Newer
Older
# Script: filter different markets( just for cake oil market and sugar, cereals...)
#' function reads balance detailed, split it into "biofuels" and "non-biofuels"
#'
#' @param balance_detailed A data frame.
#' @param p_biofuels boolean.
#' @return links
#' @export
#'
prelinks <- function(balance_detailed, p_biofuels = TRUE){
# balance_detailed <- balance_detailed_Scenario
# add variable key, itemName and color from products
balance_detailed <- left_join(balance_detailed, products[,c(2,3,9)], by = c(".i4" = "key"))
# rename new variable
colnames(balance_detailed)[13] <- "Commodities"
colnames(balance_detailed)[1] <- "key"
balance_detailed$Commodities <- sub("\\[.*?\\]", "", balance_detailed$Commodities)
library(magrittr)
# prod_list_fullname <- balance_detailed %>% select(Commodities) %>% extract2(1)
# remove character begin with [ and end with ]
# prod_list_fullname <- gsub("\\[.*?\\]", "", prod_list_fullname)
# p_biofuels = T
if (p_biofuels){
selected_list = c("Rape seed",
"Soya seed",
"Sunflower seed",
"Rape seed oil",
"Soya oil",
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
"Soya cake",
"Sunflowe seed cake",
"Rape seed cake",
"Destilled dried grains from bio-ethanol processing",
"Pulses",
"Bio ethanol")
}else{
selected_list = c("Wheat","Grain maize","Barley", "Other cereals","Poultry meat",
"Sugar", "Fish and other acquatic products", "Rice milled")
}
#
balance_detailed_ <- balance_detailed %>% as.data.frame()
balance_detailed_selected <- filter(balance_detailed_, balance_detailed_[,13] %in% selected_list)
as_tibble(balance_detailed_selected)
#rename "Destilled dried grains from bio-ethanol processing" -> "DDGS"
# balance_detailed_selected <- balance_detailed_selected %>%
# mutate(Commodities = ifelse(Commodities!="Destilled dried grains from bio-ethanol processing",Commodities,"DDGS"))
# filter
balance_detailed_all <- balance_detailed_selected %>%
select(-.i1) %>% # remove region
select(-.i5) %>% # remove year: 2030
select(-interv_ch) %>%
group_by(Commodities) %>%
summarise_at(vars(supply:exports),sum,na.rm=TRUE)
library(magrittr)
prod_list_fullname <- balance_detailed_selected %>% select(Commodities) %>% extract2(1)
#print(prod_list_fullname)
vars_to_run <- prod_list_fullname
left_side <- c("supply","imports")
links_list <- list()
#nodes_list <- list()
vars_to_run <- vars_to_run[!vars_to_run %in%
c("Soya oil",
"Soya cake",
"Rape seed oil",
"Rape seed cake",
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
"Sunflowe seed cake",
"Bio ethanol",
"Destilled dried grains from bio-ethanol processing"
)]
# "Rape seed" "Sunflower seed" "Soya seed" "Pulses"
for (i in vars_to_run){
data_df <- reshape2::melt(balance_detailed_all)
data_df_1 <- data_df %>% filter(Commodities == i)
data_df_1_l <- data_df_1 %>% filter(variable %in% left_side )
data_df_1_r <- data_df_1 %>% filter(!variable %in% left_side)
data_df_1_l_agg <- data_df_1_l %>% group_by(Commodities) %>% summarise(value=sum(value))
data_df_1_l_agg$variable <- paste0(i)
first <- merge(data_df_1_l,data_df_1_l_agg,by="Commodities",all=TRUE)
first <- first[,c("variable.x","variable.y","value.x")]
colnames(first)[3] <- "value"
second <- merge(data_df_1_l_agg,data_df_1_r,by="Commodities",all=TRUE)
second <- second[,c("variable.x","variable.y","value.y")]
colnames(second)[3] <- "value"
if (p_biofuels){
if (i != "Pulses"){
if (i == "Soya seed") {produced = c("Soya oil","Soya cake")}
else if (i == "Rape seed") {produced = c("Rape seed oil","Rape seed cake") }
else if (i == "Sunflower seed") {produced = c("Sunflower seed oil","Sunflowe seed cake") }
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
else {print(i);break}
data_df_proc_help <- balance_detailed_all %>% filter(Commodities %in% produced)
data_df_proc_help <- reshape2::melt(data_df_proc_help)
data_df_proc_help <- data_df_proc_help %>% filter(variable== "supply" )
data_df_proc_help$variable <- paste(data_df_proc_help$Commodities)
data_df_proc_help$Commodities <- i
colnames(data_df_proc_help) <- c("variable.x","variable.y","value")
second <- rbind(second,data_df_proc_help)
second <- second %>% filter(variable.y!="processing")
data_df_proc_help <- balance_detailed_all %>% filter(Commodities %in% produced)
data_df_proc_help <- reshape2::melt(data_df_proc_help)
data_df_proc_help <- data_df_proc_help %>% filter(variable!="supply")
colnames(data_df_proc_help) <- c("variable.x","variable.y","value")
second <- rbind(second,data_df_proc_help)
second_imports <- second %>% filter(variable.y=="imports")
second <- second %>% filter(!variable.y=="imports")
colnames(second_imports) <- c("variable.y","variable.x","value")
second <- rbind(second,second_imports)
}
}
combined <- rbind(first,second)
links_list[[i]] <- combined
links <- rbindlist(links_list)
links <- unique(links)
}
if(p_biofuels){
# links[grepl("Bio",variable.x) | grepl("Bio",variable.y)]
setnames(links,old=c("variable.x","variable.y"),new=c("source","target"))
# links[grepl("Bio",source) | grepl("Bio",target)]
i <- c("Bio ethanol","Destilled dried grains from bio-ethanol processing")
data_df_proc_help <- balance_detailed_all %>% filter(Commodities %in% i)
data_df_proc_help <- reshape2::melt(data_df_proc_help)
data_df_proc_help <- data_df_proc_help %>% filter(variable=="supply")
data_df_proc_help$variable <- paste(data_df_proc_help$Commodities)
data_df_proc_help$Commodities <- "biofuels"
colnames(data_df_proc_help) <- c("variable.x","variable.y","value")
data_df <- reshape2::melt(balance_detailed_all)
data_df_1 <- data_df[data_df$Commodities %in% i,]
data_df_1_l <- data_df_1[data_df_1$variable %in% left_side,]
data_df_1_l$variable <- as.character(data_df_1_l$variable)
data_df_1_l <- data_df_1_l %>% filter(variable!="supply")
data_df_1_r <- data_df_1[!data_df_1$variable %in% left_side,]
data_df_1_r$variable <- as.character(data_df_1_r$variable)
data_df_1_l_agg <- data_df_1_l%>%group_by(Commodities) %>% summarise(value=sum(value))
data_df_1_l_agg$variable <- paste0(i)
first <- merge(data_df_1_l,data_df_1_l_agg,by="Commodities",all=TRUE)
first <- first[,c("variable.x","variable.y","value.x")]
colnames(first)[3] <- "value"
second <- merge(data_df_1_l_agg,data_df_1_r,by="Commodities",all=TRUE)
second <- second[,c("variable.x","variable.y","value.y")]
colnames(second)[3] <- "value"
second <- rbind(second,data_df_proc_help)
second <- second %>% filter(!variable.y %in% c("processing","biofuels"))
#nodes <- rbind(nodes,data.frame("name"=i))
combined <- rbind(first,second)
setnames(combined,old=c("variable.x","variable.y"),new=c("source","target"))
links <- rbind(links,combined)
}
else{
setnames(links,old=c("variable.x","variable.y"),new=c("source","target"))
}
# remove 0 value and <=900
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
# links[,value:=ifelse(value<=50,NA,value)]
links<-na.omit(links)
return(links)
}
#' function calulates sum of links
#'
#' @param balance_detailed_links links.
#' @return sum
#'
linksum <- function(balance_detailed_links){
# the total value of different products from source
linkcopysum<- aggregate(balance_detailed_links$value,
by=list(Category=balance_detailed_links$source),
FUN=sum)
# add a column: the total value of different prodcuts from target
linkcopysum <- merge(linkcopysum,
aggregate(balance_detailed_links$value,
by=list(Category=balance_detailed_links$target),
FUN=sum),
by="Category")
return(linkcopysum)
}
#' function Replaces specific characters within strings
#' @param old old column names.
#' @param new new column names.
#' @param df replace data frame
#' @return df
#'
#'
#'
replacestr <- function(old, new, df){
df$label <- gsub(old, new, df$label)
return(df)
}
#' function get links and nodes for drawing a sankey diagram
#' @param baseline Baseline balance market.
#' @param scenario Scenario balance market.
#' @param p_baseline boolean.
#' @param fixedNodePosition boolean.
#'
#' @return links and nodes
#' @export
#'
#'
links_nodes <- function(baseline, scenario, p_baseline, fixedNodePosition = TRUE){
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
linksum_baseline = linksum(baseline)
linksum_scenario = linksum(scenario)
# percentage diferences between scenario and baseline
linksum_scenario$percent_source = (linksum_scenario$x.x*100/linksum_baseline$x.x)-100
linksum_scenario$percent_target = (linksum_scenario$x.y*100/linksum_baseline$x.y)-100
#links = baseline
if (p_baseline){
links = baseline
}else{links= scenario}
nodes <- data.frame("name" = c(unique(c(as.character(links$source),links$target)) ))
# product: names -> number
links$source <- sapply(links$source,function(x)which(nodes$name %in% x))
links$source <- links$source -1
links$target <- sapply(links$target,function(x)which(nodes$name %in% x))
links$target <- links$target-1
if (p_baseline){
links$group <- paste0(gsub(" ", "_", baseline$source),"->", gsub(" ", "_", baseline$target))
}else{
links$group <- paste0(gsub(" ", "_", scenario$source),"->", gsub(" ", "_", scenario$target))
}
source_baseline <- baseline %>% as.data.frame() %>%
filter(source %in% c("imports", "supply")) %>%
group_by(source) %>%
summarise(sum(value))%>%
rename(name=source, sum = `sum(value)`)
source_scenario <- scenario %>% as.data.frame() %>%
filter(source %in% c("imports", "supply")) %>%
group_by(source) %>%
summarise(sum(value))%>%
rename(name=source, sum = `sum(value)`)
# calculate percent
source_scenario$percent<- (source_scenario$sum/source_baseline$sum)*100-100
target_baseline <- baseline %>% as.data.frame() %>%
filter(!(target %in% c("imports", "supply"))) %>%
group_by(target) %>%
summarise(sum(value)) %>%
rename(name=target, sum = `sum(value)`)
target_scenario <- scenario %>% as.data.frame() %>%
filter(!(target %in% c("imports", "supply"))) %>%
group_by(target) %>%
summarise(sum(value)) %>%
rename(name=target, sum = `sum(value)`)
# calculate percent
target_scenario$percent<- (target_scenario$sum/target_baseline$sum)*100-100
all_baseline <- rbind(source_baseline, target_baseline)
all_scenario <- rbind(source_scenario, target_scenario)
# if p_basline != TRUE -> add percent for nodes
if(!p_baseline){
# nodes$label <- ifelse(nodes$percent_target=="",
# as.character(nodes$label),
# paste0(nodes$label, " (", nodes$percent_target,"%", ")"))
all_scenario$label <- paste0(all_scenario$name, ": ",
format(round((all_scenario$sum)/1000,1), decimal.mark = ","),
"K (",round(all_scenario$percent,1), "%)" )
all <- all_scenario
#nodes$levels <- 1:3
}else{
all_baseline$label <- paste0(all_baseline$name, ": ",
format(round((all_baseline$sum)/1000,1), decimal.mark = "," ), "K" )
all <- all_baseline }
nodes_ <- left_join(nodes, all)
rm(nodes)
nodes <- nodes_
nodes$rank=c(1:nrow(nodes))
nodes = merge(nodes, linksum_scenario[,c("Category","percent_target")],
by.x="name",
by.y="Category",
all.x=TRUE)
nodes = nodes[order(nodes$rank,decreasing = FALSE),]
nodes$percent_target = round(nodes$percent_target,0)
nodes$percent_target <- ifelse(nodes$name=="biofuels","",nodes$percent_target)
nodes[is.na(nodes)]<-""
selectedcolor <- bind_rows(products %>% select(c(itemName,color)),
dim5s %>% select(c(itemName,color)))
selectedcolor$itemName <- sub("\\[.*?\\]", "", selectedcolor$itemName)
nodes <- left_join(nodes, selectedcolor, by = c("name" = "itemName"))
#print(nodes)
nodes <- nodes %>% select(c(rank,name,label,color))
# add links color
links_temp <- links
links_temp$color <- ""
colorlink <- c("Rape", "Sunflowe", "Soya", "Pulses",
"Wheat","Grain","Barley",
"Other","Poultry",
"Sugar", "Fish",
"Rice")
# get colors from nodes
color_nodes <- c("Wheat|Grain|Barley|Other|Poultry|Sugar|Fish|Rice|Rape seed$|Sunflowe seed|Soya seed$|Pulses")
getcolor <- nodes[grep(color_nodes, nodes$name), ] %>% select(c(name, color))
library(colorspace)
for (colorcol in colorlink){
temp <- getcolor %>%
filter(grepl(colorcol, getcolor$name))
# message(colorcol)
# print(links_temp[grepl(colorcol, group), ])
# print(links_temp[grep(colorcol, group), ])
links_temp[grep(colorcol, group), ]$color <- paste0(colorspace::lighten(temp$color, 0.05),"80")
}
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
links <- links_temp %>%
mutate(color = ifelse(color=="","rgba(123,123,123,0.1)",color))
# Replace specific characters within strings
old=c("supply",
"human_cons",
"processing",
"biofuels",
"feed",
"imports",
"exports",
"Destilled dried grains from bio-ethanol Processing")
new=c("Production",
"Human Consumption",
"Processing",
"Biofuels",
"Feed",
"Imports",
"Exports",
"DDG")
for (i in 1:length(old)){
nodes <- replacestr(old[i],new[i], nodes)
}
# add nodes position
if (fixedNodePosition) {nodes_lv <- nodes_position(nodes);nodes <- nodes %>% left_join(nodes_lv)}
return(list(links,nodes))
}
#' sets the position of nodes
#' @param nodes nodes for drawing a sankey diagram.
#' @return nodes with fixed position
#'
nodes_position <- function(nodes){
# nodes$level_x = ""
# nodes$level_y = ""
# if( "Pulses" %in% nodes$name){
# # for oil cake
# nodes <- nodes %>%
# mutate(level_x = ifelse(name %in% c("supply", "imports"), 0.01,
# ifelse(name %in% c("human_cons", "feed", "exports", "processing"),0.8,
# ifelse(name %in% c("Rape seed", "Soya seed", "Sunflower seed","Pulses"), 0.2,
# ifelse(name %in% c("Rape seed oil", "Soya oil", "Sunflower seed oil"),0.35,
# ifelse(name %in% c("Rape seed cake", "Soya cake", "Sunflowe seed cake", "biofuels"), 0.5, 0.65))) ))) %>%
# mutate(level_y = ifelse(name %in% c("supply", "imports"), c(0.01,0.8),
# ifelse(name %in% c("human_cons","Bio ethanol", "feed", "exports", "processing"),c(0.01,0.3,0.5,0.7,0.9),
# ifelse(name %in% c("Rape seed", "Soya seed","Sunflower seed","Pulses"), c(0.1,0.3,0.6,0.9),
# ifelse(name %in% c("Rape seed oil", "Soya oil", "Sunflower seed oil"),c(0.1, 0.4, 0.7),
# ifelse(name %in% c("Rape seed cake", "Soya cake", "Sunflowe seed cake", "biofuels"), c(0.1, 0.3, 0.5, 0.7), 0.5))) )))
#
#
# }else{
# nodes <- nodes %>%
# # for rest
# mutate(level_x = ifelse(name %in% c("supply", "imports"),0.01, ifelse(name %in% c("human_cons","biofuels", "feed", "exports", "processing"),0.8,0.4)))%>%
# mutate(level_y = ifelse(name %in% c("supply", "imports"),c(0.1,0.8), ifelse(name %in% c("human_cons","biofuels", "feed", "exports", "processing"),c(0.01,0.3,0.5,0.7,0.9),c(0.01,0.1,0.2,0.3,0.5,0.6,0.7,0.8))))
#
# }
nodes_level <- read.table(text="name level_x level_y
'supply' 0.01 0.1
'imports' 0.01 0.5
'Rape seed' 0.2 0.01
'Sunflowe seed cake' 0.5 0.5
'Soya seed' 0.2 0.3
'Bio ethanol' 0.65 0.14
'biofuels' 0.5 0.7
'feed' 0.8 0.3
'processing' 0.8 0.8
'Rape seed cake' 0.5 0.1
'Sunflower seed' 0.2 0.45
'Sunflower seed oil' 0.35 0.55
'Pulses' 0.2 0.6
'Destilled dried grains from bio-ethanol processing' 0.65 0.6
'human_cons' 0.8 0.01
'exports' 0.8 0.5", header=TRUE)}
else{nodes_level <- read.table(text = "name level_x level_y
'supply' 0.01 0.2
'imports' 0.01 0.5
'biofuels' 0.8 0.7
'feed' 0.8 0.3
'processing' 0.8 0.8
'human_cons' 0.8 0.01
'exports' 0.8 0.5
'Grain maize' 0.4 0.01
'Barley' 0.4 0.15
'Wheat' 0.4 0.35
'Poultry meat' 0.4 0.5
'Rice milled' 0.4 0.6
'Sugar' 0.4 0.7
'Fish and other acquatic products' 0.4 0.8"
, header=TRUE)}
return(nodes_level)
}
#' function draws sankey diagram and saves html pages for sankey
#'
#'
#' @param data a data frame object has two lists, which contains the links between the nodes and the nodes, the nodes has node id and properties of the nodes.links should have include the Source and Target for each link. An optional Value variable can be included to specify how close the nodes are to one another.
#' If no ID is specified then the nodes must be in the same order as the Source variable column in the Links data frame. Currently only grouping variable is allowed.
#' @param p_baseline boolean.
#' @param png boolean. if TRUE, the sankey chart will be saved.
#'
#' @return sankey diagram.
#' @export
plot_sankey <- function(data, p_baseline, png, outdata.dir){
links <- data[[1]]
nodes <- data[[2]]
N_ethanol <- nodes %>%
mutate_all(list(~ . %in% "Bio ethanol")) %>%
as.matrix %>%
which(., arr.ind = TRUE) %>%
nrow()
N_cereals <- nodes %>%
mutate_all(list(~ . %in% "Other cereals")) %>%
as.matrix %>%
which(., arr.ind = TRUE) %>%
nrow()
if (N_ethanol==2) NAME <- "comparison_cake_oil"
else if(N_cereals == 2 ) NAME <- "comparison_cereals"
else if(p_baseline){if("Bio ethanol" %in% nodes$name){ NAME <- "baseline"; print("Biofuels")}else{ NAME <- "baseline_Rest"}}
else{ if("Bio ethanol" %in% nodes$name){ NAME <- "scenario"; print("Biofuels")} else{ NAME <- "scenario_Rest"}}
# arrangement = 'perpendicular',
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
valueformat = ".0f",
valuesuffix = " tonnes",
node = list(
label = nodes$label,
color = nodes$color,
x = nodes$level_x,
y= nodes$level_y,
pad = 20,
thickness = 15,
line = list(
color = "blau",
width = 1
)
),
link = list(
source = links$source,
target = links$target,
value = links$value,
color = links$color
#label = links$group
)
) %>% layout(
autosize = F,
margin = list(l=10, r=10, b=50, t=50, pad=40),
title = "Market Balance in Tonnens",
font = list(size = 13, family ="Arial"),
xaxis = list(showgrid = F, zeroline = F),
yaxis = list(showgrid = F, zeroline = F)
)
#
library(htmlwidgets)
print(paste0("Export sankey diagram: ",paste0(outdata.dir,"/",NAME,"_fig.html")))
saveWidget(fig, paste0(outdata.dir,"/",NAME,"_fig.html"), selfcontained = F, libdir = "lib")
library(webshot)
#you convert it as png
if(png){
webshot(paste0(outdata.dir,"/",NAME,"_fig.html"),paste0(outdata.dir,"/",NAME,"_fig.png"),