only 10 rows of differences shown when file= argument is used

[dwoollacott]

df1 <- data.frame(
    rn = 1:100,
    lt = letters[1:20]
)

df2 <- data.frame(
    rn = 1:100,
    lt = c(letters[1:5], rev(letters[6:20]))
)

diff1 <- diffdf(
    df1,
    df2,
    file = "test.out"
)

produces in "test.out"

Differences found between the objects!

Summary of BASE and COMPARE
  ====================================
    PROPERTY      BASE        COMP    
  ------------------------------------
      Name        df1         df2     
     Class     data.frame  data.frame 
    Rows(#)       100         100     
   Columns(#)      2           2      
  ------------------------------------


Not all Values Compared Equal
  =============================
   Variable  No of Differences 
  -----------------------------
      lt            70         
  -----------------------------


First 10 of 70 rows are shown in table below
  ========================================
   VARIABLE  ..ROWNUMBER..  BASE  COMPARE 
  ----------------------------------------
      lt           6         f       t    
      lt           7         g       s    
      lt           8         h       r    
      lt           9         i       q    
      lt          10         j       p    
      lt          11         k       o    
      lt          12         l       n    
      lt          14         n       l    
      lt          15         o       k    
      lt          16         p       j    
  ----------------------------------------

[gowerc]

Hey @dwoollacott,

I think the fact it prints 10 rows by default is expected and not an issue. That being said it looks like we don't actually expose the n_row option to allow users to specify the number of rows to be printed in the file which is an issue :(

Tbh I never really liked this functionality all that much and wouldn't be against just binning it (@kieranjmartin ?)

For reference the file argument here is just a wrapper around:

x <- print(diffdf_obj, as_string = TRUE)
sink(file)
cat(x, sep = "\n")
sink()

So if you want to change the number of rows in the file then as a workaround you can currently just use:

x <- print(diffdf_obj, as_string = TRUE, row_limit = NULL) # NULL = all rows
sink(file)
cat(x, sep = "\n")
sink()

[dwoollacott]

Thanks @gowerc - the other option would be to change line 373 in diffdf.R from

        x <- print(COMPARE, as_string = TRUE)

to

        x <- print(COMPARE, row_limit=NULL, as_string = TRUE)

this is only based on my thought that if the user is writing the output to a file, they'll normally want all the differences printed so this should be the default behaviour.

[kieranjmartin]

I think the original reason we kept the limit is that this could lead to some degenerate behaviour; if you have two large datasets with lots of differences and no limits, you could end up with a truly enormous report file which may be surprising.