# pudl.extract.rus12

Extract USDA RUS Form 12 data from CSVs.

## Attributes

| [`raw_rus12__all_dfs`](#pudl.extract.rus12.raw_rus12__all_dfs)   |    |
|-----------------------------------------------------------------------|----|
| [`raw_rus12_assets`](#pudl.extract.rus12.raw_rus12_assets)     |    |

## Classes

| [`Extractor`](#pudl.extract.rus12.Extractor)   | Extractor for USDA RUS Form 12.   |
|--------------------------------------------------------------|-----------------------------------|

## Functions

| [`raw_rus12_asset_factory`](#pudl.extract.rus12.raw_rus12_asset_factory)(table_name)   | Create raw RUS 12 asset for a specific page.   |
|----------------------------------------------------------------------------------------|------------------------------------------------|

## Module Contents

### *class* pudl.extract.rus12.Extractor(\*args, \*\*kwargs)

Bases: [`pudl.extract.csv.CsvExtractor`](../csv/index.html.md#pudl.extract.csv.CsvExtractor)

Extractor for USDA RUS Form 12.

#### METADATA

Instance of metadata object to use with this extractor.

#### READ_CSV_KWARGS

Keyword arguments that are passed to `pandas.read_csv()`.

These allow customization of the CSV parsing process. For example, you can specify
the column delimiter, data types, date parsing, etc. This can greatly reduce peak
memory usage and speed up the extraction process. Unfortunately you must refer to
the column headers using their original names as they appear in the CSV.

TODO[zaneselvans] 2024-04-19: it would be useful to be able to specify different CSV
reading options for different pages within the same dataset. At the moment the same
arguments will be applied to all pages. This still allows some flexibility because
some `pandas.read_csv()` arguments like `dtype` don’t raise errors if the
columns they apply to aren’t present.

#### source_filename(page: [str](https://docs.python.org/3/builtins/stdtypes.html#str), \*\*partition: [pudl.extract.extractor.PartitionSelection](../extractor/index.html.md#pudl.extract.extractor.PartitionSelection)) → [str](https://docs.python.org/3/builtins/stdtypes.html#str)

Get the file name for the right page and part.

In this instance we are using the same methodology from the excel metadata extractor.

#### process_raw(df, page, \*\*partition)

Adds source column and report_year column if missing.

#### load_source(page: [str](https://docs.python.org/3/builtins/stdtypes.html#str), \*\*partition: [pudl.extract.extractor.PartitionSelection](../extractor/index.html.md#pudl.extract.extractor.PartitionSelection)) → [pandas.DataFrame](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.html#pandas.DataFrame)

Produce the dataframe object for the given partition.

In this instance we need to specify both year and geography to get the right zipfile.
In 2010 there is a zipfile at the federal and state level, but for now we only want
to use the federal-level data.

* **Parameters:**
  * **page** – pudl name for the dataset contents, eg “boiler_generator_assn” or
    “data”
  * **partition** – partition to load. Examples:
    {‘year’: 2009}
    {‘year_month’: ‘2020-08’}
* **Returns:**
  pd.DataFrame instance containing CSV data

### pudl.extract.rus12.raw_rus12_\_all_dfs

### pudl.extract.rus12.raw_rus12_asset_factory(table_name: [str](https://docs.python.org/3/builtins/stdtypes.html#str))

Create raw RUS 12 asset for a specific page.

### pudl.extract.rus12.raw_rus12_assets