Working Draft: KVAL (Key Value Access Language)

Grammar for accessing K,V databases with a bit more ease. First binding intended to be for Golang's BoltDB.

Operations:

Basic inserts and gets

  INS Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> Key :: Value
  INS Prime Bucket >> Secondary Bucket >> Tertiary Bucket
  GET Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> Key :: Value
  GET Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> Key
  GET Prime Bucket >> Secondary Bucket >> Tertiary Bucket
  GET Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> {PAT}
  GET Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> _ :: Value
  GET Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> _ :: {PAT}
  LIS Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> Key
  LIS Prime Bucket >> Secondary Bucket >> Tertiary Bucket 
  DEL Prime Bucket >> Secondary Bucket >> Tertiary Bucket
  DEL Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> Key
  DEL Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> Key :: _
  DEL Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> _      
  REN Prime Bucket >> Secondary Bucket >> Tertiary Bucket >>>> Key => Key
  REN Prime Bucket >> Secondary Bucket >> Tertiary Bucket => Third Bucket
  GET _

Functions

  INS   Insert
  GET   Get values
  LIS   Check existence
  DEL   Delete
  REN   Rename

Operators

  >>    Bucket:Bucket relationship
  >>>>  Bucket:Key relationship
  ::    Key::Value releationship
  =>    Name assignment
  _     Wildcard

Capabilities

  {PAT} Given a regex {PAT} for Key XOR Value, find match.

Restrictions

  Must be >= 1 Buckets for data. 
  {PAT} is not a valid option for an INS query.

Examples

  Bucket B exists inside Bucket A:                A >> B  
  Check existence of Bucket B inside Bucket A:    LIS A >> B              [Return: True]
  Get key value pairs in B:                       GET A >> B
  Get Root Bucket Contents                        GET _

  Key K1 exists inside Bucket A:                  A >>>> K1
  Check K1 exists inside Bucket A:                LIS A >>>> K1
  Get value for K1:                               GET A >>>> K1
  Add value for K1:                               INS A >>>> K1 :: V1
  Delete K1 value:                                DEL A >>>> K1 :: _
  Delete Values for Keys in Bucket A              DEL A >>>> _
  
  Delete A:                                       DEL A
  Delete B:                                       DEL A >> B
  
  Rename K1:                                      REN A >>>> K1 => K2
  Rename B:                                       REN A >>>> B => C

Recommendations

Idomatic methods for working with KVAL and storing date.

Data:

Data should be stored as Base64 and for those wanting to retrieve it, it helps to mark it up, like in the following example:

INS Data Bucket >>>> Data Key :: data:image/png:base64:<base64 encoded data>

We can break this down and see the following information in our simple encoding:

data:                    # let's us know we have a data blob
image/png:               # mimetype to help users decode the information
base64:                  # letting us know the data is stored in the db as base64
<based 64 encoded data>: # the encoded data portion

Therefore we can describe the following needed elements to encode data blobs:

data:<mimetype>:<encoding type>:<encoded data>

You can see how this is implemented in our BoltDB reference implementation: https://github.com/kval-access-language/kval-bbolt