hardlink: add new option: -S/--maximum-size

Support limiting comparison by maximum file size.

[kzak@redhat.com: - resolve conflicts]

Signed-off-by: Daniele Pizzolli <dan+dev@toel.it>
Signed-off-by: Karel Zak <kzak@redhat.com>

diff --git a/misc-utils/hardlink.1.adoc b/misc-utils/hardlink.1.adoc
index 36f8bfdbf7ff5747423c439d3d09d5d942dfe5d7..f45b468358e6d79168f51c9412394a6a6c4745d9 100644 (file)
--- a/misc-utils/hardlink.1.adoc
+++ b/misc-utils/hardlink.1.adoc
@@ -87,6 +87,9 @@ A regular expression to include files. If the option *--exclude* has been given,
 *-s*, *--minimum-size* _size_::
 The minimum size to consider. By default this is 1, so empty files will not be linked. The _size_ argument may be followed by the multiplicative suffixes KiB (=1024), MiB (=1024*1024), and so on for GiB, TiB, PiB, EiB, ZiB and YiB (the "iB" is optional, e.g., "K" has the same meaning as "KiB").
 
+*-S*, *--maximum-size* _size_::
+The maximum size to consider. By default this is 0 and 0 has the special meaning of unlimited. The _size_ argument may be followed by the multiplicative suffixes KiB (=1024), MiB (=1024*1024), and so on for GiB, TiB, PiB, EiB, ZiB and YiB (the "iB" is optional, e.g., "K" has the same meaning as "KiB").
+
 *-b*, *--io-size* _size_::
 The size of the read() or sendfile() buffer used when comparing file contents.
 The _size_ argument may be followed by the multiplicative suffixes KiB, MiB,

diff --git a/misc-utils/hardlink.c b/misc-utils/hardlink.c
index 20998ffeaf8b40d41c08dd3b5d3a265029ea88ed..ed49b0a7f51176ab3e2adb2fb2a471864cca9fb1 100644 (file)
--- a/misc-utils/hardlink.c
+++ b/misc-utils/hardlink.c
@@ -138,6 +138,7 @@ struct hdl_regex {
  * @keep_oldest: Choose the file with oldest timestamp as master (default = FALSE)
  * @dry_run: Specifies whether hardlink should not link files (default = FALSE)
  * @min_size: Minimum size of files to consider. (default = 1 byte)
+ * @max_size: Maximum size of files to consider, 0 means umlimited. (default = 0 byte)
  */
 static struct options {
        struct hdl_regex *include;
@@ -155,6 +156,7 @@ static struct options {
        unsigned int keep_oldest:1;
        unsigned int dry_run:1;
        uintmax_t min_size;
+       uintmax_t max_size;
        size_t io_size;
        size_t cache_size;
 } opts = {
@@ -723,6 +725,12 @@ static int inserter(const char *fpath, const struct stat *sb,
        jlog(JLOG_VERBOSE2, " %5zu: [%ld/%ld/%ld] %s",
                        stats.files, sb->st_dev, sb->st_ino, sb->st_nlink, fpath);
 
+       if ((opts.max_size > 0) && ((uintmax_t) sb->st_size > opts.max_size)) {
+               jlog(JLOG_VERBOSE1,
+                    _("Skipped %s (greater than configured size)"), fpath);
+               return 0;
+       }
+
        pathlen = strlen(fpath) + 1;
 
