Is your feature request related to a problem? Please describe.
I want to restrict what key values are permitted, but too many to generate a static list of.
Describe the solution you'd like
Regex support would be ideal, and offer enough flexibility for probably nearly any data I might accept.
Obviously in the above examples, no LUHN check occurs against cc, and ip_address technically accepts non-routable addresses, and phone_number is NA-centric.
But there is still tremendous value in being able to enforce regex constraints in policies.
Describe alternatives you've considered
Could generate massive lists of allowed values, but not feasible / maintainable when there's millions of possible combinations.
Explain any additional use-cases
Literally any (full or partial) data validation.
Additional context
Yes, data sanitization and validation should occur in the code which communicates with Vault. I get that.
But having regex constraints in Vault policy aids defense-in-depth and offers sanity checks in case of programming bugs, oversights, compromise, etc..
Is your feature request related to a problem? Please describe. I want to restrict what key values are permitted, but too many to generate a static list of.
Describe the solution you'd like Regex support would be ideal, and offer enough flexibility for probably nearly any data I might accept.
For example :
Obviously in the above examples, no LUHN check occurs against cc, and ip_address technically accepts non-routable addresses, and phone_number is NA-centric. But there is still tremendous value in being able to enforce regex constraints in policies.
Describe alternatives you've considered Could generate massive lists of allowed values, but not feasible / maintainable when there's millions of possible combinations.
Explain any additional use-cases Literally any (full or partial) data validation.
Additional context
Yes, data sanitization and validation should occur in the code which communicates with Vault. I get that. But having regex constraints in Vault policy aids defense-in-depth and offers sanity checks in case of programming bugs, oversights, compromise, etc..