-
Notifications
You must be signed in to change notification settings - Fork 26
/
gt_plt_dist.R
362 lines (312 loc) · 11.4 KB
/
gt_plt_dist.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
#' Add distribution plots into rows of a `gt` table
#' @description
#' The `gt_plt_dist` function takes an existing `gt_tbl` object and
#' adds summary distribution sparklines via `ggplot2`. Note that these sparklines
#' are limited to density, histogram, boxplot or rug/strip charts. If you're
#' wanting to plot more traditional spark**lines**, you can use `gtExtras::gt_plt_sparkline()`.
#'
#' @param gt_object An existing gt table object of class `gt_tbl`
#' @param column The column wherein the sparkline plot should replace existing data. Note that the data *must* be represented as a list of numeric values ahead of time.
#' @param type A string indicating the type of plot to generate, accepts `"boxplot"`, `"histogram"`, `"rug_strip"` or `"density"`.
#' @param fig_dim A vector of two numbers indicating the height/width of the plot in mm at a DPI of 25.4, defaults to `c(5,30)`
#' @param line_color Color for the line, defaults to `"black"`. Accepts a named color (eg 'blue') or a hex color.
#' @param fill_color Color for the fill of histograms/density plots, defaults to `"grey"`. Accepts a named color (eg `'blue'`) or a hex color.
#' @param bw The bandwidth or binwidth, passed to `density()` or `ggplot2::geom_histogram()`. If `type = "density"`, then `bw` is passed to the `bw` argument, if `type = "histogram"`, then `bw` is passed to the `binwidth` argument.
#' @param trim A logical indicating whether to trim the values in `type = "density"` to a slight expansion beyond the observable range. Can help with long tails in `density` plots.
#' @param same_limit A logical indicating that the plots will use the same axis range (`TRUE`) or have individual axis ranges (`FALSE`).
#' @param type_col A tidyselect column indicating a vector of which `type` of plot to make by row. Must be equal to the total number of rows and limited to `"boxplot"`, `"histogram"`, `"rug_strip"` or `"density"`.
#' @return An object of class `gt_tbl`.
#' @export
#' @section Examples:
#' ```r
#' library(gt)
#' gt_sparkline_tab <- mtcars %>%
#' dplyr::group_by(cyl) %>%
#' # must end up with list of data for each row in the input dataframe
#' dplyr::summarize(mpg_data = list(mpg), .groups = "drop") %>%
#' gt() %>%
#' gt_plt_dist(mpg_data)
#' ```
#' @section Figures:
#' \if{html}{\figure{gt_plt_dist.png}{options: width=50\%}}
#'
#' @family Plotting
#' @section Function ID:
#' 1-4
gt_plt_dist <- function(gt_object,
column,
type = "density",
fig_dim = c(5, 30),
line_color = "black",
fill_color = "grey",
bw = NULL,
trim = FALSE,
same_limit = TRUE,
type_col = NULL
) {
is_gt_stop(gt_object)
# convert tidyeval column to bare string
col_bare <- dplyr::select(gt_object[["_data"]], {{ column }}) %>% names()
# segment data with bare string column name
list_data_in <- gt_index(gt_object, col_bare, as_vector = TRUE)
# convert to a single vector
data_in <- unlist(list_data_in)
stopifnot("Specified column must contain list of values" = any(class(list_data_in) %in% "list"))
stopifnot("Specified column must be numeric" = is.numeric(data_in))
stopifnot("You must indicate the `type` of plot as one of 'boxplot', 'histogram', 'rug_strip' or 'density'." = isTRUE(type %in% c("boxplot", "rug_strip", "histogram", "density")))
# range to be used for plotting if same axis
total_rng <- grDevices::extendrange(data_in, r = range(data_in, na.rm = TRUE), f = 0.02)
# TODO: Need to account for bw as well.
plot_fn_spark <- function(trim, list_data_in, type_in) {
if (all(list_data_in %in% c(NA, NULL))) {
return("<div></div>")
}
vals <- as.double(stats::na.omit(list_data_in))
max_val <- max(vals, na.rm = TRUE)
min_val <- min(vals, na.rm = TRUE)
x_max <- vals[vals == max_val]
x_min <- vals[vals == min_val]
input_data <- dplyr::tibble(
x = 1:length(vals),
y = vals
)
# respect type column or value
type = type_in
if (type == "boxplot") {
plot_base <- ggplot(input_data) +
theme_void()
if (isTRUE(same_limit)) {
plot_base <- plot_base +
scale_x_continuous(expand = expansion(mult = 0.05)) +
coord_cartesian(
clip = "off",
xlim = grDevices::extendrange(total_rng, f = c(0, 0.01)),
ylim = c(0.9, 1.15)
)
} else {
plot_base <- plot_base +
scale_x_continuous(expand = expansion(mult = 0.05)) +
coord_cartesian(
clip = "off",
xlim = grDevices::extendrange(vals, f = 0.09),
ylim = c(0.9, 1.15)
)
}
plot_out <- plot_base +
geom_boxplot(
aes(x = .data$y, y = 1),
width = 0.15,
color = line_color,
fill = fill_color,
outlier.size = 0.3,
linewidth = 0.3
