ClickHouse Query Analyser

Go lib for analysis of ClickHouse columns usage in SQL

Installation

go get github.com/iimos/clickhouse_query_analyser

Usage

import "github.com/iimos/clickhouse_query_analyser/columnusage"

schema := []columnusage.Column{
    {"system", "numbers", "number"},
    {"system", "clusters", "cluster"},
    {"system", "clusters", "shard_num"},
    {"system", "clusters", "shard_weight"},
    {"system", "clusters", "replica_num"},
    {"system", "clusters", "host_name"},
    {"system", "clusters", "host_address"},
    {"system", "clusters", "port"},
    {"system", "clusters", "is_local"},
    {"system", "clusters", "user"},
    {"system", "clusters", "default_database"},
    {"system", "clusters", "errors_count"},
    {"system", "clusters", "estimated_recovery_time"},
    // ...
}

query := `
    select clu.host_name as hostName, coalesce(val.value, 0) as value
    from system.clusters as clu
    left join (
    select fullHostName() as hostName, sum(rows) as value
    from cluster('shards', system, part_log)
    where Database = 'raw'
      and table = 'events'
      and event_type = 'NewPart'
      and event_time between subtractMinutes(now(), 5) and now()
    group by hostName
    ) as val on clu.host_name = val.hostName
    where clu.cluster = 'shards'
`

analyser := columnusage.NewColumnUsageAnalyser(schema)
res, err := analyser.ParseQuery("default", query)
if err != nil {
    // ...
}
fmt.Printf("%v", res)