인덱스 정렬과 재인덱싱

import pandas as pd

s1 = pd.Series([10, 20, 30], index=['a', 'b', 'c'])
s2 = pd.Series([1, 2, 3], index=['b', 'c', 'd'])

# Alignment in action: a→NaN, d→NaN
result = s1 + s2
print('s1 + s2 with automatic alignment:')
print(result)

import pandas as pd

s1 = pd.Series({'a': 10, 'b': 20, 'c': 30})
s2 = pd.Series({'b': 1, 'c': 2, 'd': 3})

# Default: NaN where labels don't match
print('Default (NaN for unmatched):')
print(s1 + s2)

# fill_value=0: treat missing as zero
print('\nWith fill_value=0:')
print(s1.add(s2, fill_value=0))

import pandas as pd

df1 = pd.DataFrame({'revenue': [100, 200], 'cost': [80, 150]}, index=['Jan', 'Feb'])
df2 = pd.DataFrame({'revenue': [120, 210], 'headcount': [5, 6]}, index=['Feb', 'Mar'])

result = df1 + df2
print('Combined DataFrame (union of index and columns):')
print(result)

import pandas as pd

s = pd.Series({'a': 10, 'b': 20, 'c': 30})

# Reindex to a new label set
reindexed = s.reindex(['b', 'c', 'd', 'e'])
print('Original:', s.index.tolist())
print('New index: [b, c, d, e]')
print('\nReindexed Series:')
print(reindexed)
# b, c preserved; d, e → NaN; a dropped

import pandas as pd

# Sparse daily data with gaps
s = pd.Series(
    [10, 20, 30],
    index=pd.to_datetime(['2024-01-01', '2024-01-03', '2024-01-07'])
)

# Reindex to complete date range
full_range = pd.date_range('2024-01-01', '2024-01-07')
print('Forward fill (carry last known value):')
print(s.reindex(full_range, method='ffill'))

print('\nFill with 0:')
print(s.reindex(full_range, fill_value=0))

import pandas as pd

df = pd.DataFrame({
    'name': ['Alice', 'Bob'],
    'age': [30, 25],
    'city': ['NY', 'LA']
})

# Enforce a canonical column order and add missing columns
canonical_cols = ['name', 'age', 'email', 'city', 'country']
df_reindexed = df.reindex(columns=canonical_cols)
print(df_reindexed)
# 'email' and 'country' are new → NaN

import pandas as pd

s1 = pd.Series({'a': 10, 'b': 20, 'c': 30})
s2 = pd.Series({'b': 1, 'c': 2, 'd': 3})

# Align to common labels only (inner join)
s1_aligned, s2_aligned = s1.align(s2, join='inner')
print('Inner-aligned s1:')
print(s1_aligned)
print('Inner-aligned s2:')
print(s2_aligned)

print('\nProduct on common labels:')
print(s1_aligned * s2_aligned)

import pandas as pd

# Two DataFrames sharing the same index
revenue = pd.Series({'North': 500, 'South': 300, 'East': 450})
costs = pd.Series({'North': 350, 'South': 280, 'West': 400})

# Automatic alignment: 'East' and 'West' don't match → NaN
profit = revenue - costs
print('Profit by region:')
print(profit)

# If you want only common regions
profit_inner = revenue.align(costs, join='inner')[0] - revenue.align(costs, join='inner')[1]
print('\nProfit (common regions only):')
print(profit_inner)

import pandas as pd

df1 = pd.DataFrame({'sales': [100, 200, 300]}, index=['Q1', 'Q2', 'Q3'])
df2 = pd.DataFrame({'cost': [80, 160, 240]}, index=['Q1', 'Q2', 'Q3'])

# Check index equality
if (df1.index == df2.index).all():
    print('Indices match — safe to combine.')
    combined = pd.concat([df1, df2], axis=1)
    combined['profit'] = combined['sales'] - combined['cost']
    print(combined)
else:
    print('Indices differ — align before combining!')

import pandas as pd

# Two sales batches with different product sets
batch1 = pd.Series({'laptop': 50, 'phone': 80, 'tablet': 30})
batch2 = pd.Series({'phone': 90, 'tablet': 40, 'headphones': 60})

# Full product catalogue
all_products = ['laptop', 'phone', 'tablet', 'headphones']

# Standardise both to the full catalogue
b1 = batch1.reindex(all_products, fill_value=0)
b2 = batch2.reindex(all_products, fill_value=0)

total = b1 + b2
print('Total sales per product:')
print(total)

import pandas as pd

# Duplicate label alignment — produces unexpected expansion
s1 = pd.Series([1, 2, 3], index=['a', 'a', 'b'])
s2 = pd.Series([10, 20], index=['a', 'b'])

result = s1 + s2
print('Result with duplicate indices (unexpected expansion):')
print(result)
print('\nAlways ensure unique indices before arithmetic!')