From c84895a5ebada96af5e430593d306cd2880cd259 Mon Sep 17 00:00:00 2001
From: Thomas Waldmann <tw@waldmann-edv.de>
Date: Wed, 9 Jun 2021 23:02:21 +0200
Subject: [PATCH] dedup get_tar_filter code

also:
- check for short tarfile extensions
- add .lz4 and .zstd
- fix docs about extensions and decompression commands
---
 src/borg/archiver.py     | 42 +++++++++++++---------------------------
 src/borg/helpers/misc.py | 18 +++++++++++++++++
 2 files changed, 31 insertions(+), 29 deletions(-)

diff --git a/src/borg/archiver.py b/src/borg/archiver.py
index 8f82da2d8..1dd4f1142 100644
--- a/src/borg/archiver.py
+++ b/src/borg/archiver.py
@@ -75,6 +75,7 @@ try:
     from .helpers import msgpack
     from .helpers import sig_int
     from .helpers import iter_separated
+    from .helpers import get_tar_filter
     from .nanorst import rst_to_terminal
     from .patterns import ArgparsePatternAction, ArgparseExcludeFileAction, ArgparsePatternFileAction, parse_exclude_pattern
     from .patterns import PatternMatcher
@@ -951,18 +952,7 @@ class Archiver:
         # that it has to be installed -- hardly a problem, considering that
         # the decompressor must be installed as well to make use of the exported tarball!
 
-        filter = None
-        if args.tar_filter == 'auto':
-            # Note that filter remains None if tarfile is '-'.
-            if args.tarfile.endswith('.tar.gz'):
-                filter = 'gzip'
-            elif args.tarfile.endswith('.tar.bz2'):
-                filter = 'bzip2'
-            elif args.tarfile.endswith('.tar.xz'):
-                filter = 'xz'
-            logger.debug('Automatically determined tar filter: %s', filter)
-        else:
-            filter = args.tar_filter
+        filter = get_tar_filter(args.tarfile) if args.tar_filter == 'auto' else args.tar_filter
 
         tarstream = dash_open(args.tarfile, 'wb')
         tarstream_close = args.tarfile != '-'
@@ -1705,17 +1695,7 @@ class Archiver:
         self.output_filter = args.output_filter
         self.output_list = args.output_list
 
-        filter = None
-        if args.tar_filter == 'auto' and args.tarfile != '-':
-            if args.tarfile.endswith('.tar.gz'):
-                filter = 'gunzip'
-            elif args.tarfile.endswith('.tar.bz2'):
-                filter = 'bunzip2'
-            elif args.tarfile.endswith('.tar.xz'):
-                filter = 'unxz'
-            logger.debug('Automatically determined tar filter: %s', filter)
-        else:
-            filter = args.tar_filter
+        filter = get_tar_filter(args.tarfile) if args.tar_filter == 'auto' else args.tar_filter
 
         tarstream = dash_open(args.tarfile, 'rb')
         tarstream_close = args.tarfile != '-'
@@ -3861,9 +3841,11 @@ class Archiver:
         based on its file extension and pipe the tarball through an appropriate filter
         before writing it to FILE:
 
-        - .tar.gz: gzip
-        - .tar.bz2: bzip2
-        - .tar.xz: xz
+        - .tar.gz or .tgz: gzip
+        - .tar.bz2 or .tbz: bzip2
+        - .tar.xz or .txz: xz
+        - .tar.zstd: zstd
+        - .tar.lz4: lz4
 
         Alternatively, a ``--tar-filter`` program may be explicitly specified. It should
         read the uncompressed tar stream from stdin and write a compressed/filtered
@@ -4733,9 +4715,11 @@ class Archiver:
         By default (--tar-filter=auto) Borg will detect whether the file is compressed
         based on its file extension and pipe the file through an appropriate filter:
 
-        - .tar.gz: gunzip
-        - .tar.bz2: bunzip2
-        - .tar.xz: unxz
+        - .tar.gz or .tgz: gzip -d
+        - .tar.bz2 or .tbz: bzip2 -d
+        - .tar.xz or .txz: xz -d
+        - .tar.zstd: zstd -d
+        - .tar.lz4: lz4 -d
 
         Alternatively, a --tar-filter program may be explicitly specified. It should
         read compressed data from stdin and output an uncompressed tar stream on
diff --git a/src/borg/helpers/misc.py b/src/borg/helpers/misc.py
index ea8655ab5..573ab0132 100644
--- a/src/borg/helpers/misc.py
+++ b/src/borg/helpers/misc.py
@@ -238,3 +238,21 @@ def iter_separated(fd, sep=None, read_size=4096):
     # or if there was no data before EOF
     if len(part) > 0:
         yield part
+
+
+def get_tar_filter(fname):
+    # Note that filter is None if fname is '-'.
+    if fname.endswith(('.tar.gz', '.tgz')):
+        filter = 'gzip -d' if decompress else 'gzip'
+    elif fname.endswith(('.tar.bz2', '.tbz')):
+        filter = 'bzip2 -d' if decompress else 'bzip2'
+    elif fname.endswith(('.tar.xz', '.txz')):
+        filter = 'xz -d' if decompress else 'xz'
+    elif fname.endswith(('.tar.lz4', )):
+        filter = 'lz4 -d' if decompress else 'lz4'
+    elif fname.endswith(('.tar.zstd', )):
+        filter = 'zstd -d' if decompress else 'zstd'
+    else:
+        filter = None
+    logger.debug('Automatically determined tar filter: %s', filter)
+    return filter