AnalysisPageServer (version 1.6.2)

uniquify.ids.in.svg.files: uniquify.ids.in.svg.files

Description

Uniquify IDs in a set of SVG filename

Usage

uniquify.ids.in.svg.files(svg.filenames, suffixes = unique.words(length(svg.filenames)), prefixes = .default.uniquify.ids.prefixes)

Arguments

svg.filenames

Paths to SVG files

suffixes

Charvec. Suffixes to add to IDs, corresponding to svg.filenames. These is added after the word "glyph" and before the next character. An underscore character is added on both sides, too, to separate your suffix visually from the word "glyph" and the numbers after it. Ignored if new.glyph.word is provided.

prefixes

Named list. The names are the tokens that need to be replaced, such as "glyph" and "path". The values are charvec of prefixes. Only when those words appear after one of their prefixes is it substituted. Default is taken from AnalysisPageServer:::.default.uniquify.ids.prefixes.

Value

Nothing, modifies SVG file in place.

Details

The SVG files made by R use identifiers like "glyph1-3", "glyph1-4" etc. In particular these are used to define paths for different characters in order to render text. Also there are "clip1", "clip2" etc which represent clip paths, which I think limits the viewable area in a layer of a plot, but whatever it is looks awful if it goes wrong.

If multiple SVG files are embedded in the same page then this is a big problem because they will all share the same namespace and may grab the paths defined in a different file.

This function will process a set of SVG files replacing each word "glyph" with a file-specfic suffix like "glyph_123_", and each word "clip" with "clip_123_" You can provide the suffixes explicitly or let this function generate some random words, one for each file.

This function does search-and-replace with these two cleverness-es:

1. It uses C++ so it is faster (I hope) than calling gsub.
2. It checks the context of the words "glyph" and "clip", so if you had an SVG containing that word other than identifier it should be preserved. This is not 100% bulletproof since it doesn't actually parse the SVG file but it should be 99.99% bulletproof, unless you go out of your way to break it.

Examples

svg.filenames <- sapply(1:2, function(i)  {
  fn <- tempfile(fileext = ".svg")
  svg(fn)
  plot(1:10, main = paste("Plot", i), col = i)
  dev.off()
  fn
})
grep("glyph", readLines(svg.filenames[1]), value = TRUE)
uniquify.ids.in.svg.files(svg.filenames)
grep("glyph", readLines(svg.filenames[1]), value = TRUE)