R/checkdir.R

In chartgerink/statcheckTEMP: Extract statistics from articles and recompute p values

checkdir <- structure(function(# Extract test statistics from all HTML and PDF files in a folder.
  ### Extracts statistical references from a directory with HTML and PDF files. The "pdftotext" program is used to convert PDF files to plain text files. This must be installed and PATH variables must be properly set so that this program can be used from command line.
  ### By default a gui window is opened that allows you to choose the directory (using tcltk).
  dir,
  ### String indicating the directory to be used.
  subdir = TRUE,
  ### Logical indicating whether you also want to check subfolders. Defaults to TRUE
  ...
  ### Arguments sent to  \code{\link{statcheck}}.
  )
{
  ##details<<
  ## See \code{\link{statcheck}} for more details. This function is a wrapper around both \code{\link{checkPDFdir}} for PDF files and \code{\link{checkHTMLdir}} for HTML files.
  ## Depending on the PDF file the comparison operators can sometimes not be converted correctly, causing these to not be reported in the output. Using html versions of articles is reccomended for more stable results.
  ## Note that the conversion to plain text and extraction of statistics can result in errors. Some statistical values can be missed, especially if the notation is unconvetional. It is recommended to manually check some of the results.

  ##seealso<<
  ## \code{\link{statcheck}}, \code{\link{checkPDF}}, \code{\link{checkHTMLdir}}, \code{\link{checkHTML}}, \code{\link{checkHTMLdir}}
  if (missing(dir)) dir <- tk_choose.dir()
  
  pdfs <- any(grepl("\\.pdf$",list.files(dir,recursive=subdir),ignore.case=TRUE))
  htmls <- any(grepl("\\.html?$",list.files(dir,recursive=subdir),ignore.case=TRUE))
  
   if (pdfs) pdfres <- checkPDFdir(dir,...)
   if (htmls) htmlres <- checkHTMLdir(dir,...)
  
  if (pdfs & htmls)
  {
    Res <- rbind(pdfres,htmlres)
  } else if (pdfs & !htmls) Res <- pdfres else if (!pdfs & htmls) Res <- htmlres else stop("No PDF or HTML found")
  class(Res) <- c("statcheck","data.frame")
  return(Res)
  ### A data frame containing for each extracted statistic:
  ### \item{Source}{Name of the file of which the statistic is extracted}
  ### \item{Statistic}{Character indicating the statistic that is extracted}
  ### \item{df1}{First degree of freedom}
  ### \item{df2}{Second degree of freedom (if applicable)}
  ### \item{Value}{Reported value of the statistic}
  ### \item{Reported.Comparison}{Reported comparison, when importing from pdf this will often not be converted properly}
  ### \item{Reported.P.Value}{The reported p-value, or NA if the reported value was NS}
  ### \item{Computed}{The recomputed p-value}
  ### \item{Raw}{Raw string of the statistical reference that is extracted}
  ### \item{InExactError}{Error in inexactly reported p values as compared to the recalculated p values}
  ### \item{ExactError}{Error in exactly reported p values as compared to the recalculated p values}
  ### \item{DecisionError}{The reported result is significant whereas the recomputed result is not, or vice versa.}
},ex=function(){
# with this command a menu will pop up from which you can select the directory with articles
# checkdir()

# you could also specify the directory beforehand
# for instance:
# DIR <- "C:/mydocuments/articles"
# checkdir(DIR)
})