Package 'mcparallelDo'

mcparallelDo-package placeholder

Description

A repository for a variety of useful functions.

Details

The primary function of this package is mcparallelDo(). To use mcparallelDo(), simply invoke the function with a curly braced wrapped code and the character element name to which you want to assign the results.

---

checkIfJobStillRunning

Description

checkIfJobStillRunning

Usage

checkIfJobStillRunning(targetJob, targetValue, verbose, targetEnvironment)

Arguments

targetJob	(character) The job name
targetValue	(character) The return variable name
verbose	(logical) Whether a message will be generated when complete
targetEnvironment	(environment) Target environment

Value

logical; TRUE if still running; FALSE if not running

---

jobCompleteDestructingHandler

Description

Creates a callback handler function that can be added via addTaskCallback(). These functions run at the end of each completed R statement. This particular handler watches for the completion of the target job, which is created via mcparallel()

Usage

jobCompleteSelfDestructingHandler(targetJob, targetValue, verbose,
  targetEnvironment)

Arguments

targetJob	(character) Name of the mcparallel job variable that is waiting for a result
targetValue	A character element indicating the variable that the result of that job should be assigned to targetEnvironment
verbose	A boolean element; if TRUE the completion of the fork expr will be accompanied by a message
targetEnvironment	The environment in which you want targetValue to be created

Value

callback handler function

---

mcparallelDo

Description

This function creates a fork, sets the variable named targetValue in the targetEnvironment to NULL, evaluates a segment of code evaluated in the fork, and the result of the fork returned in a variable named targetValue in the targetEnvironment after the next top-level command completes. If there is an error in the code, the returned variable will be a try-error. These effects are accomplished via the automatic creation and destruction of a taskCallback and other functions inside the mcparallelDoManager. If job results have to be collected before you return to the top level, use mcparallelDoCheck.

Usage

mcparallelDo(code, targetValue, verbose = TRUE,
  targetEnvironment = .GlobalEnv)

Arguments

code	The code to evaluate within a fork wrapped in
targetValue	A character element indicating the variable that the result of that job should be assigned to targetEnvironment
verbose	A boolean element; if TRUE the completion of the fork expr will be accompanied by a message
targetEnvironment	The environment in which you want targetValue to be created

Value

The variable name of the job, this can be manually collected via mccollect or, if on Windows, an empty string

Examples

## Create data
data(ToothGrowth)
## Trigger mcparallelDo to perform analysis on a fork
mcparallelDo({glm(len ~ supp * dose, data=ToothGrowth)},"interactionPredictorModel")
## Do other things
binaryPredictorModel <- glm(len ~ supp, data=ToothGrowth)
gaussianPredictorModel <- glm(len ~ dose, data=ToothGrowth)
## The result from mcparallelDo returns in your targetEnvironment,
## e.g. .GlobalEnv, when it is complete with a message (by default)
summary(interactionPredictorModel)

# Example of not returning a value until we return to the top level
for (i in 1:10) {
  if (i == 1) {
    mcparallelDo({2+2}, targetValue = "output")
  }
  if (exists("output")) print(i)
}

# Example of getting a value without returning to the top level
for (i in 1:10) {
  if (i == 1) {
    mcparallelDo({2+2}, targetValue = "output")
  }
  mcparallelDoCheck()
  if (exists("output")) print(i)
}

---

mcparallelDoCheck

Description

Forces a check on all mcparallelDo() jobs and returns their values to the target environment if they are complete.

Usage

mcparallelDoCheck()

Value

A named logical vector, TRUE if complete, FALSE if not complete, and an empty logical vector if on Windows

---

The mcparallelDoManager Class and Object

Description

The mcparallelDoManager Class and Object

Usage

mcparallelDoManagerClass

Format

Class 'R6ClassGenerator' <mcparallelDoManager> object generator
  Public:
    h: list
    runningJobs: list
    addJob: function (jobName, targetValue, verbose, targetEnvironment) 
    removeJob: function (x) 
    checkJobs: function () 
    clone: function (deep = FALSE) 
  Parent env: <environment: namespace:mcparallelDo>
  Locked objects: TRUE
  Locked class: FALSE
  Portable: TRUE 
 - attr(*, "name")= chr "mcparallelDoManager_generator"

---