aboutsummaryrefslogtreecommitdiff
path: root/tests/test_suffix.sh
diff options
context:
space:
mode:
Diffstat (limited to 'tests/test_suffix.sh')
-rwxr-xr-xtests/test_suffix.sh189
1 files changed, 189 insertions, 0 deletions
diff --git a/tests/test_suffix.sh b/tests/test_suffix.sh
new file mode 100755
index 00000000..f90ff7ab
--- /dev/null
+++ b/tests/test_suffix.sh
@@ -0,0 +1,189 @@
+#!/bin/sh
+
+###############################################################################
+#
+# Author: Jia Tan
+#
+# This file has been put into the public domain.
+# You can do whatever you want with this file.
+#
+###############################################################################
+
+# If xz was not built, skip this test.
+XZ=../src/xz/xz
+test -x "$XZ" || XZ=
+if test -z "$XZ"; then
+ echo "xz was not built, skipping this test."
+ exit 77
+fi
+
+# If compression or decompression support is missing, this test is skipped.
+# This isn't perfect because it does not specifically check for LZMA1/2
+# filters. Many of the other tests also assume LZMA1/2 support if encoders
+# or decoders are enabled.
+if grep 'define HAVE_ENCODERS' ../config.h > /dev/null \
+ && grep 'define HAVE_DECODERS' ../config.h > /dev/null ; then
+ :
+else
+ echo "Compression or decompression support is disabled, skipping this test."
+ exit 77
+fi
+
+# Create temporary input file. The file contents are not important.
+SUFFIX_INPUT="suffix_temp"
+SUFFIX_INPUT_FILES="$SUFFIX_INPUT"_files
+SUFFIX_INPUT_FILES0="$SUFFIX_INPUT"_files0
+
+# Remove possible leftover temporary files
+rm -f \
+ "$SUFFIX_INPUT" \
+ "$SUFFIX_INPUT.foo" \
+ "$SUFFIX_INPUT_FILES" \
+ "$SUFFIX_INPUT_FILES"
+
+echo "foobar" > "$SUFFIX_INPUT"
+
+# Test basic suffix when compressing with raw format.
+if "$XZ" -zfk --suffix=".foo" -Fraw --lzma1=preset=0 "$SUFFIX_INPUT" ; then
+ :
+else
+ echo "Failed to compress a file with a suffix set in raw format"
+ exit 1
+fi
+
+# Test the output file is named properly.
+if test -f "$SUFFIX_INPUT.foo" ; then
+ :
+else
+ echo "Raw format compressed output file not named properly"
+ exit 1
+fi
+
+# Expect an error when compressing with raw format without a suffix
+if "$XZ" -zfk -Fraw --lzma1=preset=0 "$SUFFIX_INPUT" 2> /dev/null; then
+ echo "Error not reported when compressing in raw format without a suffix"
+ exit 1
+fi
+
+# Expect an error when decompressing with raw format without a suffix
+if "$XZ" -df -Fraw --lzma1=preset=0 "$SUFFIX_INPUT.foo" 2> /dev/null; then
+ echo "Error not reported when decompressing in raw format without a suffix"
+ exit 1
+fi
+
+# Test basic decompression with raw format and a suffix. This will also
+# delete $SUFFIX_INPUT.foo
+if "$XZ" -df --suffix=".foo" -Fraw --lzma1=preset=0 "$SUFFIX_INPUT.foo"; then
+ :
+else
+ echo "Failed to decompress a file with a suffix set in raw format"
+ exit 1
+fi
+
+# Test basic compression with .xz format and a suffix
+if "$XZ" -zfk --suffix=".foo" --lzma2=preset=0 "$SUFFIX_INPUT" ; then
+ :
+else
+ echo "Failed to compress a file with a suffix set in .xz format"
+ exit 1
+fi
+
+# Test the output file is named properly.
