Cashflower

In this version, we've changed the name of the setting:

`OUTPUT_COLUMNS` is now called `OUTPUT_VARIABLES`.

We've also updated the default value:

From `[]` (empty list) to `None`.

**Rationale:**

In the model, we use the word "variable" (like with the `variable()` decorator). While "column" relates to table formats, it doesn’t cover all the ways we can show results. Changing the name to `OUTPUT_VARIABLES` helps keep things consistent throughout the model.

For the default value, using `OUTPUT_VARIABLES=None` is clearer because it means that no specific set of variables is chosen, so all variables will be included in the output. On the other hand, `OUTPUT_VARIABLES=[]` might suggest that no variables should be output at all, which could lead to confusion or extra checks in your code.

Bug fix

Fixed a bug related to the `OUTPUT_COLUMNS` setting.

The default settings have been updated. Now, only the main output file is saved by default, while the log and diagnostic files are disabled. Users can adjust these settings in the configuration file.

After:

python
settings = {
...
"SAVE_DIAGNOSTIC": False,
"SAVE_LOG": False,
"SAVE_OUTPUT": True,
...
}


Before:

python
settings = {
...
"SAVE_DIAGNOSTIC": True,
"SAVE_LOG": True,
"SAVE_OUTPUT": True,
...
}


This change focuses on the main output file, as the log and diagnostic files are helpful but not essential for cash flow modelling. By simplifying the default settings, the basic process becomes tidier.

**Important:** This version is **not** backward-compatible.

---

In this version, we've simplified the settings for grouping results. The `AGGREGATE` and `GROUP_BY_COLUMN` settings have been removed and replaced with `GROUP_BY`.

Here’s how the old settings map to the new one:

1. Results aggregated for the whole portfolio

Before:

python
AGGREGATE = True
GROUP_BY_COLUMN = None


After:

python
GROUP_BY = None


2. Results grouped by an attribute (e.g. product)

Before:

python
AGGREGATE = True
GROUP_BY_COLUMN = "product"


After:

python
GROUP_BY = "product"


3. Individual results (per model point)

Before:

python
AGGREGATE = False
GROUP_BY_COLUMN = None


After:

python
GROUP_BY = "id"

**Bug fixes:**

- **Fixed error in result aggregation** - resolved an issue where a shape mismatch could occur when calculating results if only some variables were included in the output. The `multiplier` is now correctly applied only to variables listed in `OUTPUT_COLUMNS`, preventing this error.