Get Started in ≤1 Minute

# Before: from pyspark.sql import SparkSession
# Before: from pyspark.sql import functions as F
# Before: from pyspark.sql.window import Window

# After: only update the imports
from fuse_python.session import session
import fuse_python.functions as F
from fuse_python.window import Window

# Your existing PySpark code works unchanged
spark = session.builder.getOrCreate()

# Rest of your code stays exactly the same
# S3 support also built-in
data = [
    ("CA", "Los Angeles", 2023, 100),
    ("CA", "San Francisco", 2023, 80),
    ("NY", "New York City", 2023, 150),
    ("NY", "Buffalo", 2023, 70)
]

df = spark.createDataFrame(data, ["state", "city", "year", "incidents"])

window_spec = Window.partitionBy("state") \
                   .orderBy(F.desc("incidents")) \
                   .rowsBetween(Window.unboundedPreceding, Window.currentRow)

result = df.withColumn("running_total", F.sum("incidents").over(window_spec)) \
          .withColumn("rank", F.row_number().over(window_spec)) \
          .orderBy("state", F.desc("incidents"))

result.show()
                

from fuse_python.session import session
import fuse_python.functions as F
from fuse_python.window import Window

spark = session.builder.getOrCreate()

# Sample e-commerce data
data = [
    # (date, category, product, price, quantity, customer_id, country)
    ("2023-01-01", "Electronics", "Laptop", 1200.00, 1, "C1", "USA"),
    ("2023-01-01", "Electronics", "Phone", 800.00, 2, "C2", "Canada"),
    ("2023-01-02", "Books", "Python Guide", 50.00, 3, "C1", "USA"),
    ("2023-01-02", "Electronics", "Laptop", 1200.00, 1, "C3", "USA"),
    ("2023-01-03", "Books", "Data Science", 45.00, 2, "C2", "Canada"),
    ("2023-02-01", "Electronics", "Phone", 800.00, 1, "C4", "Mexico")
]

# Create DataFrame with multiple columns
df = spark.createDataFrame(data, [
    "date", "category", "product", "price", 
    "quantity", "customer_id", "country"
])

# Multiple transformations
result = df \
    .withColumn("date", F.to_date("date")) \
    .withColumn("month", F.date_format("date", "yyyy-MM")) \
    .withColumn("revenue", F.col("price") * F.col("quantity")) \
    .withColumn("discount", F.when(F.col("quantity") > 1, 0.1).otherwise(0)) \
    .withColumn("final_revenue", 
                F.col("revenue") * (1 - F.col("discount")))

# Complex windowing
customer_window = Window.partitionBy("customer_id") \
                       .orderBy("date") \
                       .rowsBetween(Window.unboundedPreceding, Window.currentRow)

category_window = Window.partitionBy("category", "month") \
                       .orderBy(F.desc("revenue"))

# Apply multiple window functions
analysis = result \
    .withColumn("customer_total_spend", 
                F.sum("final_revenue").over(customer_window)) \
    .withColumn("category_rank", 
                F.dense_rank().over(category_window)) \
    .withColumn("category_revenue_pct", 
                F.col("revenue") / F.sum("revenue").over(Window.partitionBy("category"))) \
    .withColumn("prev_purchase_date", 
                F.lag("date").over(Window.partitionBy("customer_id").orderBy("date")))

# Complex aggregation
summary = analysis \
    .groupBy("category", "month") \
    .agg(
        F.count("*").alias("total_sales"),
        F.sum("revenue").alias("total_revenue"),
        F.avg("revenue").alias("avg_revenue"),
        F.countDistinct("customer_id").alias("unique_customers"),
        F.collect_list("product").alias("products_sold")
    ) \
    .where(F.col("total_revenue") > 1000) \
    .orderBy(F.desc("total_revenue"))

# Show results
print("Detailed Sales Analysis:")
summary.show(truncate=False)

# Advanced filtering and selection
top_customers = analysis \
    .where((F.col("country").isin("USA", "Canada")) & 
           (F.col("category_rank") == 1)) \
    .select(
        "customer_id",
        "category",
        "month",
        F.round("customer_total_spend", 2).alias("total_spend"),
        F.datediff("date", "prev_purchase_date").alias("days_between_purchases")
    ) \
    .orderBy(F.desc("total_spend"))

print("\nTop Customers by Category:")
top_customers.show()