursor pagination provides constant-time paging regardless of depth. A sort on a unique field or unique combination is required, especially on a changing index where documents may be added or updated between pages.

Start pagination with cursor=0:

http://localhost:32192/query?index=pubs&q=title:gene&rows=10&sort=id:1&cursor=0

The response includes a cursor field (Base64-encoded token). Pass it back to get the next page:

http://localhost:32192/query?index=pubs&q=title:gene&rows=10&sort=id:1&cursor=CgwKCW15aW5kZXgS...

Repeat until the results array is empty.

Batch request

Batch mode streams all results as CSV using cursor pagination internally. A sort is required.

http://localhost:32192/query?&index=pubs&q=title:gene&rows=10&batch=true&batchSize=1000

Fields Rest

http://localhost:32192/fields/pubs

The query parameter style is also supported but the path-based style above is preferred:

http://localhost:32192/fields?index=pubs

Realtime Fields

http://localhost:32192/fields/pubs?realtime=true

Indexes Rest

http://localhost:32192/indexes

Index Schema Rest

http://localhost:32192/index?index=someIndex

Nodes Rest

http://localhost:32192/nodes

Stats Rest

http://localhost:32192/stats

Fetch Rest

http://localhost:32192/fetch?index=pubs&id=13442

Realtime Fetch

http://localhost:32192/fetch?index=pubs&id=13442&realtime=true

Terms Rest

All terms endpoints also accept realtime=true to include uncommitted changes.

All Terms for a field

http://localhost:32192/terms?index=pubs&fl=title

Terms with CSV Format

CSV output is served from the /terms/csv path (there is no format parameter):

http://localhost:32192/terms/csv?index=pubs&fl=title

Terms With Minimum Term Frequency

http://localhost:32192/terms?index=pubs&fl=title&minTermFreq=5000

Fuzzy Match of Terms

http://localhost:32192/terms?index=pubs&fl=abstract&fuzzyTermJson={"term":"gene","editDistance":1,"prefixLength":3}

Limit Terms

http://localhost:32192/terms?index=pubs&fl=title&minDocFreq=1000&amount=20

Range of Terms

http://localhost:32192/terms?index=pubs&fl=title&startTerm=cancer&endTerm=cells

Specific Terms

http://localhost:32192/terms?index=pubs&fl=abstract&includeTerm=cancer&includeTerm=gene

Associated Rest

Associated files use path-based routes under /associated/{index}/{id}/{fileName}.

Fetch a file

http://localhost:32192/associated/someIndex/123/readme.txt/file

Fetch a file’s metadata

http://localhost:32192/associated/someIndex/123/readme.txt/metadata

List the filenames for a document

http://localhost:32192/associated/someIndex/123/filenames

Download all files for a document as a zip

http://localhost:32192/associated/someIndex/123/bundle

List all associated documents in an index

Optionally filter with a q query parameter:

http://localhost:32192/associated/someIndex/all
http://localhost:32192/associated/someIndex/all?q=fileName:readme.txt

Store a file

POST the file as multipart form-data (a file part, plus an optional metaJson part) to the same path used to fetch it. Send the metaJson part before the file part so the server can bind the metadata alongside the streamed file:

curl -F 'metaJson={"author":"jane"}' -F "file=@readme.txt" \
  http://localhost:32192/associated/someIndex/123/readme.txt

Replication Rest

Reports per-shard and per-replica segment-replication progress for an index (cluster mode). Useful for monitoring replica lag and circuit-breaker state.

http://localhost:32192/replication/someIndex

Example response:

[
  {
    "shardNumber": 0,
    "lastAttemptedGeneration": 42,
    "replicas": [
      {
        "serverAddress": "10.0.0.51",
        "servicePort": 32191,
        "lastReplicatedGeneration": 42,
        "lagGenerations": 0,
        "lastSuccessMs": 1716900000000,
        "msSinceLastSuccess": 1200,
        "consecutiveFailures": 0,
        "circuitOpen": false,
        "backoffUntilMs": 0
      }
    ]
  }
]

Health Rest

Returns the health status of the Zulia node including CPU usage, memory usage, disk space, and MongoDB connectivity (cluster mode only). Each indicator reports UP, DEGRADED, or DOWN with detailed metrics.

Example Usage

curl http://localhost:32192/health