From 823343381d42e055998106cb492ae18908cc8dc0 Mon Sep 17 00:00:00 2001
From: Tobias Janke <git@tojanke.de>
Date: Sun, 8 Dec 2024 11:12:39 +0100
Subject: [PATCH] [Feature] Remove duplicate database dumps to save storage
 space (#127)

* Replace gzip with deterministic bzip2 compression

* Exclude comments from SQL dump

* Add option to remove duplicates using fdupes

* Revert change to bzip2 as gzip can be deterministic using no-name parameter
---
 Dockerfile | 3 ++-
 README.md  | 1 +
 backup.sh  | 9 +++++++--
 3 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/Dockerfile b/Dockerfile
index 51fbf16..e6887c2 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -21,7 +21,8 @@ RUN apk add --update \
         mysql-client \
         gzip \
         openssl \
-        mariadb-connector-c && \
+        mariadb-connector-c \
+        fdupes && \
     rm -rf /var/cache/apk/*
 
 COPY --from=binary /go/bin/dockerize /usr/local/bin
diff --git a/README.md b/README.md
index dad5be3..16a2d65 100644
--- a/README.md
+++ b/README.md
@@ -42,6 +42,7 @@ Container is **Healthy** after the database init phase, that is after `INIT_BACK
 - `GZIP_LEVEL`: Specify the level of gzip compression from 1 (quickest, least compressed) to 9 (slowest, most compressed), default is 6.
 - `USE_PLAIN_SQL`: If set, back up and restore plain SQL files without gzip.
 - `TZ`: Specify TIMEZONE in Container. E.g. "Europe/Berlin". Default is UTC.
+- `REMOVE_DUPLICATES`: Use [fdupes](https://github.com/adrianlopezroche/fdupes) to remove duplicate database dumps
 
 If you want to make this image the perfect companion of your MySQL container, use [docker-compose](https://docs.docker.com/compose/). You can add more services that will be able to connect to the MySQL image using the name `my_mariadb`, note that you only expose the port `3306` internally to the servers and not to the host:
 
diff --git a/backup.sh b/backup.sh
index 747cd82..d7db8a7 100755
--- a/backup.sh
+++ b/backup.sh
@@ -29,13 +29,13 @@ do
     echo "==> Dumping database: $db"
     FILENAME=/backup/$DATE.$db.sql
     LATEST=/backup/latest.$db.sql
-    if mysqldump --single-transaction $MYSQLDUMP_OPTS -h "$MYSQL_HOST" -P "$MYSQL_PORT" -u "$MYSQL_USER" -p"$MYSQL_PASS" $MYSQL_SSL_OPTS "$db" > "$FILENAME"
+    if mysqldump --single-transaction --skip-comments $MYSQLDUMP_OPTS -h "$MYSQL_HOST" -P "$MYSQL_PORT" -u "$MYSQL_USER" -p"$MYSQL_PASS" $MYSQL_SSL_OPTS "$db" > "$FILENAME"
     then
       EXT=
       if [ -z "${USE_PLAIN_SQL}" ]
       then
         echo "==> Compressing $db with LEVEL $GZIP_LEVEL"
-        gzip "-$GZIP_LEVEL" -f "$FILENAME"
+        gzip "-$GZIP_LEVEL" -n -f "$FILENAME"
         EXT=.gz
         FILENAME=$FILENAME$EXT
         LATEST=$LATEST$EXT
@@ -44,6 +44,11 @@ do
       echo "==> Creating symlink to latest backup: $BASENAME"
       rm "$LATEST" 2> /dev/null
       cd /backup || exit && ln -s "$BASENAME" "$(basename "$LATEST")"
+      if [ -n "$REMOVE_DUPLICATES" ]
+      then
+        echo "=> Removing duplicate database dumps"
+        fdupes -idN /backup/
+      fi
       if [ -n "$MAX_BACKUPS" ]
       then
         # Execute the delete script, delete older backup or other custom delete script