All data.frame print methods are wrong, but some are useful
Inside you are two R users:
- One seeks meaning and truth in the numbers
- One seeks correctness and reproducibility in the code
The problem with all print() methods is that none can satisfy both
users.
Methods like paint::paint(), pillar::glimpse(), and str() are
pitched at the developer. They aim to succinctly display types, and
convey a feel for the data with examples.
The default print.data.frame() aims to give the analyst all the
numbers, and even does an okay job of it, provided you have a ultra-wide
screen monitor, and less than a couple of hundred rows.
Others try to compromise between the two modalities, and they are generally the worst to use. They’ll consider the volume of data and space available, and concoct methods of truncation. Not rows, nor columns, nor column names, nor decimal places are held sacred. Using these involves a lot of retracing your steps fiddling with print arguments or setting options. It’s enough to make you want to flip tables.
You need multiple ways of printing data to suit whatever kind of user you happen to be today, and you need them arrayed easily at your fingertips to keep you in the zone. This package places the way you view your data in your session fully within your control.
remotes::install_github("MilesMcBain/flippingtables")The setup is:
-
Create a configuration (probably in .Rprofile) with
register_flips() -
Enable the config with
flip_on()(possibly in .Rprofile)
- may seem redundant but thank me when you need to
flip_off()custom print methods because they balk at exotic text encodings or some such.
- Use
flip()to cycle between configured print methods for configured classes.
- Probably via a keyboard shortcut
Here’s an example config:
library(flippingtables)
register_flips(
printer_fns = list(
paint::paint, # a pretty good option if I do say so myself.
function(x) default_print(x, .args = list(print_arg(c("n", "max"), 100))), # a long format, uses .args see help(default_print)
function(x) withr::with_options(list(width = 300), default_print(x)) # a wide format
),
printed_classes = list(
print_override(class = "tbl", pkg_namespace = "pillar"),
print_override(class = "data.frame", pkg_namespace = "base"),
print_override(class = "data.table", pkg_namespace = "data.table")
)
)
#> [1] TRUE
flip_on() # now it's live!First in printer_fns we declare the print methods we want be able to
toggle between for use as the automatic print() in our R consolse
sessions. Using anonymous functions is a great way to prototype custom
print methods. The special function print_default() stands in for
whatever the default print for the class being printed would normally
be.
Then we nominate tbl, data.frame, and data.table as ‘flippable’
classes. We have to nominate the ‘top-level’ class that has the print
method we want to override. I.e. even though tbl (a generic tibble
used by {pillar}) and data.table are also data.frame we can’t just
configure data.frame because objects of those classes have the "tbl"
or "data.table" appear earlier in their vector of classes. This means
their corresponding print() methods dispatched instead of
print.data.frame (or the custom print method we routed that too).
Calling flippingtables::flip() will advance the binding for the
current print method to be used for all configured classes to the next
one in the list. Print methods are cycled through in the order they are
configured, and of course the cycle wraps around so cycling can happen
endlessly.
If the last result (.Last.value) has a class that is configured for
flipping, then the object is automatically re-printed with the new print
method selected by flip().
Here’s how it might look:
# assuming config above
library(palmerpenguins)
penguins
# tibble [344, 8]
# species fct Adelie Adelie Adelie Adelie Adelie Ad~
# island fct Torgersen Torgersen Torgersen Torgers~
# bill_length_mm dbl 39.1 39.5 40.3 NA 36.7 39.3
# bill_depth_mm dbl 18.7 17.4 18 NA 19.3 20.6
# flipper_length_mm int 181 186 195 NA 193 190
# body_mass_g int 3750 3800 3250 NA 3450 3650
# sex fct male female female NA female male
# year int 2007 2007 2007 2007 2007 2007
flip()
