thr3ads.net - rsync - How about a --min-size option, next to --max-size [Apr 2005]

If this information is useful, please help other people find it:
Share via:

Paul Slootman

2005-Apr-25 13:11 UTC

How about a --min-size option, next to --max-size

There's a rather old bug report in Debian's bug tracking system
(see http://bugs.debian.org/27126) about wanting to be able to specify
the maximum file size, as well as the minimum file size. Here's the
text:

    Sometimes, it's useful to specify a file size range one is
    interested in.

    For example, I'd like to keep an up-to-date mirror of Debian, but I
    currently have size problems, so I'd like to skip all files > some
    limit.

    I've also had the opposite problem (with ordinary mirror) - a server
    where there were lots of small junk files I was not interested in.


Both cases has its merits. Would a patch for --min-size be acceptable?


Paul Slootman

Paul Slootman

2005-Apr-28 15:22 UTC

head link

How about a --min-size option, next to --max-size

On Mon 25 Apr 2005, Paul Slootman wrote:
> Subject: How about a --min-size option, next to --max-size
> Both cases has its merits. Would a patch for --min-size be acceptable?
Well, I had no protests, so here's the patch :-)

Paul Slootman
-------------- next part --------------
diff -ru rsync-2.6.4/generator.c rsync-2.6.4.min-size/generator.c
--- rsync-2.6.4/generator.c	2005-04-28 17:13:57.000000000 +0200
+++ rsync-2.6.4.min-size/generator.c	2005-04-28 17:11:24.000000000 +0200
@@ -58,6 +58,7 @@
 extern int ignore_times;
 extern int size_only;
 extern OFF_T max_size;
+extern OFF_T min_size;
 extern int io_timeout;
 extern int io_error;
 extern int sock_f_out;
@@ -728,6 +729,13 @@
 		}
 		return;
 	}
+	if (min_size && S_ISREG(file->mode) && file->length <
min_size) {
+		if (verbose > 1) {
+			rprintf(FINFO, "%s is under min-size\n",
+				safe_fname(fname));
+		}
+		return;
+	}
 
 	if (preserve_links && S_ISLNK(file->mode)) {
 #ifdef SUPPORT_LINKS
@@ -1204,7 +1212,7 @@
 
 	phase++;
 	csum_length = SUM_LENGTH;
-	only_existing = max_size = opt_ignore_existing = 0;
+	only_existing = min_size = max_size = opt_ignore_existing = 0;
 	update_only = always_checksum = size_only = 0;
 	ignore_times = 1;
 	make_backups = 0; /* avoid a duplicate backup for inplace processing */
diff -ru rsync-2.6.4/options.c rsync-2.6.4.min-size/options.c
--- rsync-2.6.4/options.c	2005-04-28 17:13:57.000000000 +0200
+++ rsync-2.6.4.min-size/options.c	2005-04-28 17:06:57.000000000 +0200
@@ -98,6 +98,7 @@
 int need_messages_from_generator = 0;
 int max_delete = 0;
 OFF_T max_size = 0;
+OFF_T min_size = 0;
 int ignore_errors = 0;
 int modify_window = 0;
 int blocking_io = -1;
@@ -165,7 +166,7 @@
 static int itemize_changes = 0;
 static int refused_delete, refused_archive_part;
 static int refused_partial, refused_progress, refused_delete_before;
-static char *max_size_arg;
+static char *size_arg;
 static char partialdir_for_delayupdate[] = ".~tmp~";
 
 /** Local address to bind.  As a character string because it's
@@ -306,6 +307,7 @@
   rprintf(F,"     --ignore-errors         delete even if there are I/O
errors\n");
   rprintf(F,"     --force                 force deletion of directories
even if not empty\n");
   rprintf(F,"     --max-delete=NUM        don't delete more than NUM
files\n");
+  rprintf(F,"     --min-size=SIZE         don't transfer any file
smaller than SIZE\n");
   rprintf(F,"     --max-size=SIZE         don't transfer any file
larger than SIZE\n");
   rprintf(F,"     --partial               keep partially transferred
files\n");
   rprintf(F,"     --partial-dir=DIR       put a partially transferred file
into DIR\n");
@@ -360,7 +362,7 @@
 enum {OPT_VERSION = 1000, OPT_DAEMON, OPT_SENDER, OPT_EXCLUDE,
OPT_EXCLUDE_FROM,
       OPT_FILTER, OPT_COMPARE_DEST, OPT_COPY_DEST, OPT_LINK_DEST,
       OPT_INCLUDE, OPT_INCLUDE_FROM, OPT_MODIFY_WINDOW,
-      OPT_READ_BATCH, OPT_WRITE_BATCH, OPT_TIMEOUT, OPT_MAX_SIZE,
+      OPT_READ_BATCH, OPT_WRITE_BATCH, OPT_TIMEOUT, OPT_MAX_SIZE, OPT_MIN_SIZE,
       OPT_REFUSED_BASE = 9000};
 
 static struct poptOption long_options[] = {
@@ -423,7 +425,8 @@
   {"rsh",             'e', POPT_ARG_STRING, &shell_cmd,
0, 0, 0 },
   {"block-size",      'B', POPT_ARG_LONG,   &block_size,
0, 0, 0 },
   {"max-delete",       0,  POPT_ARG_INT,    &max_delete, 0, 0, 0
},
-  {"max-size",         0,  POPT_ARG_STRING, &max_size_arg, 
OPT_MAX_SIZE, 0, 0 },
+  {"max-size",         0,  POPT_ARG_STRING, &size_arg, 
OPT_MAX_SIZE, 0, 0 },
+  {"min-size",         0,  POPT_ARG_STRING, &size_arg, 
OPT_MIN_SIZE, 0, 0 },
   {"timeout",          0,  POPT_ARG_INT,    &io_timeout,
OPT_TIMEOUT, 0, 0 },
   {"temp-dir",        'T', POPT_ARG_STRING, &tmpdir, 0,
0, 0 },
   {"compare-dest",     0,  POPT_ARG_STRING, 0, OPT_COMPARE_DEST, 0, 0
},
@@ -798,31 +801,40 @@
 			break;
 
