ssb-query

A scuttlebot plugin for querying data. With map-filter-reduce you can write pretty flexible queries, similar to SQL, but more javascripty.

ssb-query is just a thin layer of glue, giving access to flumeview-query with secure-scuttlebutt data.

installation

ssb-query is included in the ssb-server distribution by default. see plugins documentation

usage

command line

if you are running ssb-server, run the following queries from another tab on the same machine.

ssb-server query.read --query '{MFR_QUERY}' options...

notice the json is inside single quotes ''. this is necessary, because " part of JSON but is handled specially on the command line.

see read api documentation for options

javascript

using ssb-client, connect to a locally running sbot and call query.read, which returns a pull-stream

require('ssb-client')(function (err, sbot) {
  if(err) throw err
  pull(
    sbot.query.read({
      query: MFR_QUERY,
      ...other options
      //limit: 10, reverse: true
    }),
    pull.collect(function (err, ary) {
      console.log(ary)
    })
  )
})

api

query.read ({query,limit,reverse,old,live})

perform a query. query is a map-filter-reduce query. limit,reverse,old and live are standard options supported by most ssb database stream apis, see createLogStream

query.explain ({query})

returns internal information about what index will be used my ssb-query. the name comes from the SQL "EXPLAIN" command

output might look like this:

{
  gte: [...], lte: [...] //if this is present, then the query will use an index.
  scan: true | false, //if scan is true, the entire database will be examined. this means a very slow query
  live, old, //wether to include new and old records
  sync: false//include a {sync: true} message after the old records have finished.
}

see flumeview-query for more information.

example queries

all messages in "solarpunk" channel.

[{
  "$filter": {
    value: {
      content: {channel: "solarpunk"}
    }
  }
}]

all replies in a thread

[{
  "$filter": {
    value: {
      content: {root: "%<msg_id>"}
    }
  }
}]

messages by a type by an author

[{
  "$filter": {
    value: {
	  author: "@<author_id>",
	  content: {
	    type: "<msg_type>"
	  }
	}
  }
}]

channels, with count and sort

most recently published channels, with timestamp and message count.

[
  {"$filter": {"value": {"content":{ "channel": {"$is": "string"}, "type": "post"}}}},
  {"$reduce": {
      "channel": ["value", "content", "channel"],
      "count": {"$count": true},
      "timestamp": {"$max": ["value", "timestamp"]}
  }},
  {"$sort": [["timestamp"], ["count"]]}
]

sample output:

{
  "channel": "heropunch",
  "count": 83,
  "timestamp": 1537465741596
}

{
  "channel": "walkaway",
  "count": 43,
  "timestamp": 1537471373721
}

{
  "channel": "music",
  "count": 635,
  "timestamp": 1537474414933
}

indexes

indexes make your queries much faster. read about flumeview-query indexes

currently supported indexes:

var indexes = [
  {key: 'log', value: ['timestamp']},
  {key: 'clk', value: [['value', 'author'], ['value', 'sequence']] },
  {key: 'typ', value: [['value', 'content', 'type'], ['timestamp']] },
  {key: 'tya', value: [['value', 'content', 'type'], ['value', 'timestamp']] },
  {key: 'cha', value: [['value', 'content', 'channel'], ['timestamp']] },
  {key: 'aty', value: [['value', 'author'], ['value', 'content', 'type'], ['timestamp']]},
  {key: 'ata', value: [['value', 'author'], ['value', 'content', 'type'], ['value', 'timestamp']]},
  {key: 'art', value: [['value', 'content', 'root'], ['value', 'timestamp']]},
  {key: 'lor', value: [['rts']]}
]

so, because the of cha index, a query for value.content.channel and timestamp would be quick.

License

MIT