Module grib::cookbook

A cookbook of examples for GRIB2 data handling.

Table of contents

1. Comparison of various GRIB2 data library/tool operations
   • Listing all submessages inside
   • Finding submessages inside that match some condition
   • Extracting values with location info from a submessage

Comparison of various GRIB2 data library/tool operations

This section provides example code of data operations using some GRIB processing libraries and tools.

Listing all submessages inside

GRIB tools from ecCodes:

$ grib_ls datafile.grib

wgrib2:

$ wgrib2 datafile.grib

pygrib:

import pygrib

path = "testdata/icon_global_icosahedral_single-level_2021112018_000_TOT_PREC.grib2"
grib = pygrib.open(path)
for submessage in grib:
    print(submessage)

grib-rs:

use std::{fs::File, io::BufReader, path::Path};

use grib::codetables::{CodeTable4_2, Lookup};

fn list_submessages<P>(path: P)
where
    P: AsRef<Path>,
{
    let f = File::open(path).unwrap();
    let f = BufReader::new(f);

    let grib2 = grib::from_reader(f).unwrap();

    for (_index, submessage) in grib2.iter() {
        let discipline = submessage.indicator().discipline;
        let category = submessage.prod_def().parameter_category().unwrap();
        let parameter = submessage.prod_def().parameter_number().unwrap();
        let parameter = CodeTable4_2::new(discipline, category).lookup(usize::from(parameter));

        let forecast_time = submessage.prod_def().forecast_time().unwrap();

        let (first, _second) = submessage.prod_def().fixed_surfaces().unwrap();
        let elevation_level = first.value();

        println!(
            "{:<31} {:>14} {:>17}",
            parameter.to_string(),
            forecast_time.to_string(),
            elevation_level
        );
    }
}

fn main() {
    let path = "testdata/icon_global_icosahedral_single-level_2021112018_000_TOT_PREC.grib2";
    list_submessages(&path);
}

gribber:

$ gribber list datafile.grib

Finding submessages inside that match some condition

GRIB tools from ecCodes:

$ grib_ls -w forecastTime=0,level=850,shortName=u datafile.grib

wgrib2:

$ wgrib2 datafile.grib -match ':3 hour fcst:'

pygrib:

import pygrib

path = "testdata/icon_global_icosahedral_single-level_2021112018_000_TOT_PREC.grib2"
grib = pygrib.open(path)
for submessage in grib.select(forecastTime=3):
    print(submessage)

grib-rs:

use std::{fs::File, io::BufReader, path::Path};

use grib::{codetables::grib2::*, ForecastTime, Name};

fn find_submessages<P>(path: P, forecast_time_hours: u32)
where
    P: AsRef<Path>,
{
    let f = File::open(path).unwrap();
    let f = BufReader::new(f);

    let grib2 = grib::from_reader(f).unwrap();

    for (index, submessage) in grib2.iter() {
        let ft = submessage.prod_def().forecast_time();
        match ft {
            Some(ForecastTime {
                unit: Name(Table4_4::Hour),
                value: hours,
            }) => {
                if hours == forecast_time_hours {
                    println!("{}.{}: {}", index.0, index.1, hours);
                }
            }
            _ => {}
        }
    }
}

fn main() {
    let path = "testdata/icon_global_icosahedral_single-level_2021112018_000_TOT_PREC.grib2";
    find_submessages(&path, 3);
}

gribber:

$ gribber list datafile.grib | grep '3 Hour'

(gribber’s API for finding submessages is still in the conceptual stage and is not yet available.)

Extracting values with location info from a submessage

GRIB tools from ecCodes:

$ grib_get_data -w forecastTime=0,count=1 datafile.grib

wgrib2 (creating a flat binary file of values):

$ wgrib2 -d 1.1 -order we:ns -no_header -bin output.bin datafile.grib

pygrib:

import pygrib

path = "datafile.grib"
grib = pygrib.open(path)
submessage = grib.message(1)
lats, lons = submessage.latlons()
values = submessage.values
print((lats, lons, values))

grib-rs:

use std::{
    fs::File,
    io::{BufReader, Read, Write},
    path::Path,
};

use grib::codetables::{grib2::*, *};

fn decode_layer<P>(path: P, message_index: (usize, usize))
where
    P: AsRef<Path>,
{
    let f = File::open(path).unwrap();
    let f = BufReader::new(f);

    let grib2 = grib::from_reader(f).unwrap();
    let (_index, submessage) = grib2
        .iter()
        .find(|(index, _)| *index == message_index)
        .ok_or("no such index")
        .unwrap();

    let latlons = submessage.latlons().unwrap();
    let decoder = grib::Grib2SubmessageDecoder::from(submessage).unwrap();
    let values = decoder.dispatch().unwrap();

    for ((lat, lon), value) in latlons.zip(values) {
        println!("{lat} {lon} {value}");
    }
}

fn main() {
    let path = "testdata/gdas.t12z.pgrb2.0p25.f000.0-10.xz";

    let mut buf = Vec::new();
    let mut out = tempfile::NamedTempFile::new().unwrap();

    let f = File::open(path).unwrap();
    let f = BufReader::new(f);
    let mut f = xz2::bufread::XzDecoder::new(f);
    f.read_to_end(&mut buf).unwrap();
    out.write_all(&buf).unwrap();

    decode_layer(&out.path(), (0, 0));
}

gribber (showing values along with lat/lon info):

$ gribber decode datafile.grib 0.0

gribber (creating a flat binary file of values):

$ gribber decode -b output.bin datafile.grib 0.0