Let your customers engage right at the search bar by giving them more useful, user-centric search suggestion
- Instant search suggestions and spell check
- Low Latency ~ Response within 20 ms for 1M records
- Order doesn't matter “iPhone 6s” or “6s iPhone” yield same result
- Unlimited Filters and Parameters support
- Support personalization
- Customizable parameters like Servers port and recovery methods[Refer #config.properties].
- Apache Maven >= 3.3.9
- Java ~ 1.8
1. Git clone https://github.com/nikcomestotalk/autosuggest.git
2. cd autosuggest
3. mvn install
4. mvn exec:java
1. docker run -d -p 1082:1082 -p 1081:1081 --name autosuggest autosuggest/autosuggest
- Fork this repo and clone your fork.
- Make your desired changes
- Add tests for your new feature and ensure all tests are passing
- Commit and push
- Submit a Pull Request through Github's interface and I'll review your changes to see if they make it to the next release.
Use this repo's github issues.
URL : 127.0.0.1:1081/query
METHOD : POST
POST PARAMS:
{
"query": "iphone",
"filter": {
"param1":1,
"param2":2
},
"bucket": {
"p1": {
"value" : 7007,
"weight": 90
},
"p2": {
"value":57,
"weight":10
}
},
"limit":10
}
Query : Query of which you want suggestion
Filter : Filter out suggestions, you can set unlimited filters.
Bucket : Order your suggestion on the basis of parameters specified within bucket key, Max 3 allowed
Limit : Limit the suggestions
[
{
"matched": "iphone 6s",
"parameters": {
"param1": 1,
"param2":2,
"location": 4321
},
"error": ""
},
{
"matched": "Iphone 6s 64gb ",
"parameters": {
"param1": 1,
"param2":2,
"user": 1,
"location": 4321
},
"error": ""
}
]
How to put data into Suggest Engine.
URL : 127.0.0.1:1081/update
METHOD : POST
POST PARAMS:
{
"query": "iphone fc",
"parameter": {
"p1": 16,
"p2": 12,
"p3": 34
}
Query : Text you want to save.
parameter : Tagging the text, which can be used for both filter and bucket parameters.