diff options
Diffstat (limited to 'docs')
| -rw-r--r-- | docs/about.org | 47 | ||||
| -rw-r--r-- | docs/architecture.plantuml | 72 | ||||
| -rw-r--r-- | docs/karuta-integration.org | 130 | ||||
| -rw-r--r-- | docs/transaction-algo.org | 12 |
4 files changed, 177 insertions, 84 deletions
diff --git a/docs/about.org b/docs/about.org new file mode 100644 index 0000000..6c64ceb --- /dev/null +++ b/docs/about.org @@ -0,0 +1,47 @@ +* About + +Sakura is a semantic-relational algebra engine inspired by +E. F. Codd's RM/T and D. McGoveran's (as understood by Fabian Pascal) +SRDM. It models relations in both intension and extension, providing +their manipulation as first-class elements of an algebra. + +* Motivation + +We believe the relational model was not ever fully implemented. The +fundamental necessity of expressing intent and binding meaning +enclosed to a conceptual model in data is seldom achieved by any +self-claimed relational database vendor. The emergence of different +physically concerned paradigms in the industry demonstrate not only a +large scale amnesia about fundamentals, but also a lethal +manifestation of problems that should not exist otherwise. The +relational data model as an applied theory is an universally rich and +concrete approach to reality that bases itself greatly in +moderate-realism applied to mathematics, as described in =An +Aristotelian Realist Philosophy of Mathematics: Mathematics as the +Science of Quantity and Structure= by J. Franklin. We believe that +unleashing the full potential of the relational model as refined over +the years despite overwhelming disregard of the industry as a whole, +will promote efficiency to a new scale never seen, that was very much +so cut short by relational pretenders and theoretically unsound +approaches. Our goal is to eliminate the application-database +distinction, while maintaining a clear separation between a first +order predicate logic algebra engine and a higher order predicate +logic decision engine (application) via an embedded data sublanguage +with Sakura as a possible auxiliary runtime. + +* Design Intent + +Sakura aims to treat relations as algebraic manipulation units with both: + +- extension (actual tuples, finite or generated) +- intension (schema + constraints + derivation meaning) + +The current engine already exposes the extension side robustly for +many operations, despite the lack of constraints greater than 1OP. +The main active work is making intension handling fully consistent and +reliable across planners/operators/protocols. + +* References + +- Codd, E. F. (1970), "A Relational Model of Data for Large Shared Data Banks" +- Codd, E. F. (1979), "Extending the Database Relational Model to Capture More Meaning" diff --git a/docs/architecture.plantuml b/docs/architecture.plantuml deleted file mode 100644 index 90985ba..0000000 --- a/docs/architecture.plantuml +++ /dev/null @@ -1,72 +0,0 @@ -@startuml -allowmixing - -package Client { - object "Query Interface" as query_interface -} - -package Algebra { - object "Unifier" as unifier - object "Bytecode Builder" as bytecode_builder - object "Parser" as parser - object "Lexer" as lexer -} - -query_interface --> lexer -lexer --> parser -parser --> unifier -unifier --> bytecode_builder - -package Query { - object "Schema Builder" as schema_builder - object "Facts Builder" as facts_builder - object "Optimizer" as optimizer - object "Commands Generator" as commands_generator - object "B+ Tree" as b_plus_tree -} - -bytecode_builder --> optimizer -optimizer --> commands_generator -facts_builder --> commands_generator -schema_builder --> commands_generator -commands_generator --> b_plus_tree -b_plus_tree --> query_interface - -package Coordinator { - object "File Historical Hashes" as file_historical_hashes - object "Location Hashes" as location_hashes - object "Command Parser" as command_parser - object "Curator" as curator -} - -commands_generator <-- command_parser - -package Disk { - object "Merkle Tree Root Generator" as merkle_tree - object "WAL Reader" as wal_reader - object "WAL Writer" as wal_writer - object "ARIES" as aries - object "SHA256 Hasher" as sha - object "Executor" as executor - map storage { - write => +λ - read => +λ - } - object "Courier" as courier - database Wal -} - -executor *--> merkle_tree -executor *--> sha -curator --> command_parser -command_parser --> executor -location_hashes *--> curator -file_historical_hashes *--> curator -executor --> aries -aries --> wal_reader -aries --> wal_writer -wal_writer --> Wal -wal_reader --> Wal -courier <-- wal_reader -courier --> storage -@enduml
