Create Event Based Data Architectures

Event-driven programming is a programming paradigm where the flow of execution is defined by event. In this paradigm an event can be defined as "a change in the state" of an object. This package offers a set of functions for creating event-based architectures using three basic functions: events, dispatchers, and handlers. The handlers manage the events, the dispatchers are in charge of redirecting the events to each of the handlers, finally the events are the objects that carry the information about the change of state.

eventr

The goal of eventr is to ...

Installation

You can install the released version of eventr from CRAN with:

install.packages("eventr")

# Or the development version from GitHub:
# install.packages("devtools")
devtools::install_github("alvarofranq/eventr")

Example

This is a basic example which shows you how to solve a common problem:

library(eventr)
library(dplyr)

birth_event <- event(
  id = 'first-id',
  type = 'BIRTH',
  time = '1936-11-09',
  birth_date = '1936-11-09'
)

death_event <- event(
  id = 'second-id',
  type = 'DEATH',
  time = '2019-05-22',
  death_date = '2019-05-22'
)

set_birth_date <- function(obj, event){
  obj$birth_date <- get_body_attr(event, "birth_date")
  return(obj)
}

set_death_date <- function(obj, event){
  obj$death_date <- get_body_attr(event, "death_date")
  return(obj)
}

birth_handler <- handler(type = 'BIRTH', FUN = set_birth_date)
death_handler <- handler(type = 'DEATH', FUN = set_death_date)

handlers <- handlers_list(birth_handler, death_handler)

the_dispatcher <- dispatcher(handlers)

dispatch <- get_dispatch(the_dispatcher)

events <- event_list(birth_event, death_event)

the_obj <- dispatch(events = events, accumulate = FALSE)
the_obj

the_obj <- dispatch(events = events, accumulate = TRUE)
the_obj

# transform the_obj to data.frame
the_obj %>%
  purrr::map(as.data.frame) %>%
  bind_rows