blapply: apply a function over a list or bundle

In nrauscher/corpus: Toolkit for Corpus Analysis

Description

Very similar to lapply, but applicable to bundle-objects, in particular. The purpose of the method is to supply a uniform und convenient parallel backend for the polmineR package. In particular, progress bars are supported (the naming of the method is derived from bla bla).

Usage

blapply(x, ...)

## S4 method for signature 'list'
blapply(x, f, mc = TRUE, progress = TRUE,
  verbose = FALSE, ...)

## S4 method for signature 'vector'
blapply(x, f, mc = FALSE, progress = TRUE,
  verbose = FALSE, ...)

## S4 method for signature 'bundle'
blapply(x, f, mc = FALSE, progress = TRUE,
  verbose = FALSE, ...)

Arguments

x	a list or a bundle object
...	further parameters
f	a function that can be applied to each object contained in the bundle, note that it should swallow the parameters mc, verbose and progress (use ... to catch these params )
mc	logical, whether to use multicore - if TRUE, the number of cores will be taken from the polmineR-options
progress	logical, whether to display progress bar
verbose	logical, whether to print intermediate messages

Details

Parallel backend supported so far are the parallel package (mclapply), and doMC, doParallel and doSNOW in combination with foreach. The parallel backend to be used is taken from the option 'polmineR.backend' (getOption("polmineR.backend")), the number of cores from the option 'polmineR.cores' (getOption("polmineR.cores")).

Examples

## Not run: 
  use(polmineR.sampleCorpus)
  bt <- partition("PLPRBTTXT", list(text_id=".*"), regex=TRUE)
  speeches <- as.speeches(bt, sAttributeDates="text_date", sAttributeNames="text_name")
  foo <- blapply(speeches, function(x, ...) slot(x, "cpos"))

## End(Not run)

nrauscher/corpus documentation built on May 23, 2019, 9:34 p.m.