)
} else if (type == "rug_strip") {
plot_base <- ggplot(input_data) +
theme_void()
if (isTRUE(same_limit)) {
plot_base <- plot_base +
scale_x_continuous(expand = expansion(mult = 0.05)) +
coord_cartesian(
clip = "off",
xlim = grDevices::extendrange(total_rng, f = 0.09),
ylim = c(0.75, 1.15)
)
} else {
plot_base <- plot_base +
scale_x_continuous(expand = expansion(mult = 0.05)) +
coord_cartesian(
clip = "off",
xlim = grDevices::extendrange(vals, f = 0.09),
ylim = c(0.75, 1.15)
)
}
plot_out <- plot_base +
geom_point(
aes(x = .data$y, y = 1),
alpha = 0.2,
size = 0.3,
color = line_color,
position = position_jitter(height = 0.15, seed = 37)
) +
geom_rug(
aes(x = .data$y),
length = unit(0.2, "npc"),
alpha = 0.5,
linewidth = 0.2
)
} else if (type == "histogram") {
plot_base <- ggplot(input_data) +
theme_void()
if (isTRUE(same_limit)) {
if (is.null(bw)) {
bw <- bw_calc(data_in)
} else {
bw <- bw
}
plot_out <- plot_base +
{
if(bw > 0){
geom_histogram(
aes(x = .data$y),
color = line_color, fill = fill_color, binwidth = bw,
linewidth = 0.2
)
} else if(bw == 0) {
bw <- 1
geom_histogram(
aes(x = .data$y),
color = line_color, fill = fill_color, binwidth = bw,
linewidth = 0.2
)
} else {
hist_breaks <- graphics::hist(data_in[!is.na(data_in)], breaks = "FD", plot=FALSE)$breaks
geom_histogram(
aes(x = .data$y),
color = line_color, fill = fill_color, breaks = hist_breaks,
linewidth = 0.2
)
}
} +
scale_x_continuous(expand = expansion(mult = 0.1)) +
coord_cartesian(
clip = "off",
xlim = grDevices::extendrange(
data_in,
r = range(data_in, na.rm = TRUE),
f = 0.02
)
)
} else {
if (is.null(bw)) {
bw <- 2 * stats::IQR(vals, na.rm = TRUE) / length(vals)^(1 / 3)
} else {
bw <- bw
}
plot_out <- plot_base +
geom_histogram(
aes(x = .data$y),
color = line_color,
fill = fill_color,
binwidth = bw
) +
coord_cartesian(
clip = "off",
xlim = grDevices::extendrange(
vals,
r = range(vals, na.rm = TRUE),
f = 0.02
)
)
}
} else if (type == "density") {
if (isTRUE(same_limit)) {
if (is.null(bw)) {
bw <- stats::bw.nrd0(stats::na.omit(as.vector(data_in)))
} else {
bw <- bw
}
total_rng_dens <- stats::density(
as.vector(
stats::na.omit(data_in)
),
bw = bw
)[["x"]]
density_calc <- stats::density(input_data[["y"]], bw = bw)
density_range <- density_calc[["x"]]
density_df <- dplyr::tibble(
x = density_calc[["x"]],
y = density_calc[["y"]]
)
if (trim) { # implementation of filtering values
# only to actual and slightly outside the range
filter_range <- range(vals, na.rm = TRUE) %>%
scales::expand_range(mul = 0.05)
density_df <- dplyr::filter(
density_df,
dplyr::between(.data$x, filter_range[1], filter_range[2])
)
}
plot_base <- ggplot(density_df) +
theme_void()
plot_out <- plot_base +
geom_area(
aes(x = .data$x, y = .data$y),
color = line_color,
fill = fill_color
) +
xlim(range(density_range)) +
coord_cartesian(
xlim = range(total_rng_dens, na.rm = TRUE),
expand = TRUE,
clip = "off"
)
} else {
if (is.null(bw)) {
bw <- stats::bw.nrd0(stats::na.omit(as.vector(data_in)))
} else {
bw <- bw
}
total_rng_dens <- stats::density(stats::na.omit(as.vector(vals)), bw = bw)[["x"]]
density_calc <- stats::density(input_data[["y"]], bw = bw)
density_range <- density_calc[["x"]]
density_df <- dplyr::tibble(
x = density_calc[["x"]],
y = density_calc[["y"]]
)
if (trim) { # implementation of filtering values
# only to actual and slightly outside the range
filter_range <- range(vals, na.rm = TRUE) %>%
scales::expand_range(mul = 0.05)
density_df <- dplyr::filter(
density_df,
dplyr::between(.data$x, filter_range[1], filter_range[2])
)
}
plot_base <- ggplot(density_df) +
theme_void()
plot_out <- plot_base +
geom_area(
aes(x = .data$x, y = .data$y),
color = line_color,
fill = fill_color
) +
xlim(range(density_range, na.rm = TRUE)) +
coord_cartesian(
xlim = range(total_rng_dens, na.rm = TRUE),
expand = TRUE,
clip = "off"
)
}
}
out_name <- file.path(
tempfile(pattern = "file", tmpdir = tempdir(), fileext = ".svg")
)
ggsave(
out_name,
plot = plot_out,
dpi = 25.4,
height = fig_dim[1],
width = fig_dim[2],
units = "mm"
)
img_plot <- out_name %>%
readLines() %>%
paste0(collapse = "") %>%
gt::html()
on.exit(file.remove(out_name), add = TRUE)
img_plot
}
if(!rlang::quo_is_null(rlang::enquo(type_col))){
type_vec <- gt_index(gt_object, {{ type_col }}, as_vector = TRUE)
type <- type_vec
stopifnot("You must indicate the `type` of plot as one of 'boxplot', 'histogram', 'rug_strip' or 'density'." = isTRUE(all(type %in% c("boxplot", "rug_strip", "histogram", "density"))))
}
text_transform(
gt_object,
locations = cells_body(columns = {{ column }}),
fn = function(x) {
mapply(plot_fn_spark, trim, list_data_in, type, SIMPLIFY = FALSE)
}
)
}