reifyhealth.specmonstah.core — reifyhealth/specmonstah 2.1.0

add-edge-with-id clj/s

(add-edge-with-id g ent-name related-ent-name id)

When indicating :ent-a references :ent-b, include a :relation-attrs graph attribute that includes the attributes via which :ent-a references :ent-b.

For example, if the :project named :p0 has an :owner-id and :updated-by-id that both reference the :user named :u0, then the edge from :p0 to :u0 will if a :relation-attrs attribute with value #{:owner-id :updated-by-id}.

This can be used e.g. to set the values for :owner-id and :updated-by-id.

add-ent clj/s

(add-ent {:keys [data] :as db} ent-name ent-type query-term)

Add an ent, and its related ents, to the ent-db

add-ent-type-query clj/s

(add-ent-type-query db ent-type-query ent-type)

A query is composed of ent-type-queries, where each ent-type-query specifies the ents that should be created for that type. This function adds the ents for an ent-type-query.

add-ents clj/s

(add-ents {:keys [schema] :as db} query)

Produce a new db with an ent graph that contains all ents specified by query

add-n-ents clj/s

(add-n-ents db ent-type num-ents query-term)

Used when a query is something like [3]

add-ref-ents clj/s

(add-ref-ents db)

Ents are added in two stages: first, all ents that are declared in the query are added. During that process, we keep track of ref-ents, ents which are automatically generated in order to satisfy relations. This function adds those ref ents if an ent of the same name doesn't exist already.

add-related-ents clj/s

(add-related-ents {:keys [data] :as db} ent-name ent-type query-term)

all-ent-relations clj/s

(all-ent-relations db)

(all-ent-relations db ents)

Given a db, returns a map of ent-type to map of entity relations.

An example return value is: {:patient {:p0 {:created-by :u0 :updated-by :u1} :p1 {:created-by :u0 :updated-by :u2}} :user {:u0 {:friends-with :u0}}}

assoc-referenced-val clj/s

(assoc-referenced-val ent-data relation-attr relation-val constraints)

Look up related ent's attr value and assoc with parent ent attr. :coll relations will add value to a vector.

assoc-referenced-vals clj/s

(assoc-referenced-vals db {:keys [ent-name visit-key visit-val]})

A visiting function that sets referenced values.

Given a schema like {:user {} :post {:relations {:created-by [:user :id]}}}

a :post's :created-by key gets set to the :id of the :user it references.

attr-map clj/s

(attr-map db attr)

(attr-map {:keys [data] :as db} attr ents)

Produce a map where each key is a node and its value is a graph attr on that node

bound-descendants? clj/s

(bound-descendants? {:keys [types schema relation-graph]}
                    query-bindings
                    related-ent-type)

Check whether query-relations contains bindings that apply to any descendants of related-ent-type

bound-relation-attr-name clj/s

(bound-relation-attr-name {:keys [schema]} ent-name related-ent-type index)

Template for when a binding necessitates you add a new entity

bound-relation-attr-name-source clj/s

(bound-relation-attr-name-source ent-name)

coll-relation-attr? clj/s

(coll-relation-attr? db ent relation-attr)

Given a db, ent, and relation-attr, determines whether the relation is a coll attr.

conformed-query-opts clj/s

(conformed-query-opts query-term relation-attr)

These conformed query opts allow us to 1) validate the query term and 2) dispatch on what kind of query was supplied.

This is one of the most complicated parts of SM because users can supply different types of values for the :refs key of a query:

1. an ent-name for unary relations (type: :ent-name)
2. a vector of ent-names for coll relations (type: :ent-names)
3. a number for coll relations (type: :ent-count)

These types are captured by the ::refs specs, and the specs it composes. The type for the supplied query-term is returned as :qr-type.

Each of these qr-types corresponds to a constraint of either :coll or :unary. The constraint is returned as :qr-constraint.

Conforming the query opts provides the qr-type and qr-constraint so that dependent functions can dispatch on these values.

default-node-name clj/s

(default-node-name {:keys [schema]} ent-type)

Whenever specmonstah needs to create a node that's not manually named, it uses this to generate the default name.

ent-attr clj/s

(ent-attr {:keys [data]} ent-name attr)

ent-attrs clj/s

(ent-attrs {:keys [data]} ent-name)

ent-index clj/s

(ent-index g ent-type)

Used to keep track of entity's insertion order in graph relative to other entities of the same type.

ent-related-by-attr? clj/s

(ent-related-by-attr? db ent-name related-ent relation-attr)

Is ent A related to ent B by the given relation-attr?

ent-relation-constraints clj/s

(ent-relation-constraints db ent relation-attr)

ent-relations clj/s

(ent-relations db ent)

Given a db and an ent, returns a map of relation attr to ent-name.

ent-schema clj/s

(ent-schema {:keys [schema data]} ent-name)

