separate() وunite() لأعمدة السلاسل النصية

تقسيم أعمدة السلاسل النصية ودمجها

library(tidyr)

# Date stored as a single string column
df <- data.frame(
  id = 1:4,
  date = c('2024-01-15','2024-03-22','2023-11-08','2024-06-30'),
  value = c(100, 200, 150, 175)
)

print(df)

separate() — تقسيم عمود واحد إلى عدة أعمدة

library(tidyr)

df <- data.frame(
  id = 1:4,
  date = c('2024-01-15','2024-03-22','2023-11-08','2024-06-30')
)

# Split 'date' into year, month, day
separate(
  df,
  col = date,
  into = c('year','month','day'),
  sep = '-'
)

separate() مع الوسيطة convert

library(tidyr)

df <- data.frame(
  event = c('2024-1','2024-2','2023-12'),
  score = c(85, 90, 78)
)

result <- separate(
  df,
  col = event,
  into = c('year','month'),
  sep = '-',
  convert = TRUE
)

print(result)
cat('year type:', class(result$year), '\n')
cat('month type:', class(result$month))

separate() مع remove = FALSE

library(tidyr)

df <- data.frame(
  full_code = c('US-CA-001','US-TX-002','UK-LDN-003')
)

separate(
  df,
  col = full_code,
  into = c('country','state','code'),
  sep = '-',
  remove = FALSE   # Keep original column
)

separate() عند مواضع ثابتة

library(tidyr)

# Product code: first 3 chars = category, next 4 = id
df <- data.frame(
  code = c('ABC1234','XYZ5678','DEF9012'),
  qty = c(10, 20, 15)
)

separate(
  df,
  col = code,
  into = c('category','product_id'),
  sep = 3    # Split after position 3
)

unite() — دمج الأعمدة في عمود واحد

library(tidyr)

df <- data.frame(
  first = c('Alice','Bob','Carol'),
  last = c('Smith','Jones','White'),
  score = c(85, 90, 78)
)

unite(
  df,
  col = 'full_name',
  first, last,
  sep = ' '
)

unite() مع remove = FALSE

library(tidyr)

df <- data.frame(
  year = c(2024, 2024, 2023),
  month = c(1, 3, 12),
  day = c(15, 22, 8)
)

unite(
  df,
  col = 'date_str',
  year, month, day,
  sep = '-',
  remove = FALSE
)

استخدام unite() لإنشاء المفاتيح

library(tidyr)
library(dplyr)

orders <- data.frame(
  year = c(2024, 2024, 2023),
  region = c('East','West','East'),
  seq_num = c(1, 1, 2)
)

orders_keyed <- orders %>%
  unite(col = 'order_key', year, region, seq_num, sep = '_')

print(orders_keyed)

separate_rows() — التقسيم إلى عدة صفوف

library(tidyr)

# Tags stored as comma-separated values in one cell
df <- data.frame(
  id = 1:3,
  title = c('Intro to R','Data Viz','ML Basics'),
  tags = c('r,beginner','r,ggplot,visualization','r,ml,modeling')
)

separate_rows(df, tags, sep = ',')

تسلسل separate() وunite()

library(tidyr)
library(dplyr)

# Reformat date from YYYY-MM-DD to DD/MM/YYYY
df <- data.frame(
  id = 1:3,
  date = c('2024-01-15','2024-03-22','2023-11-08')
)

df %>%
  separate(date, into=c('year','month','day'), sep='-') %>%
  unite(col='date_eu', day, month, year, sep='/')

معالجة الأجزاء الزائدة أو المفقودة

library(tidyr)

# Inconsistent data: some rows have 2 parts, some have 3
df <- data.frame(
  code = c('A-1','B-2-extra','C-3'),
  value = c(10, 20, 30)
)

# 'merge' keeps the last piece unsplit
separate(df, code, into=c('prefix','suffix'), sep='-',
         extra='merge')

تحقق سريع

مراجعة: separate() وunite()

library(tidyr)
library(dplyr)

df <- data.frame(
  id = 1:3,
  datetime = c('2024-01-15 09:30','2024-03-22 14:15','2023-11-08 11:45')
)

df %>%
  separate(datetime, into=c('date','time'), sep=' ') %>%
  separate(date, into=c('year','month','day'), sep='-', convert=TRUE) %>%
  separate(time, into=c('hour','minute'), sep=':', convert=TRUE)