Data only FALSE for category

[greboul]

Hi,
Really nice package, very useful to create multiple Euler diagrams using categories with the by option.
But I have a problem since for some of my logical vectors, I have only FALSE values corresponding to one or multiple categories.
Is there a way to actually see that instead of having an error ( Error in stats::optimize(discdisc, interval = c(abs(r1 - r2), sum(r1, : 'xmin' not less than 'xmax' )? Like a box with missing categories or something similar?
Because the fact of missing TRUE values in some categories can be a very important result.
Hope it is possible, thanks !

[greboul]

Thanks for the update, I tried it but now I get this error:
"Error in pars[void, 1L] <- pars[!void, 1L]/sum(!void) : number of items to replace is not a multiple of replacement length"
Thanks !

[jolars]

Can you please provide me with a reproducible example so that I can see what's going on?

[greboul]

Dear Jolars,
You will find attached in the eulerr_example.txt file the kind of data I have and I would like to see using Euler's graphs.
The file is a csv file with column names having 25 informative rows and 8 columns (7 logical and 1 factor).
It is representing presence or absence of taxa across 7 samples at high taxonomic rank (explaining why there is redundancy in taxa column).

Here is my code throwing the error above:

fit <- euler(dataDF, by=dataDF$taxa)

What I am expecting is something like:

• Opisthokonta: S1=6,S2=2, S3=1,S4=1, S5=1,S6=1 (S3 and S4 share the same 1, which is different from the one of S5 and also of S6)
• Stramenopiles: S3=3
• Hacrobia: S2=1,S5=1, S6=1 (all 1 are samples specific -> expecting 3 shapes of same size not crossing)
• Excavata: S2=1
• Archaeplastida: S1=3,S2=3,S3=2,S5=2,S6=2,S7=1 (one of the S3 is specific of S3 as well as one of S2 and two of S1. One is shared by all.)
• Alveolata: S7=2
  eulerr_example.txt

Hope this is understandable,
greboul

[jolars]

Does this look right?

[greboul]

Seems pretty good, and I tried with my complete data and seems to work well.
Thanks a lot for your quick and great work !