a parameter has one shape, while Calliope lets the data decide whether a limit is per technology, per node or per group

[FBumann]

Prompt: "Let’s add one parent issue for caliope and file focused issues for individual topics or link existing ones."

Note

The following content was generated by AI.

Calliope reads annual_flow_max and net_import_share at whatever shape the data gives. A mathspec parameter has one declared shape. The recommendation: refuse this by design, and add a row to limits.md that says to declare one parameter per shape.

Case, why refuse, what to change

Case. In Calliope v0.7.0, annual_energy_balance.yaml reads annual_flow_max per technology, per node, and over a group. net_import_share.yaml does the same with its share. The data decides which shape applies. The port in #770 declares one shape each. The group row reads a separate scalar, annual_flow_max_group, and net_import_share is one number.

Why refuse. A shape that the data decides is a declaration that depends on the data. limits.md already refuses "declarations that depend on the data", because to_spec could then no longer read the file without the data. A parameter per shape states the same model with the shape visible in the file. The port's form is the correct one, not a workaround.

What to change.

• Add a row to limits.md → Deliberate non-primitives: "One parameter read at several shapes. Why refused: the data would decide the declaration. Instead: one parameter per shape, each read by its own row."
• In docs(calliope): all of calliope's math is a set of fragments that merge, with its modes laid over them as patches #770, move gap 6 from "what mathspec is missing" to "out by design".

If a maintainer sees a case that one parameter per shape cannot state, that case reopens the question.