Skip to content

Comparing figures?

3 messages · Ramiro Barrantes, Suzen, Mehmet, Martin Maechler

Ramiro Barrantes
# 
Hello,

I am working on tests to compare figures.  I have been using ImageMagick, which creates a figure signature, and I can compare a "test" figure signature against a saved "reference" figure signature.  It seems to work pretty well.  However, it is slow as it requires reading from the file system.

Are there any options to compare figures on memory?  For example, if I generate a ggplot or lattice graph, I could have all my saved "reference" figures on memory (which I would have loaded all at once) and compare them. I just haven't found anything.

I just found out about the vdiffr package and was going to explore it, not sure about the speed.

Any suggestions appreciated.

Thank you,
<https://west.exch023.serverdata.net/owa/?ae=Item&a=New&t=IPM.Note&cc=MTQuMy4zMTkuMixlbi1VUyw2LEhUTUwsMCww&pspid=_1525698150389_875737489#>
Ramiro

Ramiro Barrantes Ph.D.
Precision Bioassay, Inc.
431 Pine St., Suite 110
Burlington, VT 05401
802 865 0155
802 861 2365 FAX
www.precisionbioassay.com<https://west.exch023.serverdata.net/owa/redir.aspx?SURL=wN3KzpoKXAcetH7sTOTnSyfg-iAXFIinpPUtRcduCFCtkgZrUSDTCGgAdAB0AHAAOgAvAC8AdwB3AHcALgBwAHIAZQBjAGkAcwBpAG8AbgBiAGkAbwBhAHMAcwBhAHkALgBjAG8AbQA.&URL=http%3a%2f%2fwww.precisionbioassay.com>
ramiro at precisionbioassay.com

CONFIDENTIALITY NOTICE: This email, including any attach...{{dropped:9}}
Suzen, Mehmet
# 
I suggest perceptual diff. You could write a wrapper around it.

http://pdiff.sourceforge.net

On Mon, 7 May 2018 16:49 Ramiro Barrantes, <ramiro at precisionbioassay.com>
wrote:

  
  


  


      

    

    

      
      

        


  

        

      Martin Maechler
    

    

      
      #
    

  


  

          

      
      
      
    

        
> I suggest perceptual diff. You could write a wrapper
    > around it.  http://pdiff.sourceforge.net

In (base) R's own test ('make check-devel'), we basically use

   pdf(*, compress=FALSE)

actually also setting 'encoding=.' and 'paper=.', see the R sources at
   https://svn.r-project.org/R/trunk/tests/reg-plot.R

and the check (source at
    https://svn.r-project.org/R/trunk/tests/Makefile.common)

basically is using

    R CMD Rdiff  <foo>.pdf  <foo>.pdf.save

(with <foo> := reg-plot in this case)

Martin Maechler
R Core Team


    > On Mon, 7 May 2018 16:49 Ramiro Barrantes,

        
> <ramiro at precisionbioassay.com> wrote:

        
>> Hello,
    >> 
    >> I am working on tests to compare figures.  I have been
    >> using ImageMagick, which creates a figure signature, and
    >> I can compare a "test" figure signature against a saved
    >> "reference" figure signature.  It seems to work pretty
    >> well.  However, it is slow as it requires reading from
    >> the file system.
    >> 
    >> Are there any options to compare figures on memory?  For
    >> example, if I generate a ggplot or lattice graph, I could
    >> have all my saved "reference" figures on memory (which I
    >> would have loaded all at once) and compare them.  I just
    >> haven't found anything.
    >> 
    >> I just found out about the vdiffr package and was going
    >> to explore it, not sure about the speed.
    >> 
    >> Any suggestions appreciated.
    >> 
    >> Thank you, <
    >> https://west.exch023.serverdata.net/owa/?ae=Item&a=New&t=IPM.Note&cc=MTQuMy4zMTkuMixlbi1VUyw2LEhUTUwsMCww&pspid=_1525698150389_875737489#
    >> >
    >> Ramiro
    >> 
    >> Ramiro Barrantes Ph.D.  Precision Bioassay, Inc.  431
    >> Pine St., Suite 110 Burlington, VT 05401 802 865 0155 802
    >> 861 2365 FAX www.precisionbioassay.com<
    >> https://west.exch023.serverdata.net/owa/redir.aspx?SURL=wN3KzpoKXAcetH7sTOTnSyfg-iAXFIinpPUtRcduCFCtkgZrUSDTCGgAdAB0AHAAOgAvAC8AdwB3AHcALgBwAHIAZQBjAGkAcwBpAG8AbgBiAGkAbwBhAHMAcwBhAHkALgBjAG8AbQA.&URL=http%3a%2f%2fwww.precisionbioassay.com
    >> >
    >> ramiro at precisionbioassay.com
    >> 
    >> CONFIDENTIALITY NOTICE: This email, including any
    >> attach...{{dropped:9}}
    >> 
    >> ______________________________________________
    >> R-help at r-project.org mailing list -- To UNSUBSCRIBE and
    >> more, see https://stat.ethz.ch/mailman/listinfo/r-help
    >> PLEASE do read the posting guide
    >> http://www.R-project.org/posting-guide.html and provide
    >> commented, minimal, self-contained, reproducible code.
    >> 

    > 	[[alternative HTML version deleted]]

    > ______________________________________________
    > R-help at r-project.org mailing list -- To UNSUBSCRIBE and
    > more, see https://stat.ethz.ch/mailman/listinfo/r-help
    > PLEASE do read the posting guide
    > http://www.R-project.org/posting-guide.html and provide
    > commented, minimal, self-contained, reproducible code.