zjuul’s gists

zjuul / tidy_tuesday_20230926.R

Last active September 26, 2023 16:16

	library(tidyverse)

	richmondway <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2023/2023-09-26/richmondway.csv')


	ggplot(
	richmondway,
	aes(
	x = (max(richmondway$Episode) + 1) * Season + Episode,
	y = F_count_RK

zjuul / enphase_solar_plot.R

Last active July 23, 2022 21:13

	library(httr)
	library(jsonlite)
	library(lubridate)
	library(tidyverse)



	########### config
	#
	# my info - replace this with your enphase api key, app id, user_id

zjuul / add-missing-datalayer-event.js

Created September 14, 2020 09:08 — forked from sahava/add-missing-datalayer-event.js

This piece of code adds a default 'event' key-value to each dataLayer.push() that hasn't got one.

	(function() {
	var oldPush = window.dataLayer.push;
	window.dataLayer.push = function() {
	var states = [].slice.call(arguments, 0);
	states.forEach(function(s) {
	if (!s.hasOwnProperty('event')) {
	s.event = 'default';
	}
	});
	return oldPush.apply(window.dataLayer, states);

zjuul / wide_to_long.sql

Last active April 20, 2020 07:27

Transform your long data to wide data - to break down the request in phases

	create or replace table `PROJECTNAME.DATASETNAME.performance_long` partition by DATE(timestamp) as

	with base as (
	select
	timestamp, date, event_timestamp, clientid, user_id, device_category, country, url, hostname,
	page_path, page_title, page_referrer, type
	from `PROJECTNAME.DATASETNAME.performance_wide`
	)

	,intervals as (

zjuul / transform_to_wide.sql

Created April 14, 2020 20:18

Transform Performance Navigation data to wide format

	create or replace table `PROJECTNAME.DATASETNAME.performance_wide` partition by DATE(timestamp) as
	/*
	wide version of nested events table
	*/



	SELECT
	TIMESTAMP_MICROS(event_timestamp) as timestamp,
	DATE(TIMESTAMP_MICROS(event_timestamp)) as date,

zjuul / tdf.R

Created July 9, 2019 15:11

Tour de France 2019 - plot general classification

	library(httr)
	library(jsonlite)
	library(ggplot2)
	library(dplyr)

	# url with TDF standings
	url <- "https://sportapi.widgets.sports.gracenote.com/cycling/getresult/classificationid/2148837/languagecode/1.json?c=64&module=cycling&type=classification"

	standings <- fromJSON(content(GET(url), "text")[[1]])

zjuul / GTM_track_autocomplete.js

Created March 20, 2019 08:02

	// GTM custom HTML tag for tracking incomplete searches
	// logic: if user starts typing, start a timer. Stop the timer when user is idle 2.5 seconds
	// then push the event on the DL

	(function() {
	var timer, firstResult, searchText;
	var waiting = false;

	var waitTime = 2500; // milliseconds

zjuul / all_join_types.sql

Created August 23, 2018 13:58

SQL joins: all in one

	-- create two tables: L and R
	-- content of tables a "val" column with two rows.
	-- rows in L: "left only" and "both"
	-- rows in R: "right only" and "both"

	with l as (
	select 'both' as val
	union
	select 'left_only' as val
	), r as (

zjuul / cohort_chart.R

Last active December 26, 2020 15:37

	# From: https://stuifbergen.com/2018/03/cohort-analysis-with-snowplow-and-r/

	########## 1: get data
	appname <- "NAME" # used as selection criterium, replace that with the name of your app
	q <- paste0("
	SELECT
	user_id,
	TO_CHAR(CONVERT_TIMEZONE('UTC', 'Europe/Amsterdam', derived_tstamp),'YYYY ww') AS yw
	FROM atomic.events
	WHERE app_id = '", appname, "' AND derived_tstamp >= '2018-01-01'

zjuul / supershort_manual.txt

Last active August 7, 2018 13:32

G-dash on synology nas

	Step 1: Download the binaries
	https://github.com/Gulden/gulden-official/releases/tag/v2.0.0.9
	(choose Gulden-2.0.0.9-arm-linux-eabihf.tar.gz )

	Step 2: unpack the binaries on your NAS somewhere, create a data directory
	(outside home, outside web root)

	Step 3: create Gulden.conf file inside the data directory
	https://g-dash.nl/manual/index.html

Jules zjuul