Skip to content

Source reference

Choose a source by read semantics, then reference the configured environment ID from the business model. Connection endpoints, credentials, topics, queues, and buckets remain environment-owned.

database

Configured RDBMS client reading a physical table or SQL selector.

Canonical URI: database://<identifier>

Supported XML consumers: generate, iterate, variable

Read forms: database_table, database_query

Read forms by XML consumer: generate=database_table, database_query; iterate=database_table, database_query; variable=database_table, database_query

Distributions: ordered, random, round_robin

Selection capabilities: selector=yes; iterationSelector=yes; pagination=yes; random access=yes; cyclic=yes; unique=yes

Environment boundary: In Platform projects declare only <database id="sourceDB" /> and reference that configured environment id; self-managed projects configure the same client in setup or its environment profile.

Complete workflows:

file

Project-relative file selected by its declared suffix contract.

Canonical URI: file://<identifier>

Supported XML consumers: generate, iterate, variable, nestedKey

Read forms: file_csv, file_xlsx, file_json, file_ndjson, file_parquet, file_xml, file_template

Read forms by XML consumer: generate=file_csv, file_xlsx, file_json, file_ndjson, file_parquet, file_xml, file_template; iterate=file_csv, file_xlsx, file_json, file_ndjson, file_parquet, file_xml, file_template; variable=file_csv, file_xlsx, file_json, file_ndjson, file_parquet, file_xml; nestedKey=file_csv, file_json

Distributions: ordered, random, weighted, stratified, round_robin, reservoir, cumulated

Selection capabilities: selector=no; iterationSelector=no; pagination=yes; random access=yes; cyclic=yes; unique=yes

Environment boundary: Reference project-relative data with <generate source="file://data/customers.ent.csv">.

Complete workflows:

kafka

Configured Kafka consumer reading the topic declared on its importer.

Canonical URI: kafka://<identifier>

Supported XML consumers: generate, iterate

Read forms: kafka

Read forms by XML consumer: generate=kafka; iterate=kafka

Distributions: ordered

Selection capabilities: selector=no; iterationSelector=no; pagination=no; random access=no; cyclic=no; unique=no

Environment boundary: Reference the configured Kafka importer id; topic and connection details belong to its environment-backed importer configuration.

Complete workflows:

memstore

Named dataset stored by the Redis memstore exporter.

Canonical URI: memstore://<identifier>

Supported XML consumers: generate, iterate, variable, nestedKey

Read forms: redis_memstore

Read forms by XML consumer: generate=redis_memstore; iterate=redis_memstore; variable=redis_memstore; nestedKey=redis_memstore

Distributions: ordered, random, weighted, stratified, round_robin, reservoir, cumulated

Selection capabilities: selector=no; iterationSelector=no; pagination=yes; random access=yes; cyclic=yes; unique=yes

Environment boundary: Produce via memstore target, consume via memstore source with the same id.

Complete workflows:

ml_model

Persisted ML generator read through its ml:// model URI.

Canonical URI: ml://<identifier>

Supported XML consumers: generate, iterate

Read forms: ml_model

Read forms by XML consumer: generate=ml_model; iterate=ml_model

Distributions: ordered

Selection capabilities: selector=no; iterationSelector=no; pagination=yes; random access=no; cyclic=no; unique=no

Environment boundary: Train and persist a model with <ml-train>; reference it from a bounded source product as source="ml://<model-name>".

Complete workflows:

mongodb

Configured MongoDB client reading a collection or Mongo selector.

Canonical URI: mongodb://<identifier>

Supported XML consumers: generate, iterate, variable

Read forms: database_table, database_query

Read forms by XML consumer: generate=database_table, database_query; iterate=database_table, database_query; variable=database_table, database_query

Distributions: ordered, random, round_robin

Selection capabilities: selector=yes; iterationSelector=yes; pagination=yes; random access=yes; cyclic=yes; unique=yes

Environment boundary: In Platform projects declare only <mongodb id="sourceMongo" /> and reference that configured environment id; self-managed projects configure the same client in setup or its environment profile.

Complete workflows:

object_storage

Configured object-storage client reading one object by its source URI.

Canonical URI: object-storage://<identifier>

Supported XML consumers: generate, iterate

Read forms: object_storage

Read forms by XML consumer: generate=object_storage; iterate=object_storage

Distributions: ordered, random, weighted, stratified, round_robin, reservoir, cumulated

Selection capabilities: selector=no; iterationSelector=no; pagination=yes; random access=yes; cyclic=yes; unique=yes

Environment boundary: In Platform projects declare only <object-storage id="archive" /> and reference that configured environment id; self-managed projects configure the same client in setup or its environment profile.

Complete workflows:

rabbitmq

Configured RabbitMQ consumer reading currently available messages from its declared queue.

Canonical URI: rabbitmq://<identifier>

Supported XML consumers: generate, iterate

Read forms: rabbitmq

Read forms by XML consumer: generate=rabbitmq; iterate=rabbitmq

Distributions: ordered, round_robin

Selection capabilities: selector=no; iterationSelector=no; pagination=no; random access=no; cyclic=no; unique=no

Environment boundary: Reference the configured RabbitMQ importer id; queue topology and connection details belong to its environment-backed importer configuration.

Complete workflows: