R for Data Science（二）

library(tidyverse)
x <- tribble(
  ~key, ~val_x,
  1, "x1",
  2, "x2",
  3, "x3"
)
y <- tribble(
  ~key, ~val_y,
  1, "y1",
  2, "y2",
  4, "y3"
)

x %>%
  inner_join(y)

Joining, by = "key"
# A tibble: 2 x 3
    key val_x val_y
  <dbl> <chr> <chr>
1     1 x1    y1   
2     2 x2    y2

> x %>%
+   left_join(y)
Joining, by = "key"
# A tibble: 3 x 3
    key val_x val_y
  <dbl> <chr> <chr>
1     1 x1    y1   
2     2 x2    y2   
3     3 x3    NA   

> x %>%
+   right_join(y)
Joining, by = "key"
# A tibble: 3 x 3
    key val_x val_y
  <dbl> <chr> <chr>
1     1 x1    y1   
2     2 x2    y2   
3     4 NA    y3   

> x %>%
+   full_join(y)
Joining, by = "key"
# A tibble: 4 x 3
    key val_x val_y
  <dbl> <chr> <chr>
1     1 x1    y1   
2     2 x2    y2   
3     3 x3    NA   
4     4 NA    y3

> x <- tribble(
+   ~key, ~val_x,
+   1, "x1",
+   2, "x2",
+   2, "x3",
+   1, "x4"
+ )

> y <- tribble(
+   ~key, ~val_y,
+   1, "y1",
+   2, "y2"
+ )

> x %>%
+   left_join(y, by = "key")
# A tibble: 4 x 3
    key val_x val_y
  <dbl> <chr> <chr>
1     1 x1    y1   
2     2 x2    y2   
3     2 x3    y2   
4     1 x4    y1

> x <- tribble(
+   ~key, ~val_x,
+   1, "x1",
+   2, "x2",
+   2, "x3",
+   3, "x4"
+ )

> y <- tribble(
+   ~key, ~val_y,
+   1, "y1",
+   2, "y2",
+   2, "y3",
+   3, "y4"
+ )

> x %>% left_join(y)
Joining, by = "key"
# A tibble: 6 x 3
    key val_x val_y
  <dbl> <chr> <chr>
1     1 x1    y1   
2     2 x2    y2   
3     2 x2    y3   
4     2 x3    y2   
5     2 x3    y3   
6     3 x4    y4

library(nycflights13)
flights %>%
  left_join(airports, by = c("dest" = "faa"))

merge(x, y) # 内连接
merge(x, y, all.x = T) # 左连接
merge(x, y, all.y = T) # 右连接
merge(x, y, all.x = T, all.y = T) # 全连接

> x <- tribble(
+   ~key, ~val_x,
+   1, "x1",
+   2, "x2",
+   3, "x3"
+ )

> y <- tribble(
+   ~key, ~val_y,
+   1, "y1",
+   2, "y2",
+   4, "y3"
+ )

> x %>%
+   semi_join(y)
Joining, by = "key"
# A tibble: 2 x 2
    key val_x
  <dbl> <chr>
1     1 x1   
2     2 x2   

> x %>%
+   anti_join(y)
Joining, by = "key"
# A tibble: 1 x 2
    key val_x
  <dbl> <chr>
1     3 x3

df1 <- tribble(
  ~x, ~y,
  1,  1,
  2,  1
)
df2 <- tribble(
  ~x, ~y,
  1,  1,
  1,  2
)

intersect(df1, df2) # 返回在两个数据框中都有的观测值
union(df1, df2) # 返回两个数据框的并集
> dplyr::setdiff(df1, df2) # df1中逐个去除df2中的元素
# A tibble: 1 x 2
      x     y
  <dbl> <dbl>
1     2     1

> dplyr::setdiff(df2, df1) # df2中逐个去除df1中的元素
# A tibble: 1 x 2
      x     y
  <dbl> <dbl>
1     1     2

library(stringr)

# 字符串长度
str_length(c("a", "R for data science", NA))

# 字符串合并
str_c("x", "y", "z")
str_c("x", "y", "z", sep = ", ")

x <- c("a", NA)
str_c("|-", x, "-|")
# 如果想把NA作为字符串处理，可以使用str_replace_na()函数
str_c("|-", str_replace_na(x), "-|")

name <- "程振兴"
time_of_day <- "早上好！"
birthday <- FALSE

str_c(
  time_of_day, " ", name,
  "！",
  if (birthday) "另外祝你生日快乐!"
)