\ No newline at end of file diff --git a/docs/karuta-integration.org b/docs/karuta-integration.org new file mode 100644 index 0000000..a144a25 --- /dev/null +++ b/docs/karuta-integration.org @@ -0,0 +1,130 @@ +* About Karuta + +Just like PostgreSQL and MySQL have SQL, Sakura has Karuta, a prolog-like language targetting the BEAM. [[karuta][https://github.com/dont-rely-on-nulls/karuta]] is the programming language that presents the different sublanguages to manage an RDBMS. + +We believe that what E. F. Codd envisioned was a stack of sublanguages that are embedded in a more powerful language, that expand it's abstraction scope as you move up the stack, trying to provide guarantees at each level. A data retrieval language should halt and has no business into being Turing Complete, but a predicate procedure language might. In light of this, Sakura is supposed to provide a common DSL that can differentiate between such sublanguages and execute things accordingly. + +The approach is that everything is a relation itself. So a constraint for instance is expressed as a relation less_than over NxN instead of simply being an operator '<'. + +* Sublanguages +Here's an overview of the sublanguages and how they are manifested in Karuta: + +** DRL (Data Retrieval Language) + +- Pure querying, total/terminating subset; +- No user recursion with no self references. + +** DCL (Data Constraint Language) + +- Integrity rules (key, FK-like, checks, assertions, aggregates-for-integrity). +- Semantically predicates, but with enforcement timing (on write/commit), not free query operators. + +** DML (Data Manipulation Language) + +- Insert/update/delete; +- Delta semantics, can trigger constraint evaluation. + +** DDL (Data Definition Language) + +- Define relations/views (ephemeral relations)/indexes/types/modules/predicate signatures and capabilities. + +** TCL (Transaction/Control Language) + +- Begin/commit/rollback, isolation/consistency controls, maybe savepoints. + +** PPL (Procedural Predicate Language) + +- Stored procedures/functions, recursion allowed, orchestration logic. +- Can call DRL/DML/DCL-aware operations, but not all PPL predicates are DRL-safe. + +** ACL (Authorization/Policy Language) (optional early, important later) + +- Who can query/mutate/define; row/predicate-level access policies. + +* Prototype + +#+begin_src prolog + % DDL + :- schema[users] { + :- signature[people-sig] { + person[_, _]. + another-person[string, natural]. + valid_age[natural]. + a-procedure[int, int]. + a-view[_, _, _]. + }. + :- module[people, people-sig] { + % This is using the arity directly, because the types are being specified in the signature. + % These Sakura-provided type extensions after '::' are the what informs Sakura through metadata on connection which type of Karuta predicate this is used as in Karuta. + another-person/2 :: sakura:persisted + another-person[Name, FavoriteNumber]. + + person[string, natural] :: sakura:persisted + person[Name, Age] :- valid_age[Age]. + + % DCL + another-constraint[list[person], string, natural] :: sakura:constraint + another-constraint[[], _, _]. + another-constraint[[person[Name, Age]|Rest], Whatever1, Whatever2] :- + (eq[Name, Whatever1] ; !eq[Name, Whatever1], !eq[Age, Whatever2]), + another-constraint[Rest, Whatever1, Whatever2]. + + % DCL + my-predicate[int, string, nat] :: sakura:constraint + my-predicate[Arg1, Arg2, Arg3] :- + another-module:my-predicate[Arg1, Arg2, Arg3], + on[insert, update], + timing[immediate]. + + % DCL + valid_age/1 :: sakura:constraint + valid_age[Age] with [karuta, sakura] :- + nat[Age], + greater_or_equal[Age, 0], + % DRL + aggregate[person, 2, Tuples], + another-constraint[Tuples, "lemos", 26], + on[insert, update], + timing[immediate], + target[age]. + + % PPL + a-procedure/2 :: sakura:stored + a-procedure[Arg1, Arg2] :- + eq[Arg1, Arg2]. + + % DDL + a-view[string, natural, natural] :: sakura:ephemeral + a-view[Name, Age, FavoriteNumber] :- + person[Name, Age], + another-person[Name, FavoriteNumber]. + }. + }. + + my-predicate[users:people:person, string, nat] :: signature + my-predicate[Arg1, Arg2, Arg3] :- + another-module:my-predicate[Arg1, Arg2, Arg3]. + + main[Arg1, Arg2] :- + karuta:plus[Arg1, Arg2, Out], + karuta:aggregate[numbers:inc, 2, Tuples], + + % Project directive should give us these configurations + project:sakura[Secret, Port, Host], + project:sakura:config[Users1, Commit], + + sakura:connect[Secret, Port, Host, Connection], + sakura:database[Connection, Users1, Commit, DB1Handle], + + % TCL + sakura:transaction[TXContext, DB1Handle], + + % DML + sakura:insert[users:people:person["Toad", 18], TXContext], + (sakura:error?[TXContext, Reason], sakura:rollback[TXContext], print["We rolled back: ~s!", Reason]) ; + (sakura:new[TXContext, [name], Difference1], + sakura:commit[TXContext, DB2Handle], + sakura:delta[DB1Handle, DB2Handle, Difference2], + % DRL + @relation1[Arg2, Out]). +#+end_src
\ No newline at end of file diff --git a/docs/transaction-algo.org b/docs/transaction-algo.org deleted file mode 100644 index 22903ec..0000000 --- a/docs/transaction-algo.org +++ /dev/null @@ -1,12 +0,0 @@ -* Check if temporary file with the command hash with exists -To hash the command, we must hash its bytes without the timestamp, -since it will always vary given it's a transactional time, and add -instead the commit HEAD of the state of the database. - -If such file exists, load the transaction and storage offsets -alongside the command and attempt to write it as a recovery. -* Get the transaction offset of the transaction graph file AND storage file -* Write the offsets to a tmp file -* Write the command with a flag **flushed** set to false to a tmp file -* Attempt to execute the command -* Delete the temporary file |