>A simple way to run a data cube
>[https://github.com/opendatacube/cube-in-a-box](https://)

# How to use postgreSQL

1. Use this command to open psql: psql -U postgres
2. Use pgAdmin to create a user ([follow this page](https://www.guru99.com/postgresql-create-alter-add-user.html))

# Indexing the data
> Documentation
> https://datacube-core.readthedocs.io/en/stable/ops/product.html

> Scripts for indexing data into ODC instances
> https://github.com/opendatacube/datacube-dataset-config

## Create a Product Definition
ODC requires a bit of information up front to know what to do with them

1. Create Our Definition File:

---
name: Nimrod
description: Nimrod c-band Rain Radar 1km
metadata_type: eo3

license: CC-BY-SA-4.0

metadata:
  product:
    name: UKMO

storage:
  crs: EPSG:27700
  resolution:
    x: 1000
    y: 1000
  
measurements:
  - name: "rainrate"
    dtype: "int8"
    units: "mm/h"
    nodata: -1
    aliases: []

2. Use Command Line tool to load the product

## Ensure Dataset Documents are complete
Every dataset that you intend to index requires a metadata document describing what the data represents and where it has come from, as well has what format it is stored in. At a minimum, you need the dimensions or fields your want to search by, such as lat, lon and time, but you can include any information you deem useful.

### EO3 Format
EO3 is an intermediate format before we move to something more standard like STAC.

# UUID of the dataset
id: f884df9b-4458-47fd-a9d2-1a52a2db8a1a
$schema: 'https://schemas.opendatacube.org/dataset'

# Product name
product:
  name: Nimrod_c-band_rain_radar_1km

# Native CRS, assumed to be the same across all bands
crs: "epsg:27700"

# Optional GeoJSON object in the units of native CRS.
# Defines a polygon such that, all valid pixels across all bands
# are inside this polygon.
geometry:
  type: Polygon
  coordinates: [[..]]

# Mapping name:str -> { shape:     Tuple[ny: int, nx: int]
#                       transform: Tuple[float x 9]}
# Captures image size, and geo-registration
grids:
    default:  # "default" grid must be present
       shape: [7811, 7691]
       transform: [30, 0, 618285, 0, -30, -1642485, 0, 0, 1]
    pan:  # Landsat Panchromatic band is higher res image than other bands
       shape: [15621, 15381]
       transform: [15, 0, 618292.5, 0, -15, -1642492.5, 0, 0, 1]

# Per band storage information and references into `grids`
# Bands using "default" grid should not need to reference it
measurements:
   pan:               # Band using non-default "pan" grid
     grid: "pan"      # should match the name used in `grids` mapping above
     path: "pan.tif"
   red:               # Band using "default" grid should omit `grid` key
     path: red.tif    # Path relative to the dataset location
   blue:
     path: blue.tif
   multiband_example:
     path: multi_band.tif
     band: 2          # int: 1-based index into multi-band file
   netcdf_example:    # just example, mixing TIFF and netcdf in one product is not recommended
     path: some.nc
     layer: some_var  # str: netcdf variable to read

# Dataset properties, prefer STAC standard names here
# Timestamp is the only compulsory field here
properties:
  eo:platform: landsat-8
  eo:instrument: OLI_TIRS

  # If it's a single time instance use datetime
  datetime: 2020-01-01T07:02:54.188Z  # Use UTC

  # When recording time range use dtr:{start,end}_datetime
  dtr:start_datetime: 2020-01-01T07:02:02.233Z
  dtr:end_datetime:   2020-01-01T07:03:04.397Z

  # ODC specific "extensions"
  odc:processing_datetime: 2020-02-02T08:10:00.000Z

  odc:file_format: GeoTIFF
  odc:region_code: "074071"   # provider specific unique identified for the same location
                              # for Landsat '{:03d}{:03d}'.format(path, row)

  dea:dataset_maturity: final # one of: final| interim| nrt (near real time)
  odc:product_family: ard     # can be useful for larger installations

# Lineage only references UUIDs of direct source datasets
# Mapping name:str -> [UUID]
lineage: {}  # set to empty object if no lineage is defined

## Run the indexing process
Use Command Line tool to load the product