+if test -f "$SUFFIX_INPUT.foo" ; then
+ :
+else
+ echo ".xz format compressed output file named properly"
+ exit 1
+fi
+
+# This will delete $SUFFIX_INPUT.foo
+if "$XZ" -df --suffix=".foo" "$SUFFIX_INPUT.foo"; then
+ :
+else
+ echo "Failed to decompress a file with a suffix set in .xz format"
+ exit 1
+fi
+
+# Test reading from stdin in raw mode. This was broken in
+# cc5aa9ab138beeecaee5a1e81197591893ee9ca0 and fixed in
+# 837ea40b1c9d4998cac4500b55171bf33e0c31a6
+if echo foo | "$XZ" -Fraw --lzma1=preset=0 > /dev/null ; then
+ :
+else
+ echo "Implicit write to stdout not detected"
+ exit 1
+fi
+
+# Create two temporary files to be used with --files and --files0.
+printf "$SUFFIX_INPUT\n" > "$SUFFIX_INPUT_FILES"
+printf "$SUFFIX_INPUT\0" > "$SUFFIX_INPUT_FILES0"
+
+# Test proper handling of --files/--files0 when no suffix is set. This
+# must result in an error because xz does not know how to rename the output
+# file from the input files. This caused a segmentation fault due to a
+# mistake in f481523baac946fa3bc13d79186ffaf0c0b818a7, which was fixed by
+# 0a601ddc89fd7e1370807c6b58964f361dfcd34a.
+if "$XZ" -Fraw --lzma1=preset=0 --files="$SUFFIX_INPUT_FILES" 2> /dev/null ; then
+ echo "Failed to report error when compressing a file specified by --files in raw mode without a suffix"
+ exit 1
+fi
+
+if "$XZ" -Fraw --lzma1=preset=0 --files0="$SUFFIX_INPUT_FILES0" 2> /dev/null ; then
+ echo "Failed to report error when compressing a file specified by --files0 in raw mode without a suffix"
+ exit 1
+fi
+
+# Test proper suffix usage in raw mode with --files and --files0.
+if "$XZ" -zfk -Fraw --lzma1=preset=0 --suffix=.foo --files="$SUFFIX_INPUT_FILES" ; then
+ :
+else
+ echo "Error compressing a file specified by --files in raw mode with a suffix set"
+ exit 1
+fi
+
+if test -f "$SUFFIX_INPUT.foo" ; then
+ :
+else
+ echo "Entry processed by --files not named properly"
+ exit 1
+fi
+
+# Remove the artifact so we can be sure the next test executes properly.
+rm "$SUFFIX_INPUT.foo"
+
+if "$XZ" -zfk -Fraw --lzma1=preset=0 --suffix=.foo --files0="$SUFFIX_INPUT_FILES0" ; then
+ :
+else
+ echo "Error compressing a file specified by --files0 in raw mode with a suffix set"
+ exit 1
+fi
+
+if test -f "$SUFFIX_INPUT.foo" ; then
+ :
+else
+ echo "Entry processed by --files0 not named properly"
+ exit 1
+fi
+
+# When the file type cannot be determined by xz, it will copy the contents
+# of the file only if -c,--stdout is used. This was broken by
+# 837ea40b1c9d4998cac4500b55171bf33e0c31a6 and fixed by
+# f481523baac946fa3bc13d79186ffaf0c0b818a7.
+if echo foo | "$XZ" -df > /dev/null 2>&1; then
+ echo "Failed to report error when decompressing unknown file type without -c,--stdout"
+ exit 1
+fi
+
+if echo foo | "$XZ" -dfc > /dev/null; then
+ :
+else
+ echo "Failed to copy input to standard out when decompressing unknown file type with -c,--stdout"
+ exit 1
+fi
+
+# Remove remaining test artifacts
+rm -f \
+ "$SUFFIX_INPUT" \
+ "$SUFFIX_INPUT.foo" \
+ "$SUFFIX_INPUT_FILES" \
+ "$SUFFIX_INPUT_FILES0"