Skip to content

API Reference

Jump to bottom
Yasser Mustafa edited this page Feb 15, 2026 · 1 revision

API Reference

Complete reference for all PipeFrame functions and classes.

Table of Contents

Core Classes

DataFrame

PipeFrame's enhanced DataFrame class.

from pipeframe import DataFrame

df = DataFrame(data, columns=None, index=None)

Arguments:

  • data: dict, array, or pandas DataFrame
  • columns: column labels (optional)
  • index: row labels (optional)

Returns: PipeFrame DataFrame (fully compatible with pandas)

Example:

df = DataFrame({'x': [1, 2, 3], 'y': [4, 5, 6]})

Series

Enhanced Series class with pipe support.

from pipeframe import Series

s = Series(data, index=None, name=None)

Data Selection

select()

Select columns from DataFrame.

df >> select(*columns)

Arguments:

  • *columns: Column names to select

Special patterns:

  • starts_with('prefix'): Columns starting with prefix
  • ends_with('suffix'): Columns ending with suffix
  • contains('text'): Columns containing text
  • matches('regex'): Columns matching regex
  • -column: Exclude column

Examples:

# Select specific columns
df >> select('name', 'age', 'salary')

# Select with patterns
df >> select(starts_with('sales_'))

# Exclude columns
df >> select('-temp', '-id')

Data Filtering

filter() / where()

Filter rows based on conditions.

df >> filter(condition)
df >> where(condition)  # Alias

Arguments:

  • condition: String expression for filtering

Examples:

# Simple condition
df >> filter('age > 25')

# Multiple conditions with &
df >> filter('age > 25 & salary > 50000')

# Or conditions with |
df >> filter('department == "Sales" | department == "IT"')

# String operations
df >> filter('name.str.startswith("A")')
df >> filter('city.str.contains("York")')

# IN operator
df >> filter('status in ["Active", "Pending"]')

Data Transformation

define() / mutate()

Create or modify columns.

df >> define(**kwargs)
df >> mutate(**kwargs)  # Alias

Arguments:

  • **kwargs: name=expression pairs

Examples:

# Create new column
df >> define(bonus='salary * 0.1')

# Multiple columns
df >> define(
    bonus='salary * 0.1',
    total='salary + bonus',
    is_senior='age > 35'
)

# Complex expressions
df >> define(
    category='''
        "High" if amount > 1000 else
        "Medium" if amount > 500 else
        "Low"
    '''
)

drop()

Drop columns.

df >> drop(*columns)

Examples:

df >> drop('temp_column')
df >> drop('col1', 'col2', 'col3')

rename()

Rename columns.

df >> rename(mapper, **kwargs)

Examples:

# Using dict
df >> rename({'old_name': 'new_name'})

# Using kwargs
df >> rename(old_name='new_name')

Grouping & Aggregation

group_by()

Group DataFrame by columns.

df >> group_by(*columns)

Examples:

df >> group_by('department')
df >> group_by('department', 'location')

summarize() / summarise()

Aggregate grouped data.

grouped_df >> summarize(**kwargs)
grouped_df >> summarise(**kwargs)  # British spelling

Arguments:

  • **kwargs: name='aggregation_function(column)' pairs

Common aggregations:

  • mean(column): Average
  • sum(column): Total
  • count(): Count rows
  • min(column): Minimum
  • max(column): Maximum
  • std(column): Standard deviation
  • nunique(column): Count unique values

Examples:

result = (df
    >> group_by('department')
    >> summarize(
        avg_salary='mean(salary)',
        total_employees='count()',
        max_salary='max(salary)'
    )
)

count()

Count observations.

df >> count(*columns)

Examples:

# Count all rows
df >> count()

# Count by group
df >> count('department')
df >> count('department', 'location')

Joining & Combining

left_join()

Left outer join.

df >> left_join(right, on=None, left_on=None, right_on=None)

right_join()

Right outer join.

df >> right_join(right, on=None, left_on=None, right_on=None)

inner_join()

Inner join.

df >> inner_join(right, on=None, left_on=None, right_on=None)

full_join()

Full outer join.

df >> full_join(right, on=None, left_on=None, right_on=None)

Examples:

# Join on common column
result = orders >> left_join(customers, on='customer_id')

# Join on different column names
result = orders >> left_join(
    customers, 
    left_on='cust_id', 
    right_on='id'
)

bind_rows()

Concatenate DataFrames vertically.

df >> bind_rows(other)

bind_cols()

Concatenate DataFrames horizontally.

df >> bind_cols(other)

Sorting & Ordering

arrange() / order_by()

Sort DataFrame by columns.

df >> arrange(*columns)
df >> order_by(*columns)  # Alias

Arguments:

  • *columns: Column names (prefix with - for descending)

Examples:

# Ascending
df >> arrange('age')

# Descending (use - prefix)
df >> arrange('-salary')

# Multiple columns
df >> arrange('department', '-salary')

Reshaping

pivot_wider()

Pivot from long to wide format.

df >> pivot_wider(id_cols, names_from, values_from)

pivot_longer()

Pivot from wide to long format.

df >> pivot_longer(cols, names_to='name', values_to='value')

Examples:

# Wide to long
long_df = (wide_df
    >> pivot_longer(
        cols=['Q1', 'Q2', 'Q3', 'Q4'],
        names_to='quarter',
        values_to='sales'
    )
)

# Long to wide
wide_df = (long_df
    >> pivot_wider(
        id_cols='product',
        names_from='quarter',
        values_from='sales'
    )
)

Utilities

head()

Return first n rows.

df >> head(n=5)

tail()

Return last n rows.

df >> tail(n=5)

sample()

Random sample of rows.

df >> sample(n=None, frac=None)

Examples:

# Sample 10 rows
df >> sample(n=10)

# Sample 10% of rows
df >> sample(frac=0.1)

distinct()

Remove duplicate rows.

df >> distinct(*columns)

Examples:

# Remove all duplicate rows
df >> distinct()

# Remove duplicates based on columns
df >> distinct('customer_id', 'date')

drop_na()

Drop rows with missing values.

df >> drop_na(*columns)

Examples:

# Drop rows with any NA
df >> drop_na()

# Drop rows with NA in specific columns
df >> drop_na('critical_column')

fill_na()

Fill missing values.

df >> fill_na(value, **kwargs)

Examples:

# Fill all NAs with 0
df >> fill_na(0)

# Fill specific columns
df >> fill_na(age=0, salary=50000)

peek()

View intermediate results (for debugging).

df >> peek(n=5)

Example:

result = (df
    >> filter('age > 25')
    >> peek(n=3)  # Shows first 3 rows
    >> define(bonus='salary * 0.1')
    >> peek()  # Shows first 5 rows
)

Column Selectors

starts_with()

Select columns starting with prefix.

df >> select(starts_with('sales_'))

ends_with()

Select columns ending with suffix.

df >> select(ends_with('_total'))

contains()

Select columns containing text.

df >> select(contains('amount'))

matches()

Select columns matching regex pattern.

df >> select(matches(r'\\d{4}'))  # Columns with 4 digits

More Information

Need help? Open an issue https://github.com/Yasser03/pipeframe/issues

Clone this wiki locally