Harlan Harris HarlanH

## run_dbt.sh
#!/usr/bin/env bash

# When run, first tries to figure out if the full-refresh flag is set.
# Then, runs dbt with or without the flag.

# Do not run me in dev mode!

full_refresh_needed="$(
psql $REDSHIFT_URI -t <<EOF
SELECT full_refresh

## dbt-Metrics.Rmd
---
title: "dbt Metrics"
output:
  flexdashboard::flex_dashboard:
    orientation: rows
---

<!--

If dbt is being run on an AWS server with logs being pulled into Cloudwatch,

## dbt_doc.py
#!/usr/bin/env python

# This Q&D script scans through SQL files in the models directory and outputs a Markdown document
# with per-model comments.

import os
from os import path
import re
import warnings
import time

## dplyr_db_script.R
library(dplyr)
library(RPostgreSQL)
library(httr)
library(Lahman)
library(ggplot2)

# connect to the db
con <- src_postgres(dbname="harlan", host="localhost", user="harlan")

# upload the Batting db

## randy_venn.jl
using VennEuler

# 1		364	23.30%	SAS only
# 10	350	22.41%	Python only
# 11	70	4.48%	Python and SAS
# 100	490	31.37%	R only
# 101	68	4.35%	R and SAS
# 110	200	12.80%	R and Python
# 111	20	1.28%	R, Python and SAS

## polygontest.R
library(plyr)
library(rMaps)

venues <- structure(list(name = c("pivotal", "aol", "columbia"), lat = c(40.7403372,
                    40.7308948, 40.8074358), lon = c(-73.9951462, -73.9917096, -73.9625858 )),
                    .Names = c("name", "lat", "lon"), row.names = c(NA, -3L), class = "data.frame")

times_square <- c(40.7577, -73.9857)

map <- Leaflet$new()

## test1.js
var http = require('http');
http.createServer(function (req, res) {
  res.writeHead(200, {'Content-Type': 'text/plain'});
  res.end('Hello Koding');
}).listen(1337, '0.0.0.0');
console.log('Server running at http://host:1337/');

## vect_set.jl
# to do set stuff with vectors, turn the B vector into a set, then
# iterate over the a vector, creating a set to remove dupes.
# This is moderately efficient, preserves order, and removes dupes.
function intersect{T}(a::Vector{T}, b::Vector{T})
    bset = Set(b...)
	aset = Set()
	ret = T[]
	for a_elem in a
		if has(bset, a_elem) && !has(aset, a_elem)
			push(ret, a_elem)

## IndexDict.jl
type IndexDict{V} <: Associative{ByteString,V}
    idx::Index
    arr::Vector{V}

    IndexDict() = new(Index(), Array(V,0))
end
IndexDict() = IndexDict{Any}()

# assignment by a string replaces or appends
function assign(id::IndexDict, v, key::ByteString)

## vswitch.R
vswitch <- function(namedList, default=NA, selector) {
  # Function adapted from Bill Dunlap that implements something along the lines of a vectorized switch statement.
  # http://tolstoy.newcastle.edu.au/R/e8/devel/09/12/1122.html
  #
  # Args:
  #    namedList - e.g., list(times=df$a * df$b, plus=df$a + df$b)
  #    default - a value to assign to elements of selector that aren't matched in namedList
  #    selector - e.g., c('times', 'times', 'plus', 'exp', 'plus')
  #
  # Returns: a vector of values selected from namedList
	#!/usr/bin/env bash

	# When run, first tries to figure out if the full-refresh flag is set.
	# Then, runs dbt with or without the flag.

	# Do not run me in dev mode!

	full_refresh_needed="$(
	psql $REDSHIFT_URI -t <<EOF
	SELECT full_refresh
	---
	title: "dbt Metrics"
	output:
	flexdashboard::flex_dashboard:
	orientation: rows
	---

	<!--

	If dbt is being run on an AWS server with logs being pulled into Cloudwatch,
	#!/usr/bin/env python

	# This Q&D script scans through SQL files in the models directory and outputs a Markdown document
	# with per-model comments.

	import os
	from os import path
	import re
	import warnings
	import time
	library(dplyr)
	library(RPostgreSQL)
	library(httr)
	library(Lahman)
	library(ggplot2)

	# connect to the db
	con <- src_postgres(dbname="harlan", host="localhost", user="harlan")

	# upload the Batting db
	using VennEuler

	# 1 364 23.30% SAS only
	# 10 350 22.41% Python only
	# 11 70 4.48% Python and SAS
	# 100 490 31.37% R only
	# 101 68 4.35% R and SAS
	# 110 200 12.80% R and Python
	# 111 20 1.28% R, Python and SAS
	library(plyr)
	library(rMaps)

	venues <- structure(list(name = c("pivotal", "aol", "columbia"), lat = c(40.7403372,
	40.7308948, 40.8074358), lon = c(-73.9951462, -73.9917096, -73.9625858 )),
	.Names = c("name", "lat", "lon"), row.names = c(NA, -3L), class = "data.frame")

	times_square <- c(40.7577, -73.9857)

	map <- Leaflet$new()
	var http = require('http');
	http.createServer(function (req, res) {
	res.writeHead(200, {'Content-Type': 'text/plain'});
	res.end('Hello Koding');
	}).listen(1337, '0.0.0.0');
	console.log('Server running at http://host:1337/');
	# to do set stuff with vectors, turn the B vector into a set, then
	# iterate over the a vector, creating a set to remove dupes.
	# This is moderately efficient, preserves order, and removes dupes.
	function intersect{T}(a::Vector{T}, b::Vector{T})
	bset = Set(b...)
	aset = Set()
	ret = T[]
	for a_elem in a
	if has(bset, a_elem) && !has(aset, a_elem)
	push(ret, a_elem)
	type IndexDict{V} <: Associative{ByteString,V}
	idx::Index
	arr::Vector{V}

	IndexDict() = new(Index(), Array(V,0))
	end
	IndexDict() = IndexDict{Any}()

	# assignment by a string replaces or appends
	function assign(id::IndexDict, v, key::ByteString)
	vswitch <- function(namedList, default=NA, selector) {
	# Function adapted from Bill Dunlap that implements something along the lines of a vectorized switch statement.
	# http://tolstoy.newcastle.edu.au/R/e8/devel/09/12/1122.html
	#
	# Args:
	# namedList - e.g., list(times=df$a * df$b, plus=df$a + df$b)
	# default - a value to assign to elements of selector that aren't matched in namedList
	# selector - e.g., c('times', 'times', 'plus', 'exp', 'plus')
	#
	# Returns: a vector of values selected from namedList