-After applying this patch, run these commands for a successful build:
-
- ./prepare-source
- ./configure (optional if already run)
- make
-
-Jason M. Felice writes:
+Jason M. Felice wrote:
This patch adds the --link-by-hash=DIR option, which hard links received
files in a link farm arranged by MD4 file hash. The result is that the system
will only store one copy of the unique contents of each file, regardless of
the file's name.
+To use this patch, run these commands for a successful build:
+
+ patch -p1 <patches/link-by-hash.diff
+ ./prepare-source
+ ./configure
+ make
--- old/Makefile.in
+++ new/Makefile.in
DAEMON_OBJ = params.o loadparm.o clientserver.o access.o connection.o authenticate.o
popt_OBJS=popt/findme.o popt/popt.o popt/poptconfig.o \
popt/popthelp.o popt/poptparse.o
+--- old/flist.c
++++ new/flist.c
+@@ -57,6 +57,7 @@ extern int copy_unsafe_links;
+ extern int protocol_version;
+ extern int sanitize_paths;
+ extern struct stats stats;
++extern char *link_by_hash_dir;
+ extern struct file_list *the_file_list;
+ extern alloc_pool_t hlink_pool;
+
+@@ -624,7 +625,7 @@ static struct file_struct *recv_file_ent
+ }
+ #endif
+
+- if (always_checksum && S_ISREG(mode))
++ if ((always_checksum || link_by_hash_dir) && S_ISREG(mode))
+ extra_len += SUM_EXTRA_CNT * EXTRA_LEN;
+
+ if (file_length > 0xFFFFFFFFu && S_ISREG(mode))
--- old/hashlink.c
+++ new/hashlink.c
-@@ -0,0 +1,339 @@
+@@ -0,0 +1,336 @@
+/*
+ Copyright (C) Cronosys, LLC 2004
+
+
+extern char *link_by_hash_dir;
+
-+#if HAVE_LINK
++#ifdef HAVE_LINK
+
-+char* make_hash_name(struct file_struct *file)
++char *make_hash_name(struct file_struct *file)
+{
+ char hash[33], *dst;
-+ unsigned char *src;
-+ unsigned char c;
++ uchar c, *src = (uchar*)F_SUM(file);
+ int i;
+
-+ src = (unsigned char*)file->u.sum;
+ for (dst = hash, i = 0; i < 4; i++, src++) {
+ c = *src >> 4;
+ *(dst++) = (c >= 10) ? (c - 10 + 'a') : (c + '0');
+ char *linkname;
+ long last_fnbr;
+
-+ if (file->length == 0)
++ if (F_LENGTH(file) == 0)
+ return robust_rename(fnametmp, fname, NULL, 0644);
+
+ if (do_stat(hashname, &st) == -1) {
+ free(hashname);
+ return rc;
+}
-+
+#endif
--- old/options.c
+++ new/options.c
char *config_file = NULL;
char *shell_cmd = NULL;
char *logfile_name = NULL;
-@@ -347,6 +348,7 @@ void usage(enum logcode F)
+@@ -346,6 +347,7 @@ void usage(enum logcode F)
rprintf(F," --compare-dest=DIR also compare destination files relative to DIR\n");
rprintf(F," --copy-dest=DIR ... and include copies of unchanged files\n");
rprintf(F," --link-dest=DIR hardlink to files in DIR when unchanged\n");
rprintf(F," -z, --compress compress file data during the transfer\n");
rprintf(F," --compress-level=NUM explicitly set compression level\n");
rprintf(F," -C, --cvs-exclude auto-ignore files the same way CVS does\n");
-@@ -396,7 +398,7 @@ enum {OPT_VERSION = 1000, OPT_DAEMON, OP
+@@ -395,7 +397,7 @@ enum {OPT_VERSION = 1000, OPT_DAEMON, OP
OPT_FILTER, OPT_COMPARE_DEST, OPT_COPY_DEST, OPT_LINK_DEST, OPT_HELP,
OPT_INCLUDE, OPT_INCLUDE_FROM, OPT_MODIFY_WINDOW, OPT_MIN_SIZE, OPT_CHMOD,
OPT_READ_BATCH, OPT_WRITE_BATCH, OPT_ONLY_WRITE_BATCH, OPT_MAX_SIZE,
OPT_SERVER, OPT_REFUSED_BASE = 9000};
static struct poptOption long_options[] = {
-@@ -495,6 +497,7 @@ static struct poptOption long_options[]
+@@ -497,6 +499,7 @@ static struct poptOption long_options[]
{"compare-dest", 0, POPT_ARG_STRING, 0, OPT_COMPARE_DEST, 0, 0 },
{"copy-dest", 0, POPT_ARG_STRING, 0, OPT_COPY_DEST, 0, 0 },