# str_c是向量化的函数
str_c("prefix-", c("x", "y", "z"), "-suffix")

# 合并向量
str_c(c("x", "y", "z"), collapse = "")

# 字符串替换
x <- c("Apple", "Banana", "Pear")
str_sub(x, 1, 3)
str_sub(x, -3, -1)
str_sub(x, 1, 1) <- str_to_lower(str_sub(x, 1, 1))
x

# 不同语言环境的单词排序
x <- c("apple", "eggplant", "banana")
# English
str_sort(x, locale = "en")
# Hawaiian
str_sort(x, locale = "haw")

# 去除字符串两端的空格
str_trim(" apple ", side = "left")
str_trim(" apple ", side = "right")
str_trim(" apple ", side = "both")

# 正则表达式
x <- c("apple", "banana", "pear")
str_view(x, "an")
str_view(x, ".a.")
# 转义字符
str_view(c("abc", "a.c", "bef"), "a\\.c")

(x <- "a\\b")
writeLines(x)
str_view(x, "\\\\")

# ^ 匹配字符串的开头。
# $ 匹配字符串的结尾。
x <- c("apple", "banana", "pear")
str_view(x, "^a")
str_view(x, "a$")

x <- c("apple pie", "apple", "apple cake")
str_view(x, "apple")
str_view(x, "^apple")
str_view(x, "apple$")

# 匹配以y开头的单词
(words <- stringr::words) %>% str_view("^y")
# 匹配以x结尾的单词
(words <- stringr::words) %>% str_view("x$")
# 匹配长度为3的单词
(words <- stringr::words) %>% str_view("^[a-z][a-z][a-z]$")
# 匹配有七个字母或更多。
(words <- stringr::words) %>% str_view(".[a-z][a-z][a-z][a-z][a-z][a-z][a-z].")

# \d：匹配任何数字。
# \s：匹配任何空格（例如空格，制表符，换行符）。
# [abc]：匹配a，b或c。
# [^abc]：匹配除a，b或c之外的任何内容。

str_view(c("abc", "a.c", "a*c", "a c"), "a[.]c")
str_view(c("abc", "a.c", "a*c", "a c"), ".[*]c")
str_view(c("abc", "a.c", "a*c", "a c"), "a[ ]")
str_view(c("grey", "gray"), "gr(e|a)y")

# ?：0或1
# +：1或更多
# *：0或更多
x <- "1888 is the longest year in Roman numerals: MDCCCLXXXVIII"
str_view(x, "CC?")
str_view(x, "CC+")
str_view(x, "CC*")

# {n}：完全是n
# {n,}：n或更多
# {,m}：最多m
# {n,m}：在n和m之间
str_view(x, "C{2}")
str_view(x, "C{1,3}")
str_view(x, "C{2,}")

# 默认使贪婪模式，它们将匹配可能的最长字符串。
# 也可以通过使用?来匹配最短的字符
str_view(x, "C{1,3}")
str_view(x, "C{1,3}?")
str_view(x, 'C[LX]+?')

# 查找具有重复字母对的所有水果。
str_view(fruit, "(..)\\1", match = TRUE)

# 检查匹配
x <- c("apple", "banana", "pear")
# 是否匹配到e？
str_detect(x, "e")

# 以t开头的单词数
sum(str_detect(words, "^t"))
# 以元音结尾的单词比例
mean(str_detect(words, "[aeiou]$"))
# 含有元音的单词的比例
mean(str_detect(words, "[aeiou]"))

# 是否不含元音？
no_vowels_1 <- !str_detect(words, "[aeiou]")
no_vowels_2 <- str_detect(words, "^[^aeiou]+$")
# 上面两种匹配方式是等效的
identical(no_vowels_1, no_vowels_2)

# 以x结尾的单词
words[str_detect(words, "x$")]
# 等价于
str_subset(words, "x$")

# 过滤出数据框中以x结尾的观测值
df <- tibble(
  word = words,
  i = seq_along(word)
)
df
df %>%
  filter(str_detect(word, "x$"))

# 计算匹配数量
x <- c("apple", "banana", "pear")
str_count(x, "a")

# 统计单词的元音个数、辅音个数以及单词的长度
df %>%
  mutate(
    vowels = str_count(word, "[aeiou]"),
    consonants = str_count(word, "[^aeiou]"),
    length = str_length(word)
  )

