In order to access my server, I am forced to work with an old text terminal application that does not have X windows. The best thing I have going is emacs/ESS.
Often, I wish to make a rudimentary plots such as histograms and scatter plots and not have to go through the trouble of transferring the file to a computer with a graphics display.
Is there a text terminal-based R graphics library?
There are a number of things that can do bits of it. There's stem in default R, there's this scatter plot function, but best of all, there's the package txtplot on CRAN which does scatterplots, boxplots, barplots, density traces, acfs and plots curves (like the curve function... kinda).
I only need it once in a while - but if I am trying to convey a rough idea of a graphic in pure text as I sometimes need to, it's a life-saver.
In the past I wrote a short piece of R code that made tally-style ascii graphic in very quick time (like a sideways barchart or a stem-and-leaf plot with the numbers replaced by symbols, which solved a problem I had) - but I didn't keep it since stem mostly covers that territory.
Of course the 'table' facility produces ascii output and can be manipulated to do some interesting/useful semigraphical things.
There's also the package ascii that can be used to render various R objects into ascii form in similar fashion to Sweave - handy for formatting tables and so on. Just formatting a table into ascii is not really what it's for but you might still be able to get some use out of it with a little work and the right output format.
Sample output from txtplot:
scatter plot:
> with(cars,txtplot(speed,dist))
+----+-----------+------------+-----------+-----------+--+
120 + * +
| |
100 + +
| * * |
80 + * * +
| * * * |
60 + * * +
| * * * * * |
40 + * * * * * +
| * * * * * * * |
20 + * * * * * * * +
| * * * * |
| * * * |
0 +----+-----------+------------+-----------+-----------+--+
5 10 15 20 25
acf plot:
> txtacf(ldeaths)
+-+--------------+--------------+--------------+--------+
1 + * +
| * |
| * * * * * |
0.5 + * * * * * +
| * * * * * * * * |
| * * * * * * * * |
| * * * * * * * * |
0 + * * * * * * * * * * * * * * * * * * * * * +
| * * * * * * * * * * |
| * * * * * * * * * * |
| * * * * * * * * * |
-0.5 + * * * * * * +
| * * * * |
+-+--------------+--------------+--------------+--------+
0 0.5 1 1.5
density trace:
> txtdensity(rnorm(100,m=5,s=.1))
+------+----------+----------+----------+----------+-------+
| ***** |
4 + ** *** +
| * *** |
| ** *** |
3 + ** *** +
| *** ** |
| ***** ** |
2 + *** ** +
| *** ** |
| ** ** |
1 + ** *** +
| *** ****** |
| ******** *** |
+------+----------+----------+----------+----------+-------+
4.8 4.9 5 5.1 5.2
box plot:
> vc <- ToothGrowth[,2]=="VC"
> oj <- ToothGrowth[,2]=="OJ"
> txtboxplot(ToothGrowth[vc,1],ToothGrowth[oj,1])
5 10 15 20 25 30 35
|----+-------+--------+--------+--------+--------+-------+--|
+--------+-----------+
1 -------------| | |------------------
+--------+-----------+
+------------+----+
2 -------------| | |---------
+------------+----+
Legend: 1=ToothGrowth[vc, 1], 2=ToothGrowth[oj, 1]
curve plot:
> txtcurve(sin(pi*x),from=0,to=2)
+--+-----------+------------+------------+-----------+--+
1 + ********* +
| *** ** |
| ** ** |
0.5 + ** ** +
| ** ** |
| * ** |
0 + * ** * +
| * * |
| ** ** |
-0.5 + *** ** +
| ** ** |
| ** *** |
-1 + ********* +
+--+-----------+------------+------------+-----------+--+
0 0.5 1 1.5 2
bar chart:
> txtbarchart(as.factor(res),pch="|")
+--+------------+------------+------------+------------+--+
50 + | +
| | |
40 + | +
| | |
30 + | | +
| | | |
| | | |
20 + | | | +
| | | | |
10 + | | | +
| | | | |
0 + | | | +
+--+------------+------------+------------+------------+--+
1 1.5 2 2.5 3
Legend: 1=A, 2=B, 3=C
Add in the stem function from default R graphics:
> stem(log(islands,10))
The decimal point is at the |
1 | 1111112222233444
1 | 5555556666667899999
2 | 3344
2 | 59
3 |
3 | 5678
4 | 012
and you have quite a lot of coverage.
