Print Option: Print columns that fit in single console without wrapping

[TysonStanley]

Closes #1497

This PR addresses @MichaelChirico's number 11 and, somewhat, number 16 on #1523 by providing a trunc.cols argument (with corresponding default option("datatable.print.trunc.cols")). For example:

options("width" = 40)
DT <- data.table(thing_11 = vector("integer", 3),
                 thing_21 = vector("complex", 3),
                 thing_31 = as.IDate(paste0("2016-02-0", 1:3)),
                 thing_41 = "aasdfasdfasdfasdfasdfasdfasdfasdfasdfasdf",
                 thing_51 = vector("integer", 3),
                 thing_61 = vector("complex", 3))
print(DT, trunc.cols=TRUE)
#>    thing_11 thing_21   thing_31
#> 1:        0     0+0i 2016-02-01
#> 2:        0     0+0i 2016-02-02
#> 3:        0     0+0i 2016-02-03
#> Variables not shown: thing_41, thing_51, thing_61.

Some possible ideas for adjustments:

• Include variable type in the "Variables shown..." message
• Instead of going in the order of the columns, we could go with the data.table usual of showing the first and last columns.

[codecov]

Codecov Report

Merging #4074 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff            @@
##           master   #4074      +/-   ##
=========================================
+ Coverage    99.4%   99.4%   +<.01%     
=========================================
  Files          72      72              
  Lines       13675   13710      +35     
=========================================
+ Hits        13594   13629      +35     
  Misses         81      81

Impacted Files	Coverage Δ
R/print.data.table.R	100% <100%> (ø)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 6f17d18...020d56f. Read the comment docs.

[jangorecki]

what about some unit tests?

[TysonStanley]

Added unit tests, fixed an issue that appeared with one row data tables when trunc.cols = TRUE, and updated the example with the suggestions from @jangorecki.

Should I add info to NEWS.md?

Also, regarding the classes, I was thinking we could add the classes to the "Variables not shown" message when class = TRUE so it shows up in the table and the message. Would it be worth going forward with this?

[jangorecki]

@MichaelChirico could you take a look? you are more into printing options features.

[TysonStanley]

The failed test looks like it just needs to be re-run (it looks like a LaTeX error).

I added the classes to the message about not printed columns when class=TRUE, added a message to NEWS about the changes, and updated the tests. @MichaelChirico do you have anything else you’d like me to do for this PR?

[MichaelChirico]

@TysonStanley please see the inline comments I added, still a few things to handle. Thanks!

[TysonStanley]

@MichaelChirico I’m not seeing the inline comments showing up. Am I missing it? Also, I like the updated NEWS.

[MichaelChirico]

@TysonStanley do you see this and several others on the Files Changed tab?

https://github.com/Rdatatable/data.table/pull/4074/files

[TysonStanley]

@MichaelChirico no, I looked there and couldn’t see any... All I see are the changes to the files (no comments at all).

[MichaelChirico]

this will print columns 1, 2, ..., n where n is the last column before crossing options('width') right?

pandas is doing a bracketing approach (columns (1, N), (2, N-1), ..., (n, N-n)) where the (n+1, N-n-1) pair would break the options('width') threshold.

I'm not married to one way or the other, just worth considering which we think is the best fit. Are users likely to prefer seeing the first few columns, or seeing the first + last few? I'm not sure

[TysonStanley]

Yes, you are correct. My personal preference is the way I've done it but that, by no means, is a good reason. What do you suggest? Would having another option be the fix here?

[MichaelChirico]

I think we should pick one and stick with it rather than add more options. We can play around with it in dev for a while & maybe change later.

Just some more thoughts: I guess most common is to have the first few columns with keys/"ID"-style rows. Then have "data" in the "rightmost" columns. So the tradeoff is -- showing all the key columns, vs showing some keys, some data. Still don't have a feel yet for which is more helpful to see in a preview.

[TysonStanley]

That makes sense. I like the idea of differentiating between the keys and "data" in the preview. Definitely could get to some weird edge cases with this, depending on the console width and the number of keys...

[MichaelChirico]

@TysonStanley aaahhh sorry. I forgot to hit submit on my review. Hence the Pending in my screenshot. Should be visible now.

[TysonStanley]

The "only" is in the wrong place in that commit message. It should say: "prints only the message when console too small to fit the first column". E.g.,

options("width" = 10)
DT = data.table(a = "aaaaaaaaaaaaa",
                b = "bbbbbbbbbbbbb",
                c = "ccccccccccccc",
                d = "ddddddddddddd")
print(DT, trunc.cols=TRUE)
#> 4 variables not shown: a <char>, b <char>, c <char>, d <char> 

[TysonStanley]

This one adds the brackets to the tests. I like how those brackets look.

[TysonStanley]

@MichaelChirico Alright, checks look good so I won't make any other changes unless you ask me to.

[MichaelChirico]

Looks good to me! Thanks for your diligence!!

[mattdowle]

Excellent. @TysonStanley I've invited you to be project member. You'll need to accept the invitation displayed in your GitHub profile or GitHub repositories page. Then you can create branches directly in the main project. Welcome!
I'll merge as-is. Then I'll move the news item up and bump the test numbers in a follow-up commit. (Quicker for me that way than bringing the fork up to date.)

[jangorecki]

@TysonStanley There should be better validation of argument values provided, for example following error message is very confusing.

> print(as.data.table(iris), trunc.cols=5L)
     Sepal.Length Sepal.Width Petal.Length Petal.Width   Species
            <num>       <num>        <num>       <num>    <fctr>
  1:          5.1         3.5          1.4         0.2    setosa
  2:          4.9         3.0          1.4         0.2    setosa
  3:          4.7         3.2          1.3         0.2    setosa
  4:          4.6         3.1          1.5         0.2    setosa
  5:          5.0         3.6          1.4         0.2    setosa
 ---                                                            
146:          6.7         3.0          5.2         2.3 virginica
147:          6.3         2.5          5.0         1.9 virginica
148:          6.5         3.0          5.2         2.0 virginica
149:          6.2         3.4          5.4         2.3 virginica
150:          5.9         3.0          5.1         1.8 virginica
Error in print.data.table(as.data.table(iris), trunc.cols = 5L) : 
  object 'not_printed' not found

[TysonStanley]

Just made a PR for this #4766 that adds a quick check for a logical trunc.cols argument at the top of print.data.table().