# A tibble: 344 × 8
# species island bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
# <fct> <fct> <dbl> <dbl> <int> <int>
# 1 Adelie Torgersen 39.1 18.7 181 3750
# 2 Adelie Torgersen 39.5 17.4 186 3800
# 3 Adelie Torgersen 40.3 18 195 3250
# 4 Adelie Torgersen NA NA NA NA
# 5 Adelie Torgersen 36.7 19.3 193 3450
# .. Output continues ..
# 95 Adelie Dream 36.2 17.3 187 3300
# 96 Adelie Dream 40.8 18.9 208 4300
# 97 Adelie Dream 38.1 18.6 190 3700
# 98 Adelie Dream 40.3 18.5 196 4350
# 99 Adelie Dream 33.1 16.1 178 2900
# 100 Adelie Dream 43.2 18.5 192 4100
# # ℹ 244 more rows
# # ℹ 2 more variables: sex <fct>, year <int>
# # ℹ Use `print(n = ...)` to see more rows
# it's sticky!
penguins
# A tibble: 344 × 8
# species island bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
# <fct> <fct> <dbl> <dbl> <int> <int>
# 1 Adelie Torgersen 39.1 18.7 181 3750
# 2 Adelie Torgersen 39.5 17.4 186 3800
# 3 Adelie Torgersen 40.3 18 195 3250
# 4 Adelie Torgersen NA NA NA NA
# 5 Adelie Torgersen 36.7 19.3 193 3450
# .. Output continues ..
# 95 Adelie Dream 36.2 17.3 187 3300
# 96 Adelie Dream 40.8 18.9 208 4300
# 97 Adelie Dream 38.1 18.6 190 3700
# 98 Adelie Dream 40.3 18.5 196 4350
# 99 Adelie Dream 33.1 16.1 178 2900
# 100 Adelie Dream 43.2 18.5 192 4100
# # ℹ 244 more rows
# # ℹ 2 more variables: sex <fct>, year <int>
# # ℹ Use `print(n = ...)` to see more rows
flip()
# # A tibble: 344 × 8
# species island bill_length_mm bill_depth_mm flipper_length_mm body_mass_g sex year
# <fct> <fct> <dbl> <dbl> <int> <int> <fct> <int>
# 1 Adelie Torgersen 39.1 18.7 181 3750 male 2007
# 2 Adelie Torgersen 39.5 17.4 186 3800 female 2007
# 3 Adelie Torgersen 40.3 18 195 3250 female 2007
# 4 Adelie Torgersen NA NA NA NA NA 2007
# 5 Adelie Torgersen 36.7 19.3 193 3450 female 2007
# 6 Adelie Torgersen 39.3 20.6 190 3650 male 2007
# 7 Adelie Torgersen 38.9 17.8 181 3625 female 2007
# 8 Adelie Torgersen 39.2 19.6 195 4675 male 2007
# 9 Adelie Torgersen 34.1 18.1 193 3475 NA 2007
# 10 Adelie Torgersen 42 20.2 190 4250 NA 2007
# # ℹ 334 more rows
# # ℹ Use `print(n = ...)` to see more rowsIn VSCode:
{
"description": "flip() between print methods",
"key": "<CHOOSE A KEYBINDING>",
"command": "r.runCommand",
"when": "editorTextFocus",
"args": "flippingtables::flip()"
}In RStudio:
flip()is exposed as an RStudio addin, which can be bound to a keyboard shortcut. See instructions
In other editor:
- I think you should mostly be fine since it doesn’t depend on any editor state, but let me know if I can expose anything that makes it easier.
On of my personal favourites for looking at sumarised data analytically
(provided it is sorted sensibly) is knitr::kable(). It has the added
benefit that you can easily copy-paste the output to make nice tables in
applications that speak markdown. It presents two problems for use
directly:
- it doesn’t return the data it printed invisibly, unlike most methods
intended to be used for this purpose. This will cause printing the
.Last.valuewith flip() to stop as soon as this method is hit. - it returns an object of a different class which has its own print method. R console seems not to like the idea of running through the print dispatch again, so we get nothing.
We work around this like so:
library(palmerpenguins)
library(flippingtables)
register_flips(
printer_fns = list(
paint = paint::paint, # a pretty good option if I do say so myself.
kable = function(x) {
print(knitr::kable(x))
invisible(x)
}
),
printed_classes = list(
print_override(class = "tbl", pkg_namespace = "pillar"),
print_override(class = "data.frame", pkg_namespace = "base"),
print_override(class = "data.table", pkg_namespace = "data.table")
)
)
flip_on() # now it's live!
flip()
penguins
# Flipped print method to kable
# |species |island | bill_length_mm| bill_depth_mm| flipper_length_mm| body_mass_g|sex | year|
# |:---------|:---------|--------------:|-------------:|-----------------:|-----------:|:------|----:|
# |Adelie |Torgersen | 39.1| 18.7| 181| 3750|male | 2007|
# |Adelie |Torgersen | 39.5| 17.4| 186| 3800|female | 2007|
# |Adelie |Torgersen | 40.3| 18.0| 195| 3250|female | 2007|
# .. Output continues ..