 		case OPT_MAX_SIZE:
-			for (arg = max_size_arg; isdigit(*(uchar*)arg); arg++) {}
-			if (*arg == '.')
-				for (arg++; isdigit(*(uchar*)arg); arg++) {}
-			switch (*arg) {
-			case 'k': case 'K':
-				max_size = atof(max_size_arg) * 1024;
-				break;
-			case 'm': case 'M':
-				max_size = atof(max_size_arg) * 1024*1024;
-				break;
-			case 'g': case 'G':
-				max_size = atof(max_size_arg) * 1024*1024*1024;
-				break;
-			case '\0':
-				max_size = atof(max_size_arg);
-				break;
-			default:
-				max_size = 0;
-				break;
-			}
-			if (max_size <= 0) {
-				snprintf(err_buf, sizeof err_buf,
-					"--max-size value is invalid: %s\n",
-					max_size_arg);
-				return 0;
+		case OPT_MIN_SIZE:
+			{ 
+			    OFF_T size;
+			    for (arg = size_arg; isdigit(*(uchar*)arg); arg++) {}
+			    if (*arg == '.')
+				    for (arg++; isdigit(*(uchar*)arg); arg++) {}
+			    switch (toupper(*arg)) {
+			    case 'K':
+				    size = atof(size_arg) * 1024;
+				    break;
+			    case 'M':
+				    size = atof(size_arg) * 1024*1024;
+				    break;
+			    case 'G':
+				    size = atof(size_arg) * 1024*1024*1024;
+				    break;
+			    case '\0':
+				    size = atof(size_arg);
+				    break;
+			    default:
+				    size = 0;
+				    break;
+			    }
+			    if (size <= 0) {
+				    snprintf(err_buf, sizeof err_buf,
+					    "--%s-size value is invalid: %s\n",
+					    opt == OPT_MAX_SIZE ? "max" : "min",
+					    size_arg);
+				    return 0;
+			    }
+			    if (opt == OPT_MAX_SIZE)
+				    max_size = size;
+			    else
+				    min_size = size;
 			}
 			break;
 
@@ -1333,7 +1345,7 @@
 
 	if (max_size && am_sender) {
 		args[ac++] = "--max-size";
-		args[ac++] = max_size_arg;
+		args[ac++] = size_arg;
 	}
 
 	if (io_timeout) {
diff -ru rsync-2.6.4/rsync.yo rsync-2.6.4.min-size/rsync.yo
--- rsync-2.6.4/rsync.yo	2005-03-31 05:14:09.000000000 +0200
+++ rsync-2.6.4.min-size/rsync.yo	2005-04-28 17:13:29.000000000 +0200
@@ -341,6 +341,7 @@
      --ignore-errors         delete even if there are I/O errors
      --force                 force deletion of dirs even if not empty
      --max-delete=NUM        don't delete more than NUM files
+     --min-size=SIZE         don't transfer any file smaller than SIZE
      --max-size=SIZE         don't transfer any file larger than SIZE
      --partial               keep partially transferred files
      --partial-dir=DIR       put a partially transferred file into DIR
@@ -754,6 +755,13 @@
 files or directories (NUM must be non-zero).
 This is useful when mirroring very large trees to prevent disasters.
 
+dit(bf(--min-size=SIZE)) This tells rsync to avoid transferring any
+file that is smaller than the specified SIZE. The SIZE value can be
+suffixed with a letter to indicate a size multiplier (K, M, or G) and
+may be a fractional value (e.g. "bf(--max-size=1.5m)").
+This can help in not transferring small junk files.
+This only affects regular files, not e.g. device nodes or symbolic links.
+
 dit(bf(--max-size=SIZE)) This tells rsync to avoid transferring any
 file that is larger than the specified SIZE. The SIZE value can be
 suffixed with a letter to indicate a size multiplier (K, M, or G) and

Wayne Davison

2005-Apr-28 17:25 UTC

head link

How about a --min-size option, next to --max-size

On Mon, Apr 25, 2005 at 03:10:52PM +0200, Paul Slootman
wrote:> Would a patch for --min-size be acceptable?
Firstly, thanks for the patch -- I've tweaked it a bit and checked it
into the patches subdir.  It also prompted me to move the code in
generator.c that checks the maximum size so that it doesn't apply to
non-regular files (since some non-regular files might have a bogus
size).  Finally, I used your idea of having a separate size-parsing
function in options.c (this doesn't change how the CVS version works,
but it does make the min-size patch smaller).

As for accepting --min-size, I'm considering it.  It would be nice to
have a more general solution to the problem of limiting the transfer
based on a file's attributes (rather than a profusion of rsync options),
but that doesn't seem to be on the horizon at the moment.

..wayne..

Reasonably Related Threads

Search for more apparently analagous threads

rsync - Apr 2005 - How about a --min-size option, next to --max-size

How about a --min-size option, next to --max-size

How about a --min-size option, next to --max-size

How about a --min-size option, next to --max-size

Reasonably Related Threads