From 9cb5d6eec8facd872b3fbe24e73d8d9a53c12769 Mon Sep 17 00:00:00 2001
From: Waldo Jaquith <waldo@jaquith.org>
Date: Sun, 11 Mar 2018 21:16:38 -0400
Subject: [PATCH] Turn this into an API

This satisfies most of the requirements of #1. It remains to run this
on a cron job.
---
 .gitignore           |  1 +
 api.sh => bin/api.sh |  2 ++
 bin/csv.sh           | 15 +++++++++++++++
 bin/get_file.sh      | 18 ++++++++++++++++++
 bin/upload.sh        | 10 ++++++++++
 config.sh            | 13 +++++++++++++
 run.sh               |  8 ++++----
 7 files changed, 63 insertions(+), 4 deletions(-)
 rename api.sh => bin/api.sh (94%)
 create mode 100755 bin/csv.sh
 create mode 100755 bin/get_file.sh
 create mode 100755 bin/upload.sh
 create mode 100755 config.sh

diff --git a/.gitignore b/.gitignore
index 8c6492c..3c14011 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1 +1,2 @@
 api/
+NationalFedCodes.txt
diff --git a/api.sh b/bin/api.sh
similarity index 94%
rename from api.sh
rename to bin/api.sh
index 2617147..dfef044 100755
--- a/api.sh
+++ b/bin/api.sh
@@ -9,6 +9,8 @@ OUTPUT_DIR=./api
 # Create the JSON directory.
 mkdir -p "$OUTPUT_DIR"
 
+echo "Creating JSON files"
+
 # Iterate through each line in the CSV file.
 while read -r p; do
 	FIPS_VALUE=$(echo "$p" | cut -d "," -f 1)
diff --git a/bin/csv.sh b/bin/csv.sh
new file mode 100755
index 0000000..d7449fa
--- /dev/null
+++ b/bin/csv.sh
@@ -0,0 +1,15 @@
+#/bin/bash
+
+# Import variables
+source config.sh
+
+echo "Creating CSV file"
+
+# Create the headers for the CSV file.
+echo "GNIS_ID,FIPS_CODE" > "$CSV_FILE"
+
+# Skip the first line, pipe the 1st and 4th columns into the CSV file.
+sed 1d "$SOURCE_FILE" | cut -d "|" -f 1,4  >> "$CSV_FILE"
+
+# Turn every pipe symbol into a comma.
+sed -i -e "s/|/,/g" "$CSV_FILE"
diff --git a/bin/get_file.sh b/bin/get_file.sh
new file mode 100755
index 0000000..a87bf08
--- /dev/null
+++ b/bin/get_file.sh
@@ -0,0 +1,18 @@
+#!/bin/bash
+
+# Get the current filename from the USGS download page.
+FILENAME=$(curl -s https://geonames.usgs.gov/domestic/download_data.htm |egrep -o "\/docs\/federalcodes\/NationalFedCodes_[0-9]{8}.zip")
+
+# Put together the URL for the file
+URL=https://geonames.usgs.gov"$FILENAME"
+
+echo Retreiving "$URL"
+
+# Get the file.
+curl -s "$URL" -o NationalFedCodes.zip
+
+# Extract the file's contents.
+unzip -p NationalFedCodes.zip > NationalFedCodes.txt
+
+# Get rid of the ZIP file.
+rm NationalFedCodes.zip
diff --git a/bin/upload.sh b/bin/upload.sh
new file mode 100755
index 0000000..0758adb
--- /dev/null
+++ b/bin/upload.sh
@@ -0,0 +1,10 @@
+#!/bin/bash
+
+# Import variables
+source config.sh
+
+echo "Syncing files with S3"
+
+# Sync all recent changes
+cd "$OUTPUT_DIR"
+aws s3 sync . s3://"$BUCKET"
diff --git a/config.sh b/config.sh
new file mode 100755
index 0000000..9b4b928
--- /dev/null
+++ b/config.sh
@@ -0,0 +1,13 @@
+#!/bin/bash
+
+# The file from the U.S. Board of Geographic Names' website.
+SOURCE_FILE=NationalFedCodes.txt
+
+# The CSV file.
+CSV_FILE=united-states.csv
+
+# The local directory to store the JSON files.
+OUTPUT_DIR=./api
+
+# The S3 bucket where we want to host the JSON files.
+BUCKET=fipsgnis.com
diff --git a/run.sh b/run.sh
index 7a4135d..443a30d 100755
--- a/run.sh
+++ b/run.sh
@@ -1,4 +1,4 @@
-#/bin/bash
-echo "GNIS_ID,FIPS_CODE" > united-states.csv
-sed 1d NationalFedCodes.txt | cut -d "|" -f 1,4  >> united-states.csv
-sed -i -e "s/|/,/g" ./united-states.csv
+#!/bin/bash
+
+# Run each step in turn.
+bin/get_file.sh && bin/csv.sh && bin/api.sh && bin/upload.sh