transliterate.diff

This patch adds an option --tr=BAD/GOOD to transliterate filenames.  It
can be used to remove characters illegal on the destination filesystem.
Jeff Weber expressed interest in this:

http://lists.samba.org/archive/rsync/2007-October/018996.html

To use this patch, run these commands for a successful build:

    patch -p1 <patches/transliterate.diff
    ./configure                                 (optional if already run)
    make

based-on: 6c8ca91c731b7bf2b081694bda85b7dadc2b7aff
diff --git a/flist.c b/flist.c
--- a/flist.c
+++ b/flist.c
@@ -78,6 +78,7 @@ extern uid_t our_uid;
 extern struct stats stats;
 extern char *filesfrom_host;
 extern char *usermap, *groupmap;
+extern char *tr_opt;
 
 extern struct name_num_item *file_sum_nni;
 
@@ -106,6 +107,8 @@ int file_old_total = 0; /* total of active items that will soon be gone */
 int flist_eof = 0; /* all the file-lists are now known */
 int xfer_flags_as_varint = 0;
 
+char tr_substitutions[256];
+
 #define NORMAL_NAME 0
 #define SLASH_ENDING_NAME 1
 #define DOTDIR_NAME 2
@@ -679,6 +682,23 @@ static void send_file_entry(int f, const char *fname, struct file_struct *file,
 		stats.total_size += F_LENGTH(file);
 }
 
+static void transliterate(char *path, int len)
+{
+	while (1) {
+		/* Find position of any char in tr_opt in path, or the end of the path. */
+		int span = strcspn(path, tr_opt);
+		if ((len -= span) == 0)
+			return;
+		path += span;
+		if ((*path = tr_substitutions[*(uchar*)path]) == '\0')
+			memmove(path, path+1, len--); /* copies the trailing '\0' too. */
+		else {
+			path++;
+			len--;
+		}
+	}
+}
+
 static struct file_struct *recv_file_entry(int f, struct file_list *flist, int xflags)
 {
 	static int64 modtime, atime;
@@ -750,9 +770,13 @@ static struct file_struct *recv_file_entry(int f, struct file_list *flist, int x
 			outbuf.len = 0;
 		}
 		thisname[outbuf.len] = '\0';
+		basename_len = outbuf.len;
 	}
 #endif
 
+	if (tr_opt)
+		transliterate(thisname, basename_len);
+
 	if (*thisname
 	 && (clean_fname(thisname, CFN_REFUSE_DOT_DOT_DIRS) < 0 || (!relative_paths && *thisname == '/'))) {
 		rprintf(FERROR, "ABORTING due to unsafe pathname from sender: %s\n", thisname);
@@ -2575,6 +2599,15 @@ struct file_list *recv_file_list(int f, int dir_ndx)
 			parse_name_map(usermap, True);
 		if (groupmap)
 			parse_name_map(groupmap, False);
+		if (tr_opt) { /* Parse FROM/TO string and populate tr_substitutions[] */
+			char *f, *t;
+			if ((t = strchr(tr_opt, '/')) != NULL)
+				*t++ = '\0';
+			else
+				t = "";
+			for (f = tr_opt; *f; f++)
+				tr_substitutions[*(uchar*)f] = *t ? *t++ : '\0';
+		}
 	}
 
 	start_read = stats.total_read;
diff --git a/options.c b/options.c
--- a/options.c
+++ b/options.c
@@ -211,6 +211,7 @@ int logfile_format_has_i = 0;
 int logfile_format_has_o_or_i = 0;
 int always_checksum = 0;
 int list_only = 0;
+char *tr_opt = NULL;
 
 #define MAX_BATCH_NAME_LEN 256	/* Must be less than MAXPATHLEN-13 */
 char *batch_name = NULL;
@@ -813,6 +814,7 @@ static struct poptOption long_options[] = {
   {"temp-dir",        'T', POPT_ARG_STRING, &tmpdir, 0, 0, 0 },
   {"iconv",            0,  POPT_ARG_STRING, &iconv_opt, 0, 0, 0 },
   {"no-iconv",         0,  POPT_ARG_NONE,   0, OPT_NO_ICONV, 0, 0 },
+  {"tr",               0,  POPT_ARG_STRING, &tr_opt, 0, 0, 0 },
   {"ipv4",            '4', POPT_ARG_VAL,    &default_af_hint, AF_INET, 0, 0 },
   {"ipv6",            '6', POPT_ARG_VAL,    &default_af_hint, AF_INET6, 0, 0 },
   {"8-bit-output",    '8', POPT_ARG_VAL,    &allow_8bit_chars, 1, 0, 0 },
@@ -2490,6 +2492,24 @@ int parse_arguments(int *argc_p, const char ***argv_p)
 		}
 	}
 
+	if (tr_opt) {
+		if (*tr_opt == '/' && tr_opt[1]) {
+			snprintf(err_buf, sizeof err_buf,
+				"Do not start the --tr arg with a slash\n");
+			return 0;
+		}
+		if (*tr_opt && *tr_opt != '/') {
+			need_unsorted_flist = 1;
+			arg = strchr(tr_opt, '/');
+			if (arg && strchr(arg+1, '/')) {
+				snprintf(err_buf, sizeof err_buf,
+					"--tr cannot transliterate slashes\n");
+				return 0;
+			}
+		} else
+			tr_opt = NULL;
+	}
+
 	if (trust_sender || am_server || read_batch)
 		trust_sender_args = trust_sender_filter = 1;
 	else if (old_style_args || filesfrom_host != NULL)
@@ -2958,6 +2978,12 @@ void server_options(char **args, int *argc_p)
 	if (relative_paths && !implied_dirs && (!am_sender || protocol_version >= 30))
 		args[ac++] = "--no-implied-dirs";
 
+	if (tr_opt) {
+		if (asprintf(&arg, "--tr=%s", tr_opt) < 0)
+			goto oom;
+		args[ac++] = arg;
+	}
+
 	if (write_devices && am_sender)
 		args[ac++] = "--write-devices";
 
diff --git a/rsync.1.md b/rsync.1.md
--- a/rsync.1.md
+++ b/rsync.1.md
@@ -555,6 +555,7 @@ has its own detailed description later in this manpage.
 --read-batch=FILE        read a batched update from FILE
 --protocol=NUM           force an older protocol version to be used
 --iconv=CONVERT_SPEC     request charset conversion of filenames
+--tr=BAD/GOOD            transliterate filenames
 --checksum-seed=NUM      set block/file checksum seed (advanced)
 --ipv4, -4               prefer IPv4
 --ipv6, -6               prefer IPv6
@@ -3755,6 +3756,22 @@ expand it.
     free to specify just the local charset for a daemon transfer (e.g.
     `--iconv=utf8`).
 
+0.  `--tr=BAD/GOOD`
+
+    Transliterates filenames on the receiver, after the iconv conversion (if
+    any).  This can be used to remove characters illegal on the destination
+    filesystem.  If you use this option, consider saving a "find . -ls" listing
+    of the source in the destination to help you determine the original
+    filenames in case of need.
+
+    The argument consists of a string of characters to remove, optionally
+    followed by a slash and a string of corresponding characters with which to
+    replace them.  The second string may be shorter, in which case any leftover
+    characters in the first string are simply deleted.  For example,
+    `--tr=':\/!'` replaces colons with exclamation marks and deletes
+    backslashes.  Slashes cannot be transliterated because it would cause
+    havoc.
+
 0.  `--ipv4`, `-4` or `--ipv6`, `-6`
 
     Tells rsync to prefer IPv4/IPv6 when creating sockets or running ssh.  This