Given an ent name, return the schema of its corresponding type

ents clj/s

(ents {:keys [data]})

returns all ents in the ent db

ents-by-type clj/s

(ents-by-type db)

(ents-by-type db ents)

Given a db, returns a map of ent-type to a set of entities of that type. Optionally pass in a seq of the ents that should be included.

identical-prefixes clj/s

(identical-prefixes schema)

Schemas are invalid if two types have the same prefix. This checks that.

incrementing-node-name clj/s

(incrementing-node-name {:keys [data schema]} ent-type)

A template for creating distinct node names.

init-db clj/s

(init-db {:keys [schema] :as db} query)

invalid-constraints clj/s

(invalid-constraints schema)

Constraints that reference nonexistent relation attrs

invalid-schema-relations clj/s

(invalid-schema-relations schema)

Relations that reference nonexistent types

merge-overwrites clj/s

(merge-overwrites _db {:keys [visit-val visit-query-opts schema-opts]})

Overwrites generated data with what's found in schema-opts or visit-query-opts.

numeric-node-name clj/s

(numeric-node-name schema ent-type index)

Template for generating a node name

omit-relation? clj/s

(omit-relation? db ent-name reference-key)

omit? clj/s

(omit? x)

prune-required clj/s

(prune-required {:keys [data] :as db} required-attrs)

Updates ent graph, removing edges going from a 'required' ent to the 'requiring' ent, thus eliminating cycles.

query-ents clj/s

(query-ents {:keys [data] :as _db})

Get seq of nodes that are explicitly defined in the query

query-opts clj/s

(query-opts {:keys [data]} ent-name)

query-relation clj/s

(query-relation db ent-name relation-attr)

Returns the conformed relation for an ent's relation-attr. Handles polymorphic relations.

referenced-ent-attrs clj/s

(referenced-ent-attrs {:keys [data] :as db} ent-name)

seq of [referenced-ent relation-attr]

related-ents clj/s

(related-ents {:keys [schema data] :as db}
              ent-name
              relation-attr
              related-ent-type
              query-term)

Returns all related ents for an ent's relation-attr

related-ents-by-attr clj/s

(related-ents-by-attr {:keys [data] :as db} ent-name relation-attr)

All ents related to ent via relation-attr

relation-attrs clj/s

(relation-attrs {:keys [data]} ent-name referenced-ent)

Given an ent A and an ent it references B, return the set of attrs by which A references B.

relation-attrs-with-constraint clj/s

(relation-attrs-with-constraint db ent-name constraint)

Given an ent name, return all relation attributes which include the constraint.

relation-graph clj/s

(relation-graph schema)

A graph of the type dependencies in a schema. If entities of type :project reference an entity of type :user via :owner-id, then this will return a graph where the :project node connects to the :user node

required-attrs clj/s

(required-attrs {:keys [schema]})

Returns a map of {:ent-type #{:required-attr-1 :required-attr-2}}. Used to handle cycles.

reset-relations clj/s

(reset-relations db {:keys [ent-name visit-val]})

Generated data generates values agnostic of any schema constraints that may be present. This function updates values in the generated data to match up with constraints. First, it will remove any dummy ID's for a :coll relation. Next, it will remove any dummy ID's generated for an :omit relation. The updated ent-data map will be returned.

sort-by-required clj/s

(sort-by-required db)

sort-ents clj/s

(sort-ents db)

Attempts to topsort ents. If that's not possible (cycles present), uses sort-by-required to resolve ordering of ents in cycle

throw-invalid-spec clj/s

(throw-invalid-spec arg-name spec data)

topsort-ents clj/s

(topsort-ents {:keys [data]})

uniq-relation-attr? clj/s

(uniq-relation-attr? db ent relation-attr)

Given a db, ent, and relation-attr, determines whether the relation is a uniq attr.

validate-related-ents-query clj/s

(validate-related-ents-query {:keys [schema data] :as db}
                             ent-name
                             relation-attr
                             query-term)

Check that the refs value supplied in a query is a collection if the relation type is collection, or a keyword if the relation type is unary. If the reference is omit, no further validation is required.

view clj

(view {:keys [data]})

View with loom

visit-ents clj/s

(visit-ents db visit-key visit-fns)

(visit-ents db visit-key visit-fns ents)

Perform visit-fns on ents, storing return value as a graph attribute under visit-key

visit-ents-once clj/s

(visit-ents-once db visit-key visit-fns)

(visit-ents-once db visit-key visit-fns ents)

Like visit-ents but doesn't call visit-fn if the ent already has a visit-key attribute

visit-fn-data clj/s

(visit-fn-data db ent visit-key)

When a visit fn is called, it's passed this map as its second argument

wrap-gen-data-visiting-fn clj/s

(wrap-gen-data-visiting-fn data-generating-visiting-fn)

Useful when writing visiting fns where data generated for ent A needs to be referenced by ent B.