# 字符串"abababa"中"aba"出现了多少次？
str_count("abababa", "aba") # 2
str_view_all("abababa", "aba")

# 提取含颜色的句子
colours <- c("red", "orange", "yellow", "green", "blue", "purple")
colour_match <- str_c(colours, collapse = "|")
colour_match
# 注意str_extract()仅提取第一个匹配项。
str_subset(sentences, colour_match) %>% str_extract(colour_match)
# str_extract_all()提取所有匹配项。
str_subset(sentences, colour_match) %>% str_extract_all(colour_match)
# 如果使用simplify = TRUE，str_extract_all()将返回一个矩阵，其中短匹配扩展到与最长匹配的长度：
str_subset(sentences, colour_match) %>% str_extract_all(colour_match, simplify = T) %>% as_tibble()

x <- c("a", "a b", "a b c")
str_extract_all(x, "[a-z]", simplify = TRUE) %>% as_tibble()

# 查找“a”或“the”之后的任何单词:
noun <- "(a|the) ([^ ]+)"

has_noun <- sentences %>%
  str_subset(noun) %>%
  head(10)
has_noun %>%
  str_extract(noun)

has_noun %>%
  str_match(noun) %>%
  as_tibble()

# tidyr::extract()也可以用于字符串匹配提取，并且还可以重新命名
tibble(sentences = sentences) %>%
  tidyr::extract(
    sentences, c("article", "noun"),
    "(a|the) ([^ ]+)",
    remove = F
  )

# 匹配所有的匹配项。
has_noun %>%
  str_match_all(noun)

# 替换
x <- c("apple", "pear", "banana")
str_replace(x, "[aeiou]", "-")
str_replace_all(x, "[aeiou]", "-")

x <- c("1 house", "2 cars", "3 people")
str_replace_all(x, c("1" = "one",
                     "2" = "two",
                     "3" = "three"))

# 反转第2、3个单词的顺序
sentences %>%
  str_replace("([^ ]+) ([^ ]+) ([^ ]+)", "\\1 \\3 \\2") %>%
  head(5)

# 分割
sentences %>%
  head(1) %>%
  str_split(" ")

"a|b|c|d" %>%
  str_split("\\|") %>%
  .[[1]]

# 与返回列表的其他字符串函数一样，您可以使用simplify = TRUE返回矩阵
sentences %>%
  head(5) %>%
  str_split(" ", simplify = T)
fields <- c("姓名：程振兴", "国家：中国", "年龄：35")
fields %>%
  str_split("：", n = 2, simplify = T)

x <- "This is a sentence.  This is another sentence."
str_view_all(x, boundary("word"))
x %>%
  str_split(" ") %>%
  .[[1]]

x %>%
  str_split(boundary("word")) %>%
  .[[1]]

# 定位到匹配成功的位置
str_locate(x, "is")
str_locate_all(x, "is")

str_view(fruit, "nana")
# 上面的语句是下面语句的简写
str_view(fruit, regex("nana"))
# ignore_case = TRUE允许字符匹配大写或小写形式。
bananas <- c("banana", "Banana", "BANANA")
str_view(bananas, "banana")
str_view(bananas, regex("banana",
                        ignore_case = T))

# multiline = TRUE允许^和$匹配每一行的开头和结尾，而不是整个字符串的开头和结尾。
x <- "Line 1\nLine 2\nLine 3"
writeLines(x)
str_extract_all(x, "^Line")[[1]]
str_extract_all(x, regex("^Line", multiline = T))[[1]]

# comments = TRUE允许您使用注释和空格来使复杂的正则表达式更容易理解。

phone <- regex("
 \\(? # 可选的小括号
 (\\d{3}) # 三位的数字
 [) -]? # 小括号、空格或者短线
 (\\d{3}) # 三位的数字
 [ -]? # 可选的空格或短线
 (\\d{3}) # 三个数字
               ", comment = T)
str_match("514-791-8141", phone)
# dotall = TRUE允许.匹配所有内容，包括\n。