        fil = xcalloc(1, sizeof(*fil));
@@ -915,6 +923,7 @@ static void __attribute__((__noreturn__)) usage(void)
        fputs(_(" -x, --exclude <regex>      regular expression to exclude files\n"), out);
        fputs(_(" -i, --include <regex>      regular expression to include files/dirs\n"), out);
        fputs(_(" -s, --minimum-size <size>  minimum size for files.\n"), out);
+       fputs(_(" -S, --maximum-size <size>  maximum size for files.\n"), out);
        fputs(_(" -b, --io-size <size>       I/O buffer size for file reading (speedup, using more RAM)\n"), out);
        fputs(_(" -r, --cache-size <size>    memory limit for cached file content data\n"), out);
        fputs(_(" -c, --content              compare only file contents, same as -pot\n"), out);
@@ -933,7 +942,7 @@ static void __attribute__((__noreturn__)) usage(void)
  */
 static int parse_options(int argc, char *argv[])
 {
-       static const char optstr[] = "VhvnfpotXcmMOx:y:i:r:s:b:q";
+       static const char optstr[] = "VhvnfpotXcmMOx:y:i:r:S:s:b:q";
        static const struct option long_options[] = {
                {"version", no_argument, NULL, 'V'},
                {"help", no_argument, NULL, 'h'},
@@ -951,6 +960,7 @@ static int parse_options(int argc, char *argv[])
                {"include", required_argument, NULL, 'i'},
                {"method", required_argument, NULL, 'y' },
                {"minimum-size", required_argument, NULL, 's'},
+               {"maximum-size", required_argument, NULL, 'S'},
                {"io-size", required_argument, NULL, 'b'},
                {"content", no_argument, NULL, 'c'},
                {"quiet", no_argument, NULL, 'q'},
@@ -1019,7 +1029,10 @@ static int parse_options(int argc, char *argv[])
                        register_regex(&opts.include, optarg);
                        break;
                case 's':
-                       opts.min_size = strtosize_or_err(optarg, _("failed to parse size"));
+                       opts.min_size = strtosize_or_err(optarg, _("failed to parse minimum size"));
+                       break;
+               case 'S':
+                       opts.max_size = strtosize_or_err(optarg, _("failed to parse maximum size"));
                        break;
                case 'r':
                        opts.cache_size = strtosize_or_err(optarg, _("failed to cache size"));

diff --git a/tests/expected/hardlink/options-maximum-size-8191 b/tests/expected/hardlink/options-maximum-size-8191
new file mode 100644 (file)
index 0000000..cc718d4
--- /dev/null
+++ b/tests/expected/hardlink/options-maximum-size-8191
@@ -0,0 +1,59 @@
+Skipped [Redacted]/hardlink/testdir1/file-a-2 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-c-2 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-1/sdir-1/file-a-2 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-1/sdir-1/file-c-2 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-1/sdir-1/file-b-1 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-1/sdir-1/file-c-1 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-1/sdir-1/file-b-2 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-1/sdir-1/file-c-3 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-1/sdir-1/file-b-3 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-1/sdir-1/file-a-3 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-1/sdir-1/file-a-1 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-1/sdir-2/file-a-1-abcdefghijklmnopqrstxyz-"§$%&()=?*+ (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-b-4 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-b-1 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-a-5 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-a-4 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-2/sdir-2/file-a-5 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-2/sdir-2/file-b-5 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/dir-2/sdir-3/file-b-4 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-c-1 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-b-5 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-b-2 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-c-3 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-b-3 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-a-3 (greater than configured size)
+Skipped [Redacted]/hardlink/testdir1/file-a-1 (greater than configured size)
+Mode:           real
+Files:          26
+Linked:         0 files
+Compared:       0 xattrs
+Compared:       0 files
+Saved:          0 B
+Duration: [Redacted]
+dir-1/sdir-1/file-a-1  10      8192    1540236423      644
+dir-1/sdir-1/file-a-2  10      8192    1540236423      644
+dir-1/sdir-1/file-a-3  10      8192    1540236423      644
+dir-1/sdir-1/file-b-1  10      8192    1540236430      644
+dir-1/sdir-1/file-b-2  10      8192    1540236430      644
+dir-1/sdir-1/file-b-3  10      8192    1540236430      644
+dir-1/sdir-1/file-c-1  6       8192    1540236548      644
+dir-1/sdir-1/file-c-2  6       8192    1540236548      644
+dir-1/sdir-1/file-c-3  6       8192    1540236548      644
+dir-1/sdir-2/file-a-1-abcdefghijklmnopqrstxyz-"§$%&()=?*+     10      8192    1540236423      644
+dir-2/sdir-2/file-a-5  10      8192    1540236423      644
+dir-2/sdir-2/file-b-5  10      8192    1540236430      644
+dir-2/sdir-3/file-b-4  10      8192    1540236430      644
+file-a-1       10      8192    1540236423      644
+file-a-2       10      8192    1540236423      644
+file-a-3       10      8192    1540236423      644
+file-a-4       10      8192    1540236423      644
+file-a-5       10      8192    1540236423      644
+file-b-1       10      8192    1540236430      644
+file-b-2       10      8192    1540236430      644
+file-b-3       10      8192    1540236430      644
+file-b-4       10      8192    1540236430      644
+file-b-5       10      8192    1540236430      644
+file-c-1       6       8192    1540236548      644
+file-c-2       6       8192    1540236548      644
+file-c-3       6       8192    1540236548      644

