cycat-service/backend/data-structure.md

86 lines
2.1 KiB
Markdown

# CyCAT data structure
The data structure is based on a Redis-compatible data store. [kvrocks](https://github.com/bitleak/kvrocks) is the Redis-compatible data store used for CyCAT
but any compatible Redis data store can be used.
# u:<UUID>
Each UUID inserted in CyCAT has at least an entry in the backend with the following format:
- `u:<UUID>` -> `value type`
## Value type available
|value|type|
|-----|-----------------------------------------|
| 1 | Publisher |
| 2 | Project |
| 3 | Item |
# <TYPE INT>:<UUID> (hash table)
Each UUID inserted might have a corresponding hash table
- `<type>:<UUID>` -> keys associated with the hash table type
## Reserved keys
- `raw` - raw data for the specific item
# t:<TYPE INT> (sorted set)
Each type got a sorted set (with a score of one) to easily paginate over the various types
- `t:<type>` -> `<UUID>`
# parent:<UUID> (set)
The parent(s) UUID of the UUID.
- `parent:<UUID>` -> {`UUID`, `UUID`}
# child:<UUID> (set)
The child(ren) UUID of the UUID.
- `child:<UUID>` -> {`UUID`, `UUID`}
# r:<UUID> (set)
The list of relationships from a specific UUID.
- `r:<UUID>` -> {`UUID`, `UUID`}
# rd:<UUID>:<UUID> (set)
The relationship data from an UUID tuple.
- `rd:<source UUID>:<destination UUID>` -> {`value`, `value`}
# id:<NAMESPACE>:<NAMESPACE_ID> (set)
A namespace id referenced in an UUID
- `id:<NAMESPACE>:<NAMESPACE_ID>` -> {`UUID`, `UUID`}
# idk:<NAMESPACE> (set)
Known id per namespace
- `idk:<NAMESPACE>' -> {`NAMESPACE_ID`, `NAMESPACE_ID`}
# Statistics
## Automatic API statistics
Prefix of API statistic is `stats:f:` followed by the function name called.
- `stats:f:generateuuid` : number of calls to the UUID generator API
# Fixed UUIDs for UUIDv5 generation
|value|description|
|-----|-----------|
|690b3b43-d689-481c-aa61-5351963a36f2|Used to generate the UUID of a GitHub url - concatenation key is `publisher:project`|
|39d6e10c-dac7-40e2-8e99-1ab1cefea6f4|Used to generate CyCAT OID for publisher and project|