simonw

# This file lives in /etc/systemd/system/datasette.service
[Unit]
Description=Datasette
After=network.target

[Service]
Type=simple
User=ubuntu
Environment=GITHUB_CLIENT_ID=...
Environment=GITHUB_CLIENT_SECRET=...

simonw

datasette
sqlite-utils
sqlite-transform
csvs-to-sqlite
geojson-to-sqlite
socrata2sql
datasette-cluster-map
datasette-vega
datasette-json-html
datasette-jellyfish

simonw

simonw

Making a Datasette of Finnish coronavirus data

From here: https://github.com/HS-Datadesk/koronavirus-avoindata

Tools

You'll need jq, sqlite-utils and Datasette.

• jq installation instructions: https://stedolan.github.io/jq/download/
• pip install sqlite-utils datasette

simonw

How to run Python 3.7 on Ubuntu 16.04

This is based on Method 2 in https://websiteforstudents.com/installing-the-latest-python-3-7-on-ubuntu-16-04-18-04/

First, you need to install Python 3.7 (you will currently have Python 3.5.2):

sudo apt update
sudo apt install software-properties-common
sudo add-apt-repository ppa:deadsnakes/ppa

sudo apt update

simonw

Machine learning models installed on macOS is part of Vision.framework

I found these while poking around at the list of open files for photoanalysisd in Activity Monitor.

% ls -lahS /System/Library/Frameworks/Vision.framework/Versions/A/Resources
total 465616
-rw-r--r--    1 root  wheel    40M Dec 13 16:32 landmarks_v2.bin
-rw-r--r--    1 root  wheel    31M Dec 13 16:32 scenenet_sc2.4_sa1.4_ae1.4_r9_opt_int8.espresso.weights
-rw-r--r--    1 root  wheel    29M Dec 13 16:32 scenenet_sc2.4_sa1.4_ae1.6_r13.4_opt_int8_asymetric.espresso.weights

simonw

sqlite-utils insert /tmp/loans.db foia_150k_plus --csv '150k plus/foia_150k_plus.csv'
sqlite-utils insert /tmp/loans.db foia_up_to_150k_AL --csv 'Alabama/foia_up_to_150k_AL.csv'
sqlite-utils insert /tmp/loans.db foia_up_to_150k_AK --csv 'Alaska/foia_up_to_150k_AK.csv'
sqlite-utils insert /tmp/loans.db foia_up_to_150k_AS --csv 'American Samoa/foia_up_to_150k_AS.csv'
sqlite-utils insert /tmp/loans.db foia_up_to_150k_AZ --csv 'Arizona/foia_up_to_150k_AZ.csv'
sqlite-utils insert /tmp/loans.db foia_up_to_150k_AR --csv 'Arkansas/foia_up_to_150k_AR.csv'
sqlite-utils insert /tmp/loans.db foia_up_to_150k_CA --csv 'California/foia_up_to_150k_CA.csv'
sqlite-utils insert /tmp/loans.db foia_up_to_150k_CO --csv 'Colorado/foia_up_to_150k_CO.csv'
sqlite-utils insert /tmp/loans.db foia_up_to_150k_CT --csv 'Connecticut/foia_up_to_150k_CT.csv'
sqlite-utils insert /tmp/loans.db foia_up_to_150k_DE --csv 'Delaware/foia_up_to_150k_DE.csv'

simonw

Publishing a private Datasette instance to Heroku using datasette-auth-passwords

Using this new plugin: https://github.com/simonw/datasette-auth-passwords

Create your database as usual, then create a metadata.json file that looks like this:

{
    "allow": {
        "id": "root"
 }

simonw

simonw