# 你还可以使用其它三个函数代替regex():
# fixed()：完全匹配指定的字节序列
microbenchmark::microbenchmark(
  fixed = str_detect(sentences, fixed("the")),
  regex = str_detect(sentences, "the"),
  times = 30
)

# 注意使用fixed()匹配非英语数据可能会出问题，因为非英文字符通过有多种方式表示。
a1 <- "\u00e1"
a2 <- "a\u0301"
c(a1, a2)
identical(a1, a2) # FALSE

# coll()：对于不区分大小写的匹配非常有用
i <- c("I", "İ", "i", "ı")
i
str_subset(i, coll("i", ignore_case = T))
# 对于其它语言环境
str_subset(i, coll("i", ignore_case = T,
                   locale = "tr"))
# 查看默认的语言环境
stringi::stri_locale_info()

# apropos()可以用于搜索全局环境中可用的所有对象。
apropos("replace")

# dir()函数可以用于列出目录中的所有文件
head(dir(pattern = "\\.do$"))

library(stringi)
# 计算单词的数量
stri_count_words(sentences) %>% head()

# 查找重复的字符串
stri_duplicated(c(
  "the", "brown", "cow", "jumped", "over",
  "the", "lazy", "fox"
))

# 生成随机文本
stri_rand_lipsum(nparagraphs = 1)
stri_rand_strings(n = 4, length = 5)

stri_rand_shuffle("我是程振兴")

# 字母排序
string1 <- c("hladny", "chladny", "bcade")
stri_sort(string1, locale = "en")
string2 <- c("number100", "number2")
stri_sort(string2, opts_collator = stri_opts_collator(numeric = T))

ggplot(gss_cat, aes(race)) +
  geom_bar() +
  scale_x_discrete(drop = FALSE)

relig_summary <- gss_cat %>%
  group_by(relig) %>%
  summarise(
    age = mean(age, na.rm = TRUE),
    tvhours = mean(tvhours, na.rm = TRUE),
    n = n()
  ) %>%
  mutate(
    relig = fct_reorder(relig, tvhours)
  )

# 将relig按照tvhours排序
ggplot(relig_summary, aes(tvhours, relig)) + geom_point()

gss_cat %>%
  filter(!is.na(age)) %>%
  count(age, marital) %>%
  group_by(age) %>%
  mutate(prop = n / sum(n)) %>%
  ggplot(aes(
    age, prop,
    colour = fct_reorder2(
      marital, age, prop))) +
  geom_line() +
  labs(colour = "marital")

gss_cat %>%
  mutate(
    marital = marital %>%
      fct_infreq() %>%
      fct_rev()
  ) %>%
  ggplot(aes(marital)) +
  geom_bar()

gss_cat %>% count(partyid)
gss_cat %>%
  mutate(
    partyid = fct_recode(
      partyid,
      "无回答" = "No answer",
      "不知道" = "Don't know",
      "其它" = "Other party"
    )
  ) %>%
  count(partyid)
# 你还可以把不同组修改到同一组
gss_cat %>%
  mutate(
    partyid = fct_recode(
      partyid,
      "其它" = "No answer",
      "其它" = "Don't know",
      "其它" = "Other party"
    )
  ) %>%
  count(partyid)

# fct_collapse()对于组合多个级别很方便
gss_cat %>%
  mutate(partyid = fct_collapse(
    partyid,
    "其它" = c("No answer", "Don't know", "Other party")
  )) %>%
  count(partyid)

# 如果你想把所有的小组混合在一起，使用fct_lump()：
gss_cat %>%
  mutate(relig = fct_lump(relig)) %>%
  count(relig)

# 默认情况是逐步将最小的组组合在一起，确保聚合之后的组仍然是最小组，也可以指定保留的组
gss_cat %>%
  mutate(relig = fct_lump(relig, n = 10)) %>%
  count(relig, sort = T)

library(lubridate)
today()
now()
ymd("2017-03-01")
mdy("Jan. 31st, 2017")
dmy("31-Jan-2017")
ymd(20170101)
ymd_hms("20170101201159")

# make_datetime()/make_date()
flights %>%
  select(year, month, day, hour, minute) %>%
  mutate(
    departure = make_datetime(year, month, day, hour, minute),
    date = make_date(year, month, day)
  ) %>%
  select(departure, date)

