Requirements

Regex in WHERE clause

• Users can use Regex to filter out values for a specific TEXT measurement.
• Similar to https://docs.influxdata.com/influxdb/v1.7/query_language/data_exploration/#use-a-regular-expression-to-specify-measurements-in-the-from-clause

Regex in FROM and SELECT clause

• Users can use Regex to select one or more time series based on time series path
• Similar to https://docs.influxdata.com/influxdb/v1.7/query_language/data_exploration/#use-a-regular-expression-to-specify-tag-values-in-the-where-clause and https://docs.influxdata.com/influxdb/v1.7/query_language/data_exploration/#use-a-regular-expression-to-specify-field-keys-and-tag-keys-in-the-select-clause

Grammar design

Regex in WHERE clause

Choice 1. Influx style

• Use / to delimit regular expressions
• Use =~ and !~ to denote match and does not match respectively

Examples:

select city from root.d1.s1 where city =~ /.*/ 
select city from root.d1.s1 where city !~ /Bei.*/

Pros & cons:

[+] Concise and easy to write.

[-] Introduces mental burden for users to be familiar with the new operators and delimiters; not self-explanatory

Choice 1. (Preferred) Postgres / MySQL style

• Use normal string delimiter (i.e. single quotes and double quotes)for regular expressions
• Use keywords MATCHES and NOT MATCHES to denote match and does not match respectively (since LIKE is already taken in IoTDB)