diff --git a/tests/expected/hardlink/options-maximum-size-8192 b/tests/expected/hardlink/options-maximum-size-8192
new file mode 100644 (file)
index 0000000..7308553
--- /dev/null
+++ b/tests/expected/hardlink/options-maximum-size-8192
@@ -0,0 +1,33 @@
+Mode:           real
+Files:          26
+Linked:         0 files
+Compared:       0 xattrs
+Compared:       0 files
+Saved:          0 B
+Duration: [Redacted]
+dir-1/sdir-1/file-a-1  10      8192    1540236423      644
+dir-1/sdir-1/file-a-2  10      8192    1540236423      644
+dir-1/sdir-1/file-a-3  10      8192    1540236423      644
+dir-1/sdir-1/file-b-1  10      8192    1540236430      644
+dir-1/sdir-1/file-b-2  10      8192    1540236430      644
+dir-1/sdir-1/file-b-3  10      8192    1540236430      644
+dir-1/sdir-1/file-c-1  6       8192    1540236548      644
+dir-1/sdir-1/file-c-2  6       8192    1540236548      644
+dir-1/sdir-1/file-c-3  6       8192    1540236548      644
+dir-1/sdir-2/file-a-1-abcdefghijklmnopqrstxyz-"§$%&()=?*+     10      8192    1540236423      644
+dir-2/sdir-2/file-a-5  10      8192    1540236423      644
+dir-2/sdir-2/file-b-5  10      8192    1540236430      644
+dir-2/sdir-3/file-b-4  10      8192    1540236430      644
+file-a-1       10      8192    1540236423      644
+file-a-2       10      8192    1540236423      644
+file-a-3       10      8192    1540236423      644
+file-a-4       10      8192    1540236423      644
+file-a-5       10      8192    1540236423      644
+file-b-1       10      8192    1540236430      644
+file-b-2       10      8192    1540236430      644
+file-b-3       10      8192    1540236430      644
+file-b-4       10      8192    1540236430      644
+file-b-5       10      8192    1540236430      644
+file-c-1       6       8192    1540236548      644
+file-c-2       6       8192    1540236548      644
+file-c-3       6       8192    1540236548      644

diff --git a/tests/ts/hardlink/options b/tests/ts/hardlink/options
index e293921a1f7f6fda40717c1fb2d7505a6eb4451a..d86acf395f4269eaf6864216f6a07c7c2fc74c8b 100755 (executable)
--- a/tests/ts/hardlink/options
+++ b/tests/ts/hardlink/options
@@ -76,5 +76,22 @@ $TS_CMD_HARDLINK --quiet --content "$SRCDIR" >> $TS_OUTPUT 2>> $TS_ERRLOG
 show_srcdir | sed 's/\(1540236\).*/\1xxx\tperm/' >> $TS_OUTPUT 2>> $TS_ERRLOG
 ts_finalize_subtest
 
+ts_init_subtest "maximum-size-8191"
+# Redact path and duration for reproduciblity
+$TS_CMD_HARDLINK -vv --maximum-size 8191 "$SRCDIR" | \
+    sed 's:^Skipped .*/tests/output/\(.*\):Skipped [Redacted]/\1:' | \
+       sed 's/^Duration: .*/Duration: [Redacted]/' >> $TS_OUTPUT 2>> $TS_ERRLOG
+show_srcdir >> $TS_OUTPUT 2>> $TS_ERRLOG
+ts_finalize_subtest
+
+ts_init_subtest "maximum-size-8192"
+# Redact path and duration for reproduciblity
+$TS_CMD_HARDLINK -vv --maximum-size 8192 "$SRCDIR" | \
+    sed 's:^Skipped .*/tests/output/\(.*\):Skipped [Redacted]/\1:' | \
+       sed 's/^Duration:.*/Duration: [Redacted]/' >> $TS_OUTPUT 2>> $TS_ERRLOG
+show_srcdir >> $TS_OUTPUT 2>> $TS_ERRLOG
+ts_finalize_subtest
+
+
 rm -rf "$SRCDIR"
 ts_finalize