(datetime <- ymd_hms(now()))
datetime %>% year()
datetime %>% month()
datetime %>% day()
# 获取星期
datetime %>% wday(label = T, abbr = F) # 4表示周三
datetime %>% month(label = T, abbr = F)

datetime %>%
  floor_date("week") # 返回上一个周日的日期
datetime %>%
  ceiling_date("week") # 返回下一个周日的日期

datetime %>%
  round_date("week")
datetime %>%
  round_date("month")
datetime %>%
  round_date("year")

datetime
hour(datetime) <- hour(datetime) + 1
datetime

update(datetime, year = 2020, month = 2)

> age <- today() - ymd(19960624)
> age
Time difference of 8248 days
> age %>% as.duration()
[1] "712627200s (~22.58 years)"

dseconds(15)
days_in_month("2012-02-01" %>% ymd())
dweeks(3)
dyears(22)
ddays(0:5)
dhours(c(12, 20))
# 可以在天数之间添加和减去持续时间
(tomorrow <- today() + ddays(1))
(last_year <- today() - dyears(1))

seconds(15)
minutes(10)
hours(c(12, 24))
years(1)
10 * (months(6) + days(1))

ymd("2016-01-01") + dyears(1)
ymd("2016-01-01") + years(1)

years(1)/days(1) # 365.25

next_year <- today() + years(1)
(today() %--% next_year) / ddays(1) # 365

(today() %--% next_year) %/% ddays(1)

library(magrittr)
assign("x", 10)
("x" %>% assign(100))

tryCatch(stop("!"), error = function(e) "An error!")

# 点点
commas <- function(...){
  stringr::str_c(..., collapse = ", ")
}
commas(letters[1:10])

# getOption("width")：获取控制台的宽度
rule <- function(..., pad = "-"){
  title <- paste0(...)
  width <- getOption("width") - nchar(title) - 5
  print(width)
  cat(title, " ", stringr::str_dup(pad, width), "\n", sep = "")
}

rule("重要的输出：")

# 编写可管理的函数
# invisible()函数可以隐藏输出
show_missings <- function(df){
  n <- sum(is.na(df))
  cat("缺失值的数量为：", n, "\n", sep = "")
  invisible(df)
}
show_missings(flights)
x <- show_missings(flights)
dim(x) # 可以看出x是个数据框

# 定义操作符
`%czx%` <- function(x, y){
  return(sum(x, y))
}
10 %czx% 100

set_names(1:3, c("a", "b", "c"))

# 属性
x <- 1:10
attr(x, "greeting")
attr(x, "greeting") <- "Hi!"
attr(x, "farewell") <- "Bye!"
attributes(x)

x <- ymd_hms(now())
x
unclass(x)
attr(x, "tzone") <- "US/Pacific"
x
y <- as.POSIXlt(x)
y
attributes(y)

(tb <- tibble(x = 1:5, y = 5:1))
typeof(tb)

df <- tibble(
  a = rnorm(10),
  b = rnorm(10),
  c = rnorm(10),
  d = rnorm(10)
)

output <- vector("double", ncol(df))
for(i in seq_along(df)){
  output[[i]] <- median(df[[i]])
}
output

means <- c(0, 1, 2)
out <- vector("list", length(means))
for(i in seq_along(means)){
  n <- sample(100, 1)
  out[[i]] <- rnorm(n, means[[i]])
}
str(out)
# unlist()函数可以将矢量列表展平为单个矢量。
str(unlist(out))

# map() 映射产生列表。
# map_lgl() 映射产生一个逻辑向量。
# map_int() 映射产生一个整数向量。
# map_dbl() 映射产生一个double向量。
# map_chr() 映射产生一个字符矢量。
map_dbl(df, mean)
map_dbl(df, median)
z <- list(x = 1:3, y = 4:5)
map_int(z, length)

# 将mtcars数据集拆分为三个部分（每个圆柱体一个），并为每个部分拟合相同的线性模型：
models <- mtcars %>%
  split(.$cyl) %>%
  map(function(df) lm(mpg ~ wt, data = df))
# 上面的代码等价于：
models <- mtcars %>%
  split(.$cyl) %>%
  map(~lm(mpg ~ wt, data = .))

models %>%
  map(summary) %>%
  map_dbl(~.$r.squared) %>%
  round(3)
# 下面的代码功能相同
models %>%
  map(summary) %>%
  map_dbl("r.squared") %>%
  round(3)

