mappingvermont

import io

import boto3
from boto.s3.connection import S3Connection
import smart_open


def download_csv(bucket, path, iso):
    # read and filter a single csv

mappingvermont

mappingvermont

mappingvermont

import os
import uuid
import hashlib
import shutil


tif_list = [x for x in os.listdir(".") if os.path.splitext(x)[1] == ".tif"]

for tif in tif_list:
    u = str(uuid.uuid4())

mappingvermont

# based on this guide:
https://orc.apache.org/docs/building.html

# check out http://www.mirrorservice.org/sites/ftp.apache.org/orc/ to find the newest version

# download it
curl -sSLO http://www.mirrorservice.org/sites/ftp.apache.org/orc/orc-1.5.2/orc-1.5.2.tar.gz

# unzip
tar -xvf orc-1.5.2.tar.gz

mappingvermont

mappingvermont

import boto3
import json


session = boto3.Session(profile_name='gfwpro')
client = session.client('lambda', region_name='us-east-1')

aoi = {"features":[{"properties":{},"type":"Feature","geometry":{"type":"Polygon","coordinates":[[[140.2137,-6.3999],[140.3078,-6.3685],[140.3249,-6.3924],[140.3249,-6.4606],[140.3064,-6.49],[140.2274,-6.4838],[140.2068,-6.434],[140.2137,-6.3999]]]}}],"crs":{},"type":"FeatureCollection"}

# then build an event to kick off the process

mappingvermont

var AWS = require('aws-sdk');

var run = function(feat) {
  var lambda = new AWS.Lambda({region: 'us-east-1'});
  var params = {
    FunctionName: 'geoproc-raster-umd', /* required */
    Payload: JSON.stringify({'queryStringParameters': {'analysis': 'extent', 'thresh': 30}, 'body': {'geojson': {"type":"FeatureCollection","features":[feat]}}})}

  lambda.invoke(params, function(err, data) {
    if (err) console.log(err, err.stack); // an error occurred

mappingvermont

Run split.sh to split your input file into 3 pieces - 2 small ones and one giant one.

Then output temp.json and find and replace all 0.0 values for area_loss and emissions with 0.0000001

Then create a file called `temp` with just } in it. For some reason it seems like the split process 
above drops the final } from the original JSON file.

Then combine all three files using head, because `cat` will add new lines that we don't want:
head -c -1 -q temp.json remainder temp > adm2.json

mappingvermont

import pandas as pd
import os
import sys

csv_name = sys.argv[1]
src_csv = os.path.join('data', csv_name + '.csv')
dst_csv = os.path.join('final', csv_name + '.csv')

df = pd.read_csv(src_csv)
df.loc[df['polyname'] == 'primary-forest', 'polyname'] = 'primary_forest'