Examples

These examples are checked into the PyPlyne repository. If you installed PyPlyne in your own project, copy the snippets you want to try into a local .pyplyne file and run that file with uv run pyplyne.

If you cloned this repository, run the examples from the repository root:

Filter And Map A Sequence

Start here for a minimal sequence pipeline: annotate a list with seq, use _ in filter and map, and print the resulting list.

uv run pyplyne examples/list_pipeline.pyplyne

Shows:

• runnable sequence pipeline
• seq annotation
• _ placeholder expressions
• filter()
• map()
• print() result

examples/list_pipeline.pyplyne

numbers = seq [1, 2, 3, 4, 5, 6]

result = numbers
  |> filter(_ % 2 == 0)
  |> map(_ * 10)

print(result)

Transform Rows In A Table

Start here for table data: declare inline rows with df, filter by column names, add a computed column, and project selected fields.

uv run pyplyne examples/tabular_pipeline.pyplyne

Shows:

• runnable table pipeline
• df annotation
• where()
• mutate()
• select()
• computed columns

examples/tabular_pipeline.pyplyne

rows = df [
  {"region": "north", "amount": 120},
  {"region": "south", "amount": 80},
  {"region": "north", "amount": 220},
]

large_sales = rows
  |> where(amount > 100)
  |> mutate(double=amount * 2)
  |> select(region, amount, double)

print(large_sales)

Wrap A Polars DataFrame

Use this when Python code already creates a Polars DataFrame: import polars as pl, mark the object as df, then continue with PyPlyne table verbs.

uv run pyplyne examples/polars_constructor.pyplyne

Shows:

• Python imports
• Polars interop
• pl.DataFrame construction
• df annotation
• table verbs

examples/polars_constructor.pyplyne

import polars as pl

sales = df pl.DataFrame([
  {"region": "north", "amount": 120},
  {"region": "south", "amount": 80},
  {"region": "north", "amount": 220},
])

large_sales = sales
  |> where(amount > 100)
  |> select(region, amount)

print(large_sales)

Convert Tables To Rows And Back

Use this when a table workflow needs row-wise sequence logic: convert a filtered table to rows, map or edit records, then rebuild a table.

uv run pyplyne examples/shape_conversions.pyplyne

Shows:

• shape conversion
• to_rows()
• sequence map
• set_fields()
• to_table()
• arrange()

examples/shape_conversions.pyplyne

sales = df [
  {"region": "north", "amount": 120},
  {"region": "south", "amount": 80},
  {"region": "north", "amount": 220},
]

labels = sales
  |> where(amount > 100)
  |> to_rows()
  |> map(_["region"])

reviewed = sales
  |> where(amount > 100)
  |> to_rows()
  |> set_fields(reviewed=True)
  |> to_table()
  |> arrange(region)

print(labels)
print(reviewed)

Edit Record Fields

Use this for JSON-like row dictionaries: derive new fields, remove debug-only fields, and keep the final shape explicit.

uv run pyplyne examples/record_fields.pyplyne

Shows:

• record dictionaries
• set_fields()
• drop_fields()
• keep_fields()
• field projection

examples/record_fields.pyplyne

rows = seq [
  {"region": "north", "amount": 120, "discount": 10, "debug": "a"},
  {"region": "south", "amount": 80, "discount": 5, "debug": "b"},
  {"region": "north", "amount": 220, "discount": 20, "debug": "c"},
]

with_fields = rows
  |> set_fields(
    net=amount - discount,
    reviewed=amount > 100,
    label=region + "-" + str(amount),
  )

without_debug = with_fields
  |> drop_fields(debug)

projected = without_debug
  |> keep_fields(region, amount, net, reviewed, label)

print(with_fields)
print(without_debug)
print(projected)

Run The Full Language Tour

Use this as an end-to-end smoke test: it combines imports, sequence and table verbs, grouped summaries, deferred pipelines, shape conversion, and CSV output.

uv run pyplyne examples/full_language_tour.pyplyne

Shows:

• end-to-end tour
• imports
• sequence verbs
• table verbs
• group_by() and summarize()
• defer pipeline
• write_csv() output

examples/full_language_tour.pyplyne

from pathlib import Path

numbers = seq [1, 2, 3, 4, 5, 6]

sequence_total = numbers
  |> filter(_ > 2)
  |> map(x => x * 10)
  |> reduce(_1 + _2)

rows = df [
  {"region": "north", "amount": 120, "discount": 10},
  {"region": "south", "amount": 80, "discount": 5},
  {"region": "north", "amount": 220, "discount": 20},
  {"region": "south", "amount": 180, "discount": 15},
]

summary = rows
  |> where(amount > 100)
  |> mutate(net=amount - discount)
  |> group_by(region)
  |> summarize(
    total=sum(net),
    average=mean(net),
    smallest=min(net),
    largest=max(net),
    rows=count(),
  )
  |> arrange(region)

plan = defer rows
  |> where(amount > 100)
  |> select(region, amount)

summary_rows = summary |> to_rows()
summary_table = summary_rows |> to_table()
reviewed_sales = rows
  |> where(amount > 100)
  |> to_rows()
  |> set_fields(reviewed=True)
  |> to_table()
  |> arrange(region)

output_path = Path("examples/full_language_tour_output.csv")
summary_table |> write_csv(str(output_path))

print(sequence_total)
print(summary)
print(plan)
print(summary_rows)
print(reviewed_sales)

Generated Output

The full tour writes examples/full_language_tour_output.csv. The generated file is ignored by Git.

Choosing An Example

Goal	Example
Filter And Map A Sequence	examples/list_pipeline.pyplyne
Transform Rows In A Table	examples/tabular_pipeline.pyplyne
Wrap A Polars DataFrame	examples/polars_constructor.pyplyne
Convert Tables To Rows And Back	examples/shape_conversions.pyplyne
Edit Record Fields	examples/record_fields.pyplyne
Run The Full Language Tour	examples/full_language_tour.pyplyne

Source-backed examples

This page is generated from site/data/example-catalog.json and the actual files in examples/. Update the runnable example file first, then regenerate this page with npm run docs:examples from site/.