From 4619d4d8b75be9743d42ae9e3dd6f31e1dcbf007 Mon Sep 17 00:00:00 2001
From: Vincent Batts <vbatts@hashbangbash.com>
Date: Thu, 21 May 2015 11:14:05 -0400
Subject: [PATCH] dups: deduplication tool

---
 README.md                  | 33 +++++++++++++++++++++++++++++++++
 cmd/slackware-sync/main.go | 14 ++++++++++----
 2 files changed, 43 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index eb3ea80..ab704bd 100644
--- a/README.md
+++ b/README.md
@@ -4,6 +4,39 @@ Random utilities of vbatts' being cleaned up, and published
 
 # Commands
 
+## dups
+
+building a document of file checksum info, for a directory tree. Optionally
+deduplicate the tree using hardlinks.
+
+### Install
+
+	go get github.com/vbatts/freezing-octo-hipster/cmd/dups
+
+### Usage
+
+	$ dups -h
+	Usage of dups:
+	  -H=false: hardlink the duplicate files
+	  -l="": load existing map from file
+	  -q=false: less output
+	  -s="hash-map.json": file to save map of file hashes to
+
+By default it scans the paths provided, and creates a JSON document of the file paths and their checksum:
+
+	$ dups .
+	"/home/vbatts/src/vb/freezing-octo-hipster/.git/logs/refs/heads/master" is the same content as "/home/vbatts/src/vb/freezing-octo-hipster/.git/logs/HEAD"
+	"/home/vbatts/src/vb/freezing-octo-hipster/.git/refs/remotes/origin/master" is the same content as "/home/vbatts/src/vb/freezing-octo-hipster/.git/refs/heads/master"
+	"/home/vbatts/src/vb/freezing-octo-hipster/cmd/find-todos/main.go~" is the same content as "/home/vbatts/src/vb/freezing-octo-hipster/cmd/find-todos/main.go"
+	"/home/vbatts/src/vb/freezing-octo-hipster/cmd/slackware-sync/README.md~" is the same content as "/home/vbatts/src/vb/freezing-octo-hipster/cmd/slackware-sync/README.md"
+	"/home/vbatts/src/vb/freezing-octo-hipster/cmd/slackware-sync/main.go" is the same content as "/home/vbatts/src/vb/freezing-octo-hipster/cmd/slackware-sync/main.go~"
+	Savings of 0.005681mb
+	wrote "hash-map.json"
+
+With the `-H` flag, as duplicate files are found (files with matching checksum)
+are encountered, hardlink it to the duplicate file.
+
+
 ## next-note
 
 Simple date formating for notes
diff --git a/cmd/slackware-sync/main.go b/cmd/slackware-sync/main.go
index cac7dc4..6e29ac9 100644
--- a/cmd/slackware-sync/main.go
+++ b/cmd/slackware-sync/main.go
@@ -3,11 +3,12 @@ package main
 import (
 	"flag"
 	"fmt"
-	"github.com/BurntSushi/toml"
 	"net/url"
 	"os"
 	"os/exec"
 	"path"
+
+	"github.com/BurntSushi/toml"
 )
 
 func main() {
@@ -19,6 +20,10 @@ func main() {
 		os.Exit(1)
 	}
 
+	if len(*flSyncDir) > 0 {
+		config.SyncDir = *flSyncDir
+	}
+
 	_, err = EnsureDirExists(config.SyncDir)
 	if err != nil {
 		fmt.Println(err)
@@ -47,10 +52,10 @@ func main() {
 			cmd.Stdout = os.Stdout
 		}
 
-    err = cmd.Run()
-    if err != nil {
+		err = cmd.Run()
+		if err != nil {
 			fmt.Fprintln(os.Stderr, err)
-    }
+		}
 	}
 }
 
@@ -82,6 +87,7 @@ type Mirror struct {
 }
 
 var (
+	flSyncDir    = flag.String("dir", "", "directory to sync to (this flag overrides the url in the configuration file)")
 	flConfigFile = flag.String("c", path.Join(os.Getenv("HOME"), ".slackware-sync.toml"), "config file for the sync")
 	flQuiet      = flag.Bool("q", false, "less output")
 )