# 还可以按照位置选择元素
x <- list(list(1, 2, 3), list(4, 5, 6), list(7, 8, 9))
x %>% map_dbl(2)

# lapply()基本等同于，map()
# sapply()可以自动简化输出

x1 <- list(
  c(0.27, 0.37, 0.57, 0.91, 0.20),
  c(0.90, 0.94, 0.66, 0.63, 0.06),
  c(0.21, 0.18, 0.69, 0.38, 0.77)
)
x2 <- list(
  c(0.50, 0.72, 0.99, 0.38, 0.78),
  c(0.93, 0.21, 0.65, 0.13, 0.27),
  c(0.39, 0.01, 0.38, 0.87, 0.34)
)

threshold <- function(x, cutoff = 0.8){
  x[x > cutoff]
}
x1 %>% sapply(threshold) %>% str()
x1 %>% lapply(threshold) %>% str()
x1 %>% map(threshold) %>% str()

x2 %>% sapply(threshold) %>% str()
x2 %>% lapply(threshold) %>% str()
x2 %>% map(threshold) %>% str()

# vapply()函数是一个安全的替代方法
x2 %>% vapply(threshold, FUN.VALUE = 0) %>% str()

mu <- list(5, 10, -3)
mu %>%
  map(rnorm, n = 5) %>%
  str()

sigma <- list(1, 5, 10)
seq_along(mu) %>%
  map(~rnorm(5, mu[[.]], sigma[[.]])) %>%
  str()
# 但是这样会有一种混淆的感觉，使用map2()更方便
map2(mu, sigma, rnorm, n = 5) %>%
  str()

# 如果有多个参数要映射呢？
n <- list(1, 3, 5)
args <- list(n, mu, sigma)
args %>%
  pmap(rnorm) %>%
  str()

# 下面的方法更可靠
list(mean = mu, sd = sigma, n = n) %>%
  pmap(rnorm) %>%
  str()

params <- tribble(
  ~mean, ~sd, ~n,
  5,     1,  1,
  10,     5,  3,
  -3,    10,  5
)

params %>%
  pmap(rnorm) %>%
  str()

# 调用不同的函数
f <- c("runif", "rnorm", "rpois")
param <- list(
  list(min = -1, max = 1),
  list(sd = 5),
  list(lambda = 10)
)
invoke_map(f, param, n = 5) %>% str()

sim <- tribble(
  ~f,      ~params,
  "runif", list(min = -1, max = 1),
  "rnorm", list(sd = 5),
  "rpois", list(lambda = 10)
)
sim %>%
  mutate(sim = invoke_map(f, params, n = 10))

# 当你想要调用函数的副作用而不是其返回值的时候，walk()函数是个不错的替代方法
x <- list(1, "a", 3)
x %>%
  walk(print)
# 如果您有一个图表列表和一个文件名向量，您可以使用pwalk()将每个文件保存到磁盘上的相应位置：
plots <- mtcars %>%
  split(.$cyl) %>%
  map(~ggplot(., aes(mpg, wt)) + geom_point() + theme_bw())
paths <- stringr::str_c(names(plots), ".pdf")

pwalk(list(paths, plots), ggsave, path = getwd())

iris %>%
  keep(is.factor) %>%
  str()

iris %>%
  discard(is.factor) %>%
  str()

x <- list(1:5, letters, list(10))
x %>% str()
x %>%
  some(is_character)

x %>%
  every(is_vector)

(x <- sample(10))
x %>%
  detect(~ . >5)
x %>%
  detect_index(~ . > 5)
x %>%
  head_while(~ . > 5)
x %>%
  tail_while(~ . > 5)

# reduce()函数
# 多个tibble的整合
dfs <- list(
  age = tibble(name = "John", age = 30),
  sex = tibble(name = c("John", "Mary"), sex = c("M", "F")),
  trt = tibble(name = "Mary", treatment = "A")
)
dfs %>% reduce(full_join)

# 找到向量的交集
vs <- list(
  c(1, 3, 5, 6, 10),
  c(1, 2, 3, 7, 8, 10),
  c(1, 2, 3, 4, 8, 9, 10)
)
vs %>% reduce(intersect)

# accumulate()函数
x <- sample(10)
x %>% accumulate(`+`)
x %>% accumulate(`%czx%`)