04-常见Rpackage介绍¶

## 常用方法&在本地电脑方法
install.packages("dplyr")

## 远程桌面方法（无法联网，只能从指定路径安装）
options(repos = c(CRAN = "https://nexus.hengrui.com/repository/r-proxy-cran/"))
install.packages("dplyr")

## 本地下载包安装方式
install.packages("C:/Users/wangjy35/Downloads/renv_1.1.8.tar.gz")

add_labels() ## Apply labels to multiple variables on a data frame
add_variables() ## Add Missing Variables
build_from_derived() ## Build a dataset from derived
build_qnam() ## Build the observations for a single QNAM
check_ct_col() ## Check Control Terminology for a Single Column
check_ct_data() ## Check Control Terminology for a Dataset
check_unique_keys() ## Check Uniqueness of Records by Key
check_variables() ## Check Variable Names
combine_supp() ## Combine the Domain and Supplemental Qualifier
convert_var_to_fct() ## Convert Variable to Factor with Levels Set by Control Terms
create_cat_var() ## Create Categorical Variable from Codelist
create_subgrps() ## Create Subgroups
create_var_from_codelist() ## Create Variable from Codelist
drop_unspec_vars() ## Drop Unspecified Variables
get_bad_ct() ## Gets vector of control terminology which should be there
make_supp_qual() ## Make Supplemental Qualifier
metatools_example() ## Get path to pkg example
order_cols() ## Sort Columns by Order
remove_labels() ## Remove labels to multiple variables on a data frame
set_variable_labels() ## Apply labels to a data frame using a metacore object
sort_by_key() ## Sort Rows by Key Sequence

group_summary <- mydata %>%
  group_by(group_var) %>%
  summarise(
    n = n(),
    mean_var1 = mean(var1, na.rm = TRUE),
    sd_var1 = sd(var1, na.rm = TRUE),
    median_var1 = median(var1, na.rm = TRUE),
    min_var1 = min(var1, na.rm = TRUE),
    max_var1 = max(var1, na.rm = TRUE)
  ) %>%
  ungroup() ## 在使用完group_by后一定要注意使用ungroup，避免影响后续其他计算

mydata %>%
  group_by(group_var) %>%
  mutate(
    lag_value = lag(value_var),      # 前一行
    lead_value = lead(value_var)    # 后一行
  ) %>%
  ungroup()

## 横向合并
merged <- inner_join(data1, data2, by = "key_var")
left_joined <- left_join(data1, data2, by = "key_var")
right_joined <- right_join(data1, data2, by = "key_var")
full_joined <- full_join(data1, data2, by = "key_var")

## 纵向合并，类似set操作
appended <- bind_rows(data1, data2, data3)

## 数据转置（用到tidyr包）
long <- wide %>%
  tidyr::pivot_longer(
    cols = c(var1, var2, var3),
    names_to = "variable",
    values_to = "value"
  )

wide_data <- long %>%
  tidyr::pivot_wider(
    names_from = variable,
    values_from = value
  )

sas_data <- haven::read_sas("/path/to/data.sas7bdat") 

xportr::write_xportr(sas_data, "/path/to/output.xpt")

mydata <- readxl::read_excel("/path/to/input.xlsx", sheet = "Sheet1",  # 指定工作表
                     col_names = TRUE,
                     na = c("", "NA"))

mydata <- readxl::read_excel("/path/to/input.xlsx", col_types = "text") # 直接将所有列读入为字符型，避免readxl自动判断数据类型导致的错误
mydata <- readxl::read_excel("/path/to/input.xlsx", guess_max = Inf) # 增加guess_max参数，默认值为1000，增加后可以让readxl查看更多行数据来判断数据类型，避免前面几行数据类型单一导致的错误判断

mydata <- metacore::read_all_sheets("/path/to/input.xlsx") # 批量读入，内部调用的也是read_excel(., col_types = "text")

mydata <- openxlsx::read.xlsx("/path/to/input.xlsx", sheet = "Sheet1", 
                              colNames = TRUE, na.strings = c("", "NA")) # openxlsx的读入方法，功能和readxl类似，速度更快

openxlsx::write.xlsx(mydata, "/path/to/output.xlsx")

# 保存单个对象
saveRDS(mydata, file = "/path/to/output.rds") # 保存为rds格式
mydata <- readRDS("/path/to/output.rds") # 读取rds格式

# 保存多个对象
save(mydata, file = "/path/to/output.RData") # 保存为RData格式
load("/path/to/output.RData") # 读取RData格式

diffdf::diffdf(df1, df2)

## 显示数据集中比对差异
diff <- diffdf::diffdf(df1, df2, suppress_warnings = TRUE)
diffdf::diffdf_issuerows(df1, diff)
diffdf::diffdf_issuerows(df2, diff)

# FOR循环
for(i in 1:10) {
  squared <- i^2
  print(paste(i, "squared is", squared))
}

# WHILE循环
i <- 1
while(i <= 10) {
  squared <- i^2
  print(paste(i, "squared is", squared))
  i <- i + 1
}

# if语句
x <- 10
if(x > 5) {
  print("x is greater than 5")
} else if(x == 5) {
  print("x is equal to 5")
} else {
  print("x is less than 5")
}

# switch语句
day <- "Monday"
result <- switch(day,
                 "Monday" = "Start of the week",
                 "Friday" = "End of the week",
                 "Weekend")
print(result)