jonmagic / grim

Tool for extracting pages from pdf as images and text as strings.
MIT License
216 stars 51 forks source link
ghostscript imagemagick pdf ruby
                    ,____
                    |---.\
            ___     |    `
           / .-\  ./=)
          |  |"|_/\/|
          ;  |-;| /_|
         / \_| |/ \ |
        /      \/\( |
        |   /  |` ) |
        /   \ _/    |
       /--._/  \    |
       `/|)    |    /
         /     |   |
       .'      |   |
      /         \  |
     (_.-.__.__./  /

Grim

Grim is a simple gem for extracting (reaping) a page from a pdf and converting it to an image as well as extract the text from the page as a string. It basically gives you an easy to use api to ghostscript, imagemagick, and pdftotext specific to this use case.

Prerequisites

You will need ghostscript, imagemagick, and xpdf installed. On the Mac (OSX) I highly recommend using Homebrew to get them installed.

$ brew install ghostscript imagemagick xpdf

Installation

$ gem install grim

Usage

pdf   = Grim.reap("/path/to/pdf")         # returns Grim::Pdf instance for pdf
count = pdf.count                         # returns the number of pages in the pdf
png   = pdf[3].save('/path/to/image.png') # will return true if page was saved or false if not
text  = pdf[3].text                       # returns text as a String

pdf.each do |page|
  puts page.text
end

We also support using other processors (the default is whatever version of Imagemagick/Ghostscript is in your path).

# specifying one processor with specific ImageMagick and GhostScript paths
Grim.processor =  Grim::ImageMagickProcessor.new({:imagemagick_path => "/path/to/convert", :ghostscript_path => "/path/to/gs"})

# multiple processors with fallback if first fails, useful if you need multiple versions of convert/gs
Grim.processor = Grim::MultiProcessor.new([
  Grim::ImageMagickProcessor.new({:imagemagick_path => "/path/to/6.7/convert", :ghostscript_path => "/path/to/9.04/gs"}),
  Grim::ImageMagickProcessor.new({:imagemagick_path => "/path/to/6.6/convert", :ghostscript_path => "/path/to/9.02/gs"})
])

pdf = Grim.reap('/path/to/pdf')

You can even specify a Windows executable :zap:

# specifying another ghostscript executable, win64 in this example
# the ghostscript/bin folder still has to be in the PATH for this to work
Grim.processor =  Grim::ImageMagickProcessor.new({:ghostscript_path => "gswin64c.exe"})

pdf = Grim.reap('/path/to/pdf')

Grim::ImageMagickProcessor#save supports several options as well:

pdf = Grim.reap("/path/to/pdf")
pdf[0].save('/path/to/image.png', {
  :width => 600,         # defaults to 1024
  :density => 72,        # defaults to 300
  :quality => 60,        # defaults to 90
  :colorspace => "CMYK", # defaults to "RGB"
  :alpha => "Activate"   # not used when not set
})

Grim has limited logging abilities. The default logger is Grim::NullLogger but you can also set your own logger.

require "logger"
Grim.logger = Logger.new($stdout).tap { |logger| logger.progname = 'Grim' }
Grim.processor = Grim::ImageMagickProcessor.new({:ghostscript_path => "/path/to/bin/gs"})
pdf = Grim.reap("/path/to/pdf")
pdf[3].save('/path/to/image.png')
# D, [2016-06-09T22:43:07.046532 #69344] DEBUG -- grim: Running imagemagick command
# D, [2016-06-09T22:43:07.046626 #69344] DEBUG -- grim: PATH=/path/to/bin:/usr/local/bin:/usr/bin
# D, [2016-06-09T22:43:07.046787 #69344] DEBUG -- grim: convert -resize 1024 -antialias -render -quality 90 -colorspace RGB -interlace none -density 300 /path/to/pdf /path/to/image.png

Reference

Contributors

License

See LICENSE for details.