From 1c82c33641946ba5144f270d51576b676809705f Mon Sep 17 00:00:00 2001
From: Grant Andersen <gandersen101@gmail.com>
Date: Sat, 23 Mar 2019 18:39:50 -0400
Subject: [PATCH] First Commit

---
 .gitignore       |  4 ++++
 lahman_etl.R     | 58 ++++++++++++++++++++++++++++++++++++++++++++++++
 lahman_etl.Rproj | 13 +++++++++++
 3 files changed, 75 insertions(+)
 create mode 100644 .gitignore
 create mode 100644 lahman_etl.R
 create mode 100644 lahman_etl.Rproj

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..5b6a065
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,4 @@
+.Rproj.user
+.Rhistory
+.RData
+.Ruserdata
diff --git a/lahman_etl.R b/lahman_etl.R
new file mode 100644
index 0000000..71254d0
--- /dev/null
+++ b/lahman_etl.R
@@ -0,0 +1,58 @@
+require(DBI)
+require(readr)
+require(dplyr)
+
+
+tables <- c('AllstarFull',
+'Appearances',
+'AwardsManagers',
+'AwardsPlayers',
+'AwardsShareManagers',
+'AwardsSharePlayers',
+'Batting',
+'BattingPost',
+'CollegePlaying',
+'Fielding',
+'FieldingOF',
+'FieldingOFsplit',
+'FieldingPost',
+'HallOfFame',
+'HomeGames',
+'Managers',
+'ManagersHalf',
+'Parks',
+'People',
+'Pitching',
+'PitchingPost',
+'Salaries',
+'Schools',
+'SeriesPost',
+'Teams',
+'TeamsFranchises',
+'TeamsHalf')
+
+foo<-dbConnect(MySQL(), host="localhost", dbname="lahman",user="root",port=3306)
+
+y=Sys.time()
+for (t in tables) {
+  print(t)
+  if (t != "AwardsSharePlayers"){
+  a <- read_csv(paste0("https://github.com/chadwickbureau/baseballdatabank/raw/master/core/",
+                       t,
+                      ".csv"))
+  } else {
+    a <- read_csv(paste0("https://github.com/chadwickbureau/baseballdatabank/raw/master/core/",
+                         t,
+                         ".csv"),col_names = c("awardID","yearID","lgID","playerID",
+                                               "pointsWon","pointsMax","votesFirst"),
+                  skip=1,
+                  col_types = "ciccdid")
+  }
+  
+  dbSendQuery(foo, paste0("drop table if exists ",t,";"))
+  dbWriteTable(foo, name=t, value=a %>% as.data.frame(a))
+    
+}
+Sys.time()-y
+
+dbDisconnect(foo)
\ No newline at end of file
diff --git a/lahman_etl.Rproj b/lahman_etl.Rproj
new file mode 100644
index 0000000..8e3c2eb
--- /dev/null
+++ b/lahman_etl.Rproj
@@ -0,0 +1,13 @@
+Version: 1.0
+
+RestoreWorkspace: Default
+SaveWorkspace: Default
+AlwaysSaveHistory: Default
+
+EnableCodeIndexing: Yes
+UseSpacesForTab: Yes
+NumSpacesForTab: 2
+Encoding: UTF-8
+
+RnwWeave: Sweave
+LaTeX: pdfLaTeX