Add documentation for scripts

docs/scripts.md

@@ -0,0 +1,104 @@
+# Script Documentation
+
+## Summary
+
+Some tasks are small enough that the project architecture should not change, but the large enough that they should not be performed by hand.
+Files in the `scripts` directory exist to fill this space.
+
+Currently, the following scripts are provided.
+
+* `get_counts.py`
+    * get docket, document, and comment counts from regulations.gov, a mirrulations dashboard, or a mirrulations Redis instance as json
+    * when using regulations.gov a timestamp can be given to make all dockets, documents, and comments before the timestamp count as if they were downloaded
+* `correct_counts.py`
+    * correct possible errors within a counts json file generated by `get_counts.py`
+* `set_counts.py`
+    * set values in a mirrulations Redis instance using json generated by `get_counts.py`
+
+All of the scripts above share a common format
+<details>
+<summary><code>get_counts.py</code> common format</summary>
+
+```json
+{
+  "creation_timestamp": "2024-10-16 15:00:00",
+  "dockets": {
+    "downloaded": 253807,
+    "jobs": 0,
+    "total": 253807,
+    "last_timestamp": "2024-10-13 04:04:18"
+  },
+  "documents": {
+    "downloaded": 1843774,
+    "jobs": 0,
+    "total": 1843774,
+    "last_timestamp": "2024-10-13 04:04:18"
+  },
+  "comments": {
+    "downloaded": 22240501,
+    "jobs": 10,
+    "total": 22240511,
+    "last_timestamp": "2024-10-13 04:04:18"
+  }
+}
+```
+
+</details>
+
+## Description
+
+### `get_counts.py`
+
+`get_counts.py` gets counts from one of three sources: regulations.gov, a Mirrulations Redis instance, a Mirrulations dashboard via HTTP.
+
+When reading from regulations.gov a UTC timestamp can be specified to mock having downloaded all dockets, documents, and comments from before that timestamp.
+
+When reading from a dashboard a UTC timestamp must be specified since the dashboard API does not provide one.
+
+### `correct_counts.py`
+
+`correct_counts.py` corrects counts from `get_counts.py` using one of two strategies: set downloaded counts for a type to the minimum of `downloaded` and `total` for that type, or set downloaded counts to the minimum of `total -jobs` and `downloaded`.
+By default any queued jobs will cause the script to exit and output nothing, this behavior can be changed with the `--ignore-queue` flag.
+
+### `set_counts.py`
+
+`set_counts.py` sets values from `get_counts.py` in a Redis instance.
+By default the script will prompt for user input before changing any values.
+This behavior can be changed using the `--yes` flag, which should be used **WITH GREAT CARE, ESPECIALLY IN PRODUCTION!!!**.
+
+## Setup
+
+First a virtual environment should be created to download dependencies to.
+
+```bash
+cd scripts
+python3 -m venv .venv
+source .venv/bin/activate
+pip install -r requirements.txt
+```
+
+Make sure when you are in the correct environment when running scripts.
+
+## Examples
+
+### Cap Docket, Document, and Comment downloaded counts by the counts from Regulations.gov
+
+```bash
+./get_counts.py redis | ./correct_counts.py | ./set_counts.py -y
+```
+
+### Set Docket, Document, Comment downloaded counts while jobs are in the queue
+
+```bash
+./get_counts.py dashboard | ./correct_counts.py --ignore-queue --strategy diff_total_with_jobs | ./set_counts.py -y
+```
+
+### Download Counts for a Certain Time from Regulations.gov
+
+```bash
+./get_counts.py --api-key $API_KEY -o aug_6_2022.json -t 2024-08-06T06:20:50Z
+
+EXPORT API_KEY=<REGULATIONS.GOV_API_KEY>
+./get_counts.py regulations -o oct_01_2024.json --last-timestamp 2024-10-01T15:30:10Z
+./set_counts.py -i oct_01_2024.json
+```