The data contract¶

This page lists what sources may hold and what attaching refuses and accepts, for anyone putting data on a spec. sources supplies the numbers, keyed by the names the spec declares:

import specsolve as sps

result = sps.solve(
    'dispatch.yaml',
    {'load': 'load.parquet', 'cost': cost_frame, 'p_max': p_max_frame},
)

Getting from the files an instance arrives in to these shapes is preparing the data.

What a parameter accepts¶

For a parameter declared dims: [d1, d2], the value under its key is one of:

a parquet path;
a table exposing the Arrow PyCapsule protocol (polars, pandas, pyarrow, duckdb) with columns d1, d2, value;
an int or float, one value for every coordinate the parameter covers;
a dict of label to value, for a parameter over one dimension;
a sequence (list, tuple, np.ndarray) for a parameter over one dimension, positional against that dimension's index.

The last three shapes serve models written out in Python. Each is dense, and attach materialises it: one number over (snapshot, generator) becomes one row per pair. Declare a constant as dims: [] instead. A sequence carries no labels, so its dimension's index comes from one of the three sources under where coordinates come from.

A pd.Series is unwrapped first. Its one dimension is its index rather than a column. Attach unwraps it only if pandas is already imported. An unnamed index attaches to the declared dimension. A named index attaches by that name, and a name outside the declared dimensions raises.

A MultiIndex is refused. A parameter over two dimensions arrives as a table with both as columns. series.reset_index() is the whole change.

An xr.DataArray is refused. Pass array.to_series().reset_index(). Result.to_dataarray() is the way back out.

Nothing on this path imports pandas, xarray or linopy on your behalf.

Where coordinates come from¶

Each dimension's index is resolved before any parameter loads, from a key in sources named after the dimension. That key holds a table with a column of that name, a parquet path, or a bare sequence of the labels. The first occurrence of each label is its position, and that order is what shift reads positionally.

A dimension nothing supplies raises. Attach never reads labels out of the parameters. Which labels an axis has is data's to say, and that rule is the language's.

A relation goes under its own name, as a table of the rows it has, one column per column it declares. Attach reads every column against the labels its dimension's index supplied: a key no row mentions is unmapped, and a value matching no label is refused as a typo. A keyed relation holds one row per key tuple, however many columns the key names; a bare relation holds each row at most once.

What attaching refuses and accepts¶

A coordinate has a value, or it has no row. Attach refuses a row whose value is null or NaN. Polars and parquet write a hole as a null, pandas has only NaN, and None in a pandas column is NaN by the time either lane sees it.

Refused¶

What arrives	What the message says
a declared parameter with no data	names the parameter
a source nothing can be read as a table from	names the shapes that are read
an `xr.DataArray`	names `to_series().reset_index()`
a `pd.Series` with a `MultiIndex`	names the table and the `reset_index()` that gets there
a `dims: []` parameter whose source has more than one row	one value broadcast everywhere has one row
a dict or a sequence for a parameter over more than one dimension	each runs along one dimension
a sequence whose length is not the dimension's	positional, so one entry per label
a sequence for a dimension nothing else supplies labels for	names the three ways to supply them
a key naming neither a parameter, a dimension nor a relation	names the near miss
a relation table short of a column it declares	names them, and what each is
a relation table with a null in any column	a relation is partial by omitting a row
a relation table mapping one key twice, or relating one tuple twice	a keyed relation holds one row per key, a bare one each row once
a map with both authors, or neither	names them, and says which way out
an index carrying a column named after a relation with a column over it	names the key it belongs under
a table missing a declared dimension column, or `value`	names the columns needed
a `value` column carrying a null or a NaN	names the parameter and the coordinates
a label outside the dimension's index	names the parameter and the strays
two rows for one coordinate
a relation with two values for one key
a relation value that is not a label of its column's dimension	one wording, checked once for both lanes
a dimension carrying relations with no index
a dimension nothing can supply labels for	names both ways to fix it
a dimension the spec declares and the caller also supplies	names the declaration and the colliding key
a relation whose map the spec declares and the caller also supplies	names the map and the colliding column
a declared map whose labels nothing supplies	names the map, and asks only for the labels
a declared map keyed by something the labels do not carry	names the relation and the strays
a column that is not the declared `dtype`	names both, and the declaration the data would satisfy
a divisor with no value where the spec divides by it	names the parameter and how many rows (absence)
a comparison's whole constant side with no value where the row is built	the same, naming the constraint
a bound parameter with no value where the variable exists	names both models the two repairs build

Accepted¶

What arrives	What happens
an undeclared column in a table	ignored
a coordinate with no row	sparse variables; what a missing row means where it is read is absence. `diagnostics().sparse_parameters` names the parameters that arrived short of their dims (api)
a value that is readable and wrong	bound as given

Stray labels¶

The index is what makes a stray label a stray.

cost = {'wind': 1.0, 'gsa': 2.0}  # 'gas' misspelled — refused by name

A dimension whose labels came from the parameters would read gsa as a third generator.

Growing or replacing the data¶

A built model takes new numbers with update. A sweep over slices of one dimension is solve_over. Both attach through the rules above.

The linopy lane attaches by these same rules, refusals included, held to them by tests/test_data_parity.py. The same malformed source gets the same verdict, and where one defect has one repair, the same message.