الربط بين جداول متعددة في dplyr

لماذا نستخدم ربط الجداول؟

library(dplyr)

# Two related tables
orders <- data.frame(
  order_id = 1:4,
  customer_id = c(1, 2, 1, 3),
  amount = c(250, 180, 320, 90)
)

customers <- data.frame(
  customer_id = 1:3,
  name = c('Alice','Bob','Carol'),
  city = c('NYC','LA','Chicago')
)

print(orders)
print(customers)

inner_join() — الصفوف المتطابقة فقط

library(dplyr)

orders <- data.frame(
  order_id = 1:4,
  customer_id = c(1, 2, 1, 99),
  amount = c(250, 180, 320, 90)
)

customers <- data.frame(
  customer_id = c(1, 2, 3),
  name = c('Alice','Bob','Carol')
)

# customer_id = 99 in orders has no match => dropped
# customer_id = 3 in customers has no match => dropped
inner_join(orders, customers, by = 'customer_id')

left_join() — الاحتفاظ بكل صفوف الجدول الأيسر

library(dplyr)

orders <- data.frame(
  order_id = 1:4,
  customer_id = c(1, 2, 1, 99),
  amount = c(250, 180, 320, 90)
)

customers <- data.frame(
  customer_id = c(1, 2, 3),
  name = c('Alice','Bob','Carol')
)

# Order with customer_id=99 is kept, name = NA
left_join(orders, customers, by = 'customer_id')

right_join() — الاحتفاظ بكل صفوف الجدول الأيمن

library(dplyr)

orders <- data.frame(
  order_id = c(1, 2),
  customer_id = c(1, 2),
  amount = c(250, 180)
)

customers <- data.frame(
  customer_id = c(1, 2, 3),
  name = c('Alice','Bob','Carol')
)

# Carol (customer_id=3) has no orders => NA for order columns
right_join(orders, customers, by = 'customer_id')

full_join() — الاحتفاظ بكل الصفوف

library(dplyr)

q1 <- data.frame(product = c('A','B','C'), q1_sales = c(100, 200, 150))
q2 <- data.frame(product = c('A','C','D'), q2_sales = c(120, 160, 90))

# D appears only in q2, B only in q1 => both kept
full_join(q1, q2, by = 'product')

anti_join() — العثور على حالات عدم التطابق

library(dplyr)

all_employees <- data.frame(
  id = 1:5,
  name = c('Alice','Bob','Carol','Dave','Eve')
)

trained <- data.frame(id = c(1, 3, 5))

# Who has NOT completed training?
anti_join(all_employees, trained, by = 'id')

semi_join() — التصفية حسب التطابق

library(dplyr)

products <- data.frame(
  id = 1:5,
  name = c('Widget','Gadget','Donut','Gizmo','Sprocket'),
  price = c(9.99, 24.99, 1.99, 14.99, 4.99)
)

ordered_products <- data.frame(id = c(1, 3, 5))

# Products that appear in at least one order
semi_join(products, ordered_products, by = 'id')

الربط باستخدام أسماء أعمدة مختلفة

library(dplyr)

orders <- data.frame(
  order_id = 1:3,
  user_id = c(10, 20, 10),
  amount = c(100, 200, 150)
)

users <- data.frame(
  id = c(10, 20, 30),
  username = c('alice','bob','carol')
)

# orders.user_id matches users.id
left_join(orders, users, by = c('user_id' = 'id'))

الربط باستخدام مفاتيح متعددة

library(dplyr)

actual <- data.frame(
  year = c(2023, 2023, 2024, 2024),
  quarter = c('Q1','Q2','Q1','Q2'),
  sales = c(100, 120, 130, 150)
)

target <- data.frame(
  year = c(2023, 2023, 2024, 2024),
  quarter = c('Q1','Q2','Q1','Q2'),
  target = c(110, 115, 125, 145)
)

inner_join(actual, target, by = c('year', 'quarter'))

تسلسل عمليات ربط متعددة

library(dplyr)

orders <- data.frame(order_id=1:3, cust_id=c(1,2,1), prod_id=c(10,10,20))
customers <- data.frame(cust_id=1:2, name=c('Alice','Bob'))
products <- data.frame(prod_id=c(10,20), product=c('Widget','Gadget'))

orders %>%
  left_join(customers, by = 'cust_id') %>%
  left_join(products, by = 'prod_id')

التعامل مع الأعمدة المكررة

library(dplyr)

current <- data.frame(id=1:3, value=c(10,20,30), date=Sys.Date())
historic <- data.frame(id=1:3, value=c(8,22,28), date=Sys.Date()-365)

# Both have 'value' and 'date' columns
result <- inner_join(current, historic, by='id',
                     suffix=c('_current','_historic'))
print(names(result))

تحقق سريع

مراجعة: عمليات الربط في dplyr

library(dplyr)

# Chain of joins: the typical real-world pattern
orders <- data.frame(oid=1:3, cid=c(1,2,1), pid=c(10,20,10))
customers <- data.frame(cid=1:2, name=c('Alice','Bob'))
products <- data.frame(pid=c(10,20), name=c('Widget','Gadget'), price=c(9.99,19.99))

orders %>%
  inner_join(customers, by='cid') %>%
  inner_join(products, by='pid', suffix=c('','_product')) %>%
  select(oid, name, name_product, price)