datahike-csv-loader.core — io.replikativ/datahike-csv-loader 0.1.59

load-csv clj

(load-csv csv-file)

(load-csv csv-file cfg)

(load-csv csv-file cfg {:keys [schema ref-map tuple-map composite-tuple-map]})

Reads, parses, and loads data from CSV file named csv-file into the Datahike database having (optionally specified) config cfg, with likewise optional schema-related options for the corresponding attributes. Each column represents an attribute, with keywordized column name as attribute ident, or otherwise, an element in a heterogeneous or homogeneous tuple.

If cfg is omitted, and the last argument:

1. is also absent, or has empty :schema, :ref-map, and :composite-tuple-map, cfg is inferred to be {:schema-flexibility :read}.
2. has a non-empty value for one or more of :schema, :ref-map, and :composite-tuple-map, cfg is inferred to be {}, i.e. the default value.

:schema in the last argument can be specified in two ways:

1. Full specification via the usual Datahike transaction data format, i.e. a vector of maps, each corresponding to an attribute.
2. Partial specification via an abridged format like the map returned by datahike.api/reverse-schema, albeit with slightly different keys, each having a set of attribute idents as the corresponding value. Available options:

Key	Description
:unique-id	:db/unique value :db.unique/identity
:unique-val	:db/unique value :db.unique/value
:index	:db/index value true
:cardinality-many	:db/cardinality value :db.cardinality/many

Ref- and tuple-valued attributes, i.e. those with :db/valueType :db.type/ref or :db.type/tuple, are however specified separately, via :ref-map, :tuple-map, or :composite-tuple-map, each a map as follows:

Key	Description
:ref-map	:db.type/ref attribute idents to referenced attribute idents
:composite-tuple-map	Composite :db.type/tuple attribute idents to constituent attribute idents
:tuple-map	Other (homogeneous, heterogeneous) :db.type/tuple attribute idents to constituent attribute idents

Unspecified schema attribute values are defaults or inferred from the data given.

Example invocations:

(load-csv csv-file)
(load-csv csv-file dh-cfg)
(load-csv csv-file dh-cfg {:schema [{:db/ident :name
                                     ...}
                                    ...]
                           :ref-map {...}
                           :tuple-map {...}
                           :composite-tuple-map {...}})
(load-csv csv-file dh-cfg {:schema {:unique-id #{...}
                                    ...}
                           :ref-map {...}
                           :tuple-map {...}
                           :composite-tuple-map {...}})

Please see README for more detail.