{"link-dest", 0, POPT_ARG_STRING, 0, OPT_LINK_DEST, 0, 0 },
{"fuzzy", 'y', POPT_ARG_NONE, &fuzzy_basis, 0, 0, 0 },
{"compress", 'z', POPT_ARG_NONE, 0, 'z', 0, 0 },
{"compress-level", 0, POPT_ARG_INT, &def_compress_level, 'z', 0, 0 },
-@@ -1085,6 +1088,21 @@ int parse_arguments(int *argc, const cha
+@@ -1087,6 +1090,21 @@ int parse_arguments(int *argc, const cha
usage(FINFO);
exit_cleanup(0);
+ case OPT_LINK_BY_HASH:
-+#if HAVE_LINK
++#ifdef HAVE_LINK
+ arg = poptGetOptArg(pc);
+ if (sanitize_paths)
+ arg = sanitize_path(NULL, arg, NULL, 0, NULL);
default:
/* A large opt value means that set_refuse_options()
* turned this option off. */
-@@ -1735,6 +1753,11 @@ void server_options(char **args,int *arg
+@@ -1745,6 +1763,11 @@ void server_options(char **args,int *arg
}
}
args[ac++] = "--files-from";
--- old/receiver.c
+++ new/receiver.c
-@@ -50,6 +50,7 @@ extern int delay_updates;
- extern struct stats stats;
- extern char *stdout_format;
- extern char *tmpdir;
-+extern char *link_by_hash_dir;
- extern char *partial_dir;
- extern char *basis_dir[];
- extern struct file_list *the_file_list;
-@@ -124,12 +125,13 @@ static int get_tmpname(char *fnametmp, c
+@@ -124,12 +124,14 @@ int get_tmpname(char *fnametmp, char *fn
static int receive_data(int f_in, char *fname_r, int fd_r, OFF_T size_r,
- char *fname, int fd, OFF_T total_size)
-+ char *fname, int fd, OFF_T total_size, char *md4)
++ const char *fname, int fd, OFF_T total_size,
++ const char *md4)
{
static char file_sum1[MD4_SUM_LENGTH];
static char file_sum2[MD4_SUM_LENGTH];
}
static void handle_delayed_updates(struct file_list *flist, char *local_name)
-@@ -611,8 +622,12 @@ int recv_files(int f_in, struct file_lis
- rprintf(FINFO, "%s\n", fname);
+@@ -604,7 +615,7 @@ int recv_files(int f_in, struct file_lis
/* recv file data */
-+#if HAVE_LINK
-+ if (link_by_hash_dir)
-+ file->u.sum = new_array(char, MD4_SUM_LENGTH);
-+#endif
recv_ok = receive_data(f_in, fnamecmp, fd1, st.st_size,
-- fname, fd2, file->length);
-+ fname, fd2, file->length, file->u.sum);
+- fname, fd2, F_LENGTH(file));
++ fname, fd2, F_LENGTH(file), F_SUM(file));
log_item(log_code, file, &initial_stats, iflags, NULL);
--- old/rsync.c
+++ new/rsync.c
-@@ -48,6 +48,7 @@ extern int inplace;
+@@ -49,6 +49,7 @@ extern int inplace;
extern int keep_dirlinks;
extern int make_backups;
extern mode_t orig_umask;
+extern char *link_by_hash_dir;
extern struct stats stats;
+ extern struct file_list *the_file_list;
extern struct chmod_mode_struct *daemon_chmod_modes;
-
-@@ -271,8 +272,15 @@ void finish_transfer(char *fname, char *
+@@ -324,8 +325,15 @@ void finish_transfer(char *fname, char *
/* move tmp file over real file */
if (verbose > 2)
rprintf(FINFO, "renaming %s to %s\n", fnametmp, fname);
- ret = robust_rename(fnametmp, fname, partialptr,
- file->mode & INITACCESSPERMS);
-+#if HAVE_LINK
++#ifdef HAVE_LINK
+ if (link_by_hash_dir)
+ ret = link_by_hash(fnametmp, fname, file);
+ else
ret == -2 ? "copy" : "rename",
--- old/rsync.h
+++ new/rsync.h
-@@ -651,6 +651,14 @@ struct stats {
+@@ -685,6 +685,14 @@ struct stats {
int current_file_index;
};
#include "byteorder.h"
--- old/rsync.yo
+++ new/rsync.yo
-@@ -365,6 +365,7 @@ to the detailed description below for a
+@@ -367,6 +367,7 @@ to the detailed description below for a
--compare-dest=DIR also compare received files relative to DIR
--copy-dest=DIR ... and include copies of unchanged files
--link-dest=DIR hardlink to files in DIR when unchanged
Matt McCutchen's Web Site / rsync / rsync-patches.git / blobdiff