-
Currently pandas tries to infer data types instead of using the data types stored in the database during `read_sql()` -- in the output routines, we need to ensure that we enforce the correct data type…
-
For the environment we are building, pip install catalystcoop.pudl resulted in these messages:
```
ERROR: pip's dependency resolver does not currently take into account all the packages that are i…
-
### Describe the bug
I found a number of generator records that are showing up in an annual pudl output object that have nulls in the `utility_id_eia` column. These should all be consistent because t…
-
We've got veritable 🤡 🚗 of EIA fuel type codes categorized in the current `pudl.constants` module, with a different dictionary for each table in which they appear, so we can simplify them when we crea…
-
A review of the field metadata in `src/pudl/package_data/meta/datapkg/datapackage.json` reveals the following inconsistencies in `constraints.enum` and `description` for fields of the same name across…
-
The `boiler_fuel_eia923` should have a natural primary key of:
* `report_date`
* `plant_id_eia`
* `boiler_id`
* `fuel_type_code`
However, there are 4032 rows out of 1.14 million that ac…
-
After simplifying our test suite setup (issue #942) I ran the data validation tests, to make sure they still worked with the new setup. There were a few tables with more rows than expected because (I…
-
### Describe the bug
Running the `pudl_etl` script for the fast ETL results in a KeyError.
### Bug Severity
How badly is this bug affecting you?
- **High:** This bug is preventing me from using…
-
PUDL DB tables which lack a natural primary key, and thus have no primary key declared in the new metadata structures, end up not having an implicit primary key either when they are created in SQLite.…
-
Fixing the `merge_on_date_year` issues has the potential to affect many of the MCOE and EIA 923 outputs. Do sanity checks and visualizations to make sure nothing is amiss, and if possible integrate th…