Query Param Level Option
| Field Name | Definition | Data Type | Example | Remarks |
|---|---|---|---|---|
| label | Name of label to traverse. | String | "graph_test" | Required. Must be an existing label. |
| direction | In/ out direction to traverse. | String | "out" | Optional. Default is "out". |
| limit | Number of edges to fetch. | Int | 10 | Optional. Default is 10. |
| offset | Starting position in the index. Used for paginating query results. | Int | 50 | Optional. Default is 0. |
| index | Specify index. | String | "_PK" |
Optional. Used with interval. |
| interval | Filters query results by range on indexed properties. | JSON (dictionary) | {"from": {"time": 0, "weight": 1}, "to": {"time": 1, "weight": 15}} |
Optional. Note that in order to use interval, you must specify index. (e.g. "index" : "index_name" ) |
| duration | Specify a time window and filter results by timestamp. | JSON (dictionary) | {"from": 1407616431000, "to": 1417616431000} |
Optional. |
| scoring | Scoring factors for property values in query results. These weights will be multiplied to corresponding property values, so that the query results can be sorted by the weighted sum. | JSON (dictionary) | {"time": 1, "weight": 2} |
Optional. |
| where | Filtering condition. Think of it as the "WHERE" clause of a SQL query. Currently, a selection of logical operators (and/ or) and relational operators ("equal(=)/ sets(in)/ range(between x and y)") are supported. Do not use any quotation marks for string type |
String | "((_from = 123 and _to = abcd) or gender = M) and is_hidden = false and weight between 1 and 10 or time in (1, 2, 3)". Note that it only supports long, string, and boolean types. |
Optional. |
| outputField | The usual "to" field of the result edges will be replace with a property field specified by this option. | String | "outputField": "service_user_id". This will output edge's "to" field into "service_user_id" property. | Optional. |
| exclude | For a label with exclude "true", resulting vertices of a traverse will act as a blacklist and be excluded from other labels' traverse results in the same step. | Boolean | "true" | Optional. Default is "false". |
| include | A Label with include "true" will ignore any exclude options in the step and the traverse results of the label guaranteed be included in the step result. | Boolean | "true" | Optional. Default is "false" |
| duplicate | Whether or not to allow duplicate edges; duplicate edges meaning edges with identical (from, to, label, direction) combination. | String One of "first", "sum", "countSum", or "raw". |
"first" | Optional. "first" means that only first occurrence of edge survives. "sum" means that you will sum up all scores of same edges but only one edge survive. "countSum" means to counts the occurrences of same edges but only one edge survives. "raw" means that you will allow duplicates edges. Default is "first". |
| rpcTimeout | Timeout for the request. | Integer | 100 | Optional. In milliseconds. Default is 100 and maximum is 1000. |
| maxAttempt | Max number of HBase read attempts. | Integer | 1 | Optional. Default is 1, and maximum is 5. |
| _to | Specify desired target vertex. Step results will only show edges to given vertex. | String | some vertex id | Optional. |
| threshold | Score threshold for query results. Edges with lesser scores will be dropped. | Double | 1.0 | Optional. Default is 0.0. |
| transform | "_to" values on resulting edges can be transformed to another value from their properties. | JSON (array of array) | Optional, default [["_to"]] | |
| cacheTTL | Caching the resulting edges and expire it after ttl time elapsed. | Long | 60000 | Optional. milliseconds |