Afternoon update

.gitignore

@@ -2,3 +2,4 @@
 .Rhistory
 .RData
 .Ruserdata
+*.html

README.md

@@ -4,4 +4,4 @@ Repo containing all of the class material.
 
 - `workbook` - Contains a `bookdown` with all of the exercises that will be used during the class
 
-- `dashboard` - Has a completed version of the `shinydashboard` that will be created across multiple exercises during the class
+- `dashboard` (under the `assets` folder) - Has a completed version of the `shinydashboard` that will be created across multiple exercises during the class

config.yml

@@ -0,0 +1,8 @@
+default:
+  datawarehouse-dev:
+    driver: 'PostgreSQL' 
+    server: 'localhost'
+    uid: 'rstudio_admin'
+    pwd: 'admin_user_be_careful'  
+    port: 5432
+    database: 'postgres'

options.R

@@ -0,0 +1,2 @@
+options(database_userid = "rstudio_dev")
+options(database_password = "dev_user")

workbook/01-database-access.Rmd

@@ -1,13 +1,15 @@
+```{r, section01, include = FALSE}
+knitr::opts_chunk$set(eval = TRUE)
+```
 
+# Access a database
 
 ```{r, include = FALSE}
 library(dplyr)
 library(dbplyr)
 library(DBI)
 ```
 
-# Access a database
-
 ## Connect to a database
 
 *The simpliest way to connect to a database.  More complex examples will be examined later in the class.*
@@ -133,6 +135,12 @@ con <- dbConnect(
 
 2. When prompted, type in **rstudio_dev** for the user, and **dev_user** as the password
 
+3. Disconnect from the database using `dbDisconnect()`
+```{r}
+dbDisconnect(con)
+```
+
+
 ## Secure credentials in a file
 
 *Credentials can be saved in a YAML file and then read using the `config` package: http://db.rstudio.com/best-practices/managing-credentials/#stored-in-a-file-with-config *
@@ -160,6 +168,13 @@ con <- dbConnect(odbc::odbc(),
 )
 ```
 
+5. Disconnect from the database using `dbDisconnect()`
+```{r}
+dbDisconnect(con)
+```
+
+
+
 ## Environment variables
 *Use .Renviron file to store credentials*
 
@@ -183,6 +198,12 @@ con <- dbConnect(
 )
 ```
 
+4. Disconnect from the database using `dbDisconnect()`
+```{r}
+dbDisconnect(con)
+```
+
+
 ## Use options()
 *Set options() in a separate R script*
 
@@ -211,8 +232,7 @@ con <- dbConnect(
 )
 ```
 
-
-
-```{r, include = FALSE}
+5. Disconnect from the database using `dbDisconnect()`
+```{r}
 dbDisconnect(con)
 ```

