Line Items & Data Access¶

Line item types¶

`FixedLine`¶

Hardcoded values, one per period. The simplest line item — use it for revenue assumptions, headcount, or any input that you control directly.

from pyproforma import FixedLine

revenue = FixedLine(
    values={2024: 500_000, 2025: 550_000, 2026: 605_000},
    label="Revenue",
    tags=["operating"],
)

You can also supply a value_format to control how numbers appear in tables and charts:

from pyproforma import Format

revenue = FixedLine(
    values={2024: 500_000, 2025: 550_000},
    label="Revenue",
    value_format=Format.CURRENCY_NO_DECIMALS,
)

`FormulaLine`¶

Calculated from other line items or assumptions. The formula is a lambda that receives a unified namespace li and the current period t:

from pyproforma import FormulaLine

gross_profit = FormulaLine(
    formula=lambda li, t: li.revenue[t] - li.cogs[t],
    label="Gross Profit",
)

`Assumption`¶

A scalar constant — not period-specific. Accessed in formulas as a plain attribute (no [t]):

from pyproforma import Assumption

tax_rate = Assumption(value=0.21, label="Tax Rate")

# In a formula:
net_income = FormulaLine(
    formula=lambda li, t: li.ebit[t] * (1 - li.tax_rate),
    label="Net Income",
)

`InputLine`¶

Like FixedLine but the period values are supplied at instantiation:

from pyproforma import InputLine

class Model(ProformaModel):
    revenue = InputLine(label="Revenue")

model = Model(periods=[2024, 2025], revenue={2024: 500_000, 2025: 550_000})

Debt lines¶

create_debt_lines generates a pair of DebtPrincipalLine and DebtInterestLine from loan parameters:

from pyproforma import create_debt_lines

principal, interest = create_debt_lines(
    name_prefix="loan",
    principal=1_000_000,
    rate=0.06,
    term=5,
    start_period=2024,
)

The formula namespace¶

Every formula receives two arguments: li and t.

t — the current period (an int, e.g. 2024)
li — a unified namespace giving access to both line items and assumptions

formula=lambda li, t: ...

Accessing line items¶

Line items are period-indexed — always pass [t]:

formula=lambda li, t: li.revenue[t] * 0.55
formula=lambda li, t: li.revenue[t] - li.cogs[t]

Accessing assumptions¶

Assumptions are scalars — no [t]:

formula=lambda li, t: li.ebit[t] * (1 - li.tax_rate)

Prior-period references¶

Use t-1 to reference the previous period. Add a seed value on the FixedLine or FormulaLine itself to provide the base year:

revenue = FormulaLine(
    formula=lambda li, t: li.revenue[t-1] * (1 + li.growth_rate),
    values={2024: 500_000},   # seed: used as the t-1 value for 2025
    label="Revenue",
)

Tag sums in formulas¶

li.tag["name"][t] returns the sum of all line items tagged "name" for period t:

total_operating = FormulaLine(
    formula=lambda li, t: li.tag["operating"][t],
    label="Total Operating Income",
)

Tags¶

Tags let you group line items flexibly without fixed categories. A line item can have multiple tags.

revenue = FixedLine(values={...}, tags=["revenue", "operating"])
other_income = FixedLine(values={...}, tags=["revenue"])

Access a tag group from the model:

model.tag["revenue"]           # → LineItemSelection
model.tag["revenue"].names     # → ["revenue", "other_income"]
model.tag["revenue"].sum(2024) # → sum for period 2024

Tags are also usable in table row types — see Tables.

Accessing data from an instantiated model¶

Subscript access¶

result = model["revenue"]        # → LineItemResult
result[2024]                     # → 500000.0
result.value(2024)               # same thing
result.values                    # → {2024: 500000.0, 2025: 550000.0, ...}
result.label                     # → "Revenue"
result.tags                      # → ["operating"]
result.value_format              # → NumberFormatSpec(...)

For assumptions:

result = model["tax_rate"]       # → AssumptionResult
result.value                     # → 0.21

Direct namespace access¶

model.li and model.av give lower-level access without creating result wrapper objects:

model.li.revenue[2024]    # → 500000.0  (LineItemValues namespace)
model.av.tax_rate         # → 0.21      (AssumptionValues namespace)

Convenience shortcuts¶

model["revenue"].table()   # generate a table for this item
model["revenue"].chart()   # generate a chart for this item

Value formatting¶

Every line item has a value_format (NumberFormatSpec). When not specified it defaults to Format.NO_DECIMALS. Formats flow through to tables and chart y-axis labels automatically.

Format constant	Example output
`Format.NO_DECIMALS`	`500,000`
`Format.TWO_DECIMALS`	`500,000.00`
`Format.CURRENCY`	`$500,000.00`
`Format.CURRENCY_NO_DECIMALS`	`$500,000`
`Format.PERCENT`	`21%`
`Format.PERCENT_ONE_DECIMAL`	`21.0%`
`Format.PERCENT_TWO_DECIMALS`	`21.00%`
`Format.THOUSANDS`	`500` (value divided by 1,000)
`Format.MILLIONS`	`0.5` (value divided by 1,000,000)
`Format.THOUSANDS_K`	`500K`
`Format.MILLIONS_M`	`0.5M`

For custom formats, use NumberFormatSpec directly:

from pyproforma import NumberFormatSpec

fmt = NumberFormatSpec(decimals=1, scale="millions", suffix="M", thousands=False)
# 1_500_000 → "1.5M"

NumberFormatSpec parameters: decimals, thousands (bool), prefix, suffix, multiplier, scale ("thousands" or "millions").