Ropj (version 0.3-6)

read.opj: Parse Origin(R) project file into a list of objects

Description

This function parses an OPJ file into a list of objects it consists of. Items understood by read.opj include spreadsheets, matrices, and notes.

Usage

read.opj(file, encoding = 'latin1', tree = FALSE, ...)

Value

A named list containing objects stored in the file.

• Spreadsheets are presented as data.frames, with additional attributes:

  • comment contains the fields Long name, Units, and Comment, joined by \r\n, if present. Due to a possible bug in liborigin, these values are sometimes followed by @ and some text that wasn't present in the original document. (In versions prior to v0.2-2 it was called comments, which should be still supported until v1.0.)

  • commands contains the formula that was used to create the values of the column (e.g. col(A) * 2 + 1).

• Multi-sheet spreadsheets are stored as named lists of data.frames described above.

• Matrices are presented as lists of matrix objects containing numeric data. dimnames are also assigned. The list also has attributes:

  • commands contains the formula that was used to compute the values in the matrix.

• Notes are stored as plain strings.

When tree = FALSE, the list is flat, its names are short names of the objects, and the comment attribute of the list contains the long names of the objects stored in the file.

When tree = TRUE, the list names are long names (if present; short otherwise) and the list itself represents the folder structure of the project.

Arguments

file

Path to the OPJ file to parse. Only file paths are supported, not R connections. Path is not expanded automatically.

The path must be representable in the native encoding of the R session. Unfortunately, a Unicode path that doesn't fit in the ANSI encoding will not work on Windows older than version 10, November 2019.

encoding

Encoding of the strings inside the file being opened. This should correspond to the ANSI code page of Windows installation used to produce the file. The default of 'latin1' is usually safe to use. See iconvlist() for the full list of encodings supported by your R installation.

tree

Control the structure of the returned list. When FALSE (default), returned list is flat and its unique names correspond to the short names of the objects, while the comment attribute contains their long names and comments. When TRUE, the list itself becomes a recursive data structure containing the tree of the objects, making it possible to access objects by their paths in that tree.

...

The rest of the arguments is passed to as.data.frame when converting spreadsheets from lists of columns, making it possible to set stringsAsFactors and other parameters as needed.

Examples

  x <- read.opj(system.file('test.opj', package = 'Ropj'))
  head(x$Book2, 7)
  comment(x$Book2)
  attr(x$Book2, 'commands')
  with(x$Book1, head(Sheet2 - Sheet1))
  x$MBook1$MSheet1[1:4,1:4]
  x$Note1

  if ('CP1251' %in% iconvlist()) {
    # encoding names aren't guaranteed to be supported across all platforms
    x <- read.opj(system.file('test.opj', package = 'Ropj'), 'CP1251')
    print(x$cyrillic)
  }

  str(read.opj(system.file('tree.opj', package = 'Ropj'), tree = TRUE))