workbook/02-dplyr-basics.Rmd

@@ -1,13 +1,11 @@
-
-```{r, include = FALSE}
+```{r, section02, include = FALSE}
 knitr::opts_chunk$set(eval = TRUE)
 ```
 
-
 # `dplyr` Basics
 
-```{r, include = FALSE}
 
+```{r, include = FALSE}
 library(dplyr)
 library(dbplyr)
 library(DBI)
@@ -18,16 +16,15 @@ library(DBI)
 *Basics to how to point a variable in R to a table or view inside the database*
 
 
-1. Load the `dplyr` and `dbplyr` libraries
+1. Load the `dplyr`, `DBI` and `dbplyr` libraries
 ```{r, dplyr}
 library(dplyr)
 library(dbplyr)
+library(DBI)
 ```
 
 2. *(Optional)* Open a connection to the database if it's currently closed
 ```{r}
-library(DBI)
-
 con <- dbConnect(odbc::odbc(), "Postgres Dev")
 ```
 
@@ -50,37 +47,10 @@ airports
 
 6. Set up the pointers to the other of the tables
 ```{r}
-flights <- tbl(con, in_schema("datawarehouse", "flight"))
+flights <- tbl(con, in_schema("datawarehouse", "vflight"))
 carriers <- tbl(con, in_schema("datawarehouse", "carrier"))
 ```
 
-## Basic aggregation
-*A couple of `dplyr` commands that run in-database*
-
-1. How many records are in the **airport** table?
-```{r}
-tbl(con, in_schema("datawarehouse", "vflight"))  %>%
-  group_by(month) %>%
-  tally()
-```
-
-2. What is the average character length of the airport codes? How many characters is the longest and the shortest airport name?
-```{r}
-airports %>%
-  summarise(
-    avg_airport_length = mean(length(airport), na.rm = TRUE),
-    max_airport_name = max(length(airportname), na.rm = TRUE),
-    min_airport_name = min(length(airportname), na.rm = TRUE),
-    total_records = n()
-  )
-```
-
-**Additional exercises:**
-
-1. How many records are in the **carrier** table?
-
-2. How many characters is the longest **carriername**?
-
 ## Under the hood 
 * Use `show_query()` to preview the SQL statement that will be sent to the database*
 
@@ -91,22 +61,24 @@ show_query(airports)
 
 2. Easily view the resulting query by adding `show_query()` in another piped command
 ```{r}
-carriers %>%
-  summarise(n())
+airports %>%
+  show_query()
 ```
 
-3. Run the same for last exercise in the previous section
+3. Insert `head()` in between the two statements to see how the SQL changes
 ```{r}
 airports %>%
-  summarise(
-    avg_airport_length = mean(length(airport), na.rm = TRUE),
-    max_airport_name = max(length(airportname), na.rm = TRUE),
-    min_airport_name = min(length(airportname), na.rm = TRUE),
-    total_records = n()
-  ) %>%
+  head() %>%
   show_query()
 ```
 
+4. Use `sql_render()` and `simulate_mssql()` to see how the SQL statement changes from vendor to vendor
+```{r}
+airports %>%
+  head() %>%
+  sql_render(con = simulate_mssql()) 
+```
+
 ## Un-translated R commands
 *Review of how `dbplyr` handles R commands that have not been translated into a like-SQL command*
 
@@ -131,25 +103,6 @@ airports %>%
   select(today) %>%
   head()
 ```
-## knitr SQL engine
-
-1. Copy the result of `show_query()` 
-```{r}
-airports %>%
-  summarise(
-    avg_airport_length = mean(length(airport), na.rm = TRUE),
-    max_airport_name = max(length(airportname), na.rm = TRUE),
-    min_airport_name = min(length(airportname), na.rm = TRUE),
-    total_records = n()
-  ) %>%
-  show_query()
-```
-
-2. Paste the result in this SQL chunk
-```{sql, connection = con}
-SELECT AVG(LENGTH("airport")) AS "avg_airport_length", MAX(LENGTH("airportname")) AS "max_airport_name", MIN(LENGTH("airportname")) AS "min_airport_name", COUNT(*) AS "total_records"
-FROM datawarehouse.airport
-```
 
 ## Using bang-bang
 *Intro on passing unevaluated code to a dplyr verb*
@@ -172,9 +125,64 @@ airports %>%
 ```{r}
 airports %>%
   mutate(today = !!Sys.time()) %>%
+  select(today) %>%
   head()
 ```
 
+## knitr SQL engine
+
+1. Copy the result of the latest `show_query()` exercise
+```{r}
+airports %>%
+  mutate(today = !!Sys.time()) %>%
+  show_query()
+```
+
+2. Paste the result in this SQL chunk
+```{sql, connection = con}
+SELECT "airport", "airportname", "city", "state", "country", "lat", "long", '2018-01-26T14:50:10Z' AS "today"
+FROM datawarehouse.airport
+```
+
+
+## Basic aggregation
+*A couple of `dplyr` commands that run in-database*
+
+1. How many records are in the **airport** table?
+```{r}
+tbl(con, in_schema("datawarehouse", "airport"))  %>%
+  tally()
+```
+
+2. What is the average character length of the airport codes? How many characters is the longest and the shortest airport name?
+```{r}
+airports %>%
+  summarise(
+    avg_airport_length = mean(str_length(airport), na.rm = TRUE),
+    max_airport_name = max(str_length(airportname), na.rm = TRUE),
+    min_airport_name = min(str_length(airportname), na.rm = TRUE),
+    total_records = n()
+  )
+```
+
+3. How many records are in the **carrier** table?
+```{r}
+carriers %>%
+  tally()
+```
+
+4. How many characters is the longest **carriername**?
+```{r}
+carriers %>%
+  summarise(x = max(str_length(carriername), na.rm = TRUE))
+```
+
+5. What is the SQL statement sent in exercise 4?
+```{r}
+carriers %>%
+  summarise(x = max(str_length(carriername), na.rm = TRUE)) %>%
+  show_query()
+```
 
 ```{r, include = FALSE}
 dbDisconnect(con)