-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
-@@ -34,7 +34,7 @@ OBJS1=rsync.o generator.o receiver.o cle
- main.o checksum.o match.o syscall.o log.o backup.o
- OBJS2=options.o flist.o io.o compat.o hlink.o token.o uidlist.o socket.o \
- fileio.o batch.o clientname.o chmod.o
+based-on: 28b519c93b6db30b6520d46f8cd65160213fddd2
+diff --git a/Makefile.in b/Makefile.in
+--- a/Makefile.in
++++ b/Makefile.in
+@@ -37,7 +37,7 @@ OBJS1=flist.o rsync.o generator.o receiver.o cleanup.o sender.o exclude.o \
+ util.o main.o checksum.o match.o syscall.o log.o backup.o delete.o
+ OBJS2=options.o io.o compat.o hlink.o token.o uidlist.o socket.o hashtable.o \
+ fileio.o batch.o clientname.o chmod.o acls.o xattrs.o
-OBJS3=progress.o pipe.o
+OBJS3=progress.o pipe.o hashlink.o
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/hashlink.c
-+++ new/hashlink.c
+diff --git a/flist.c b/flist.c
+--- a/flist.c
++++ b/flist.c
+@@ -72,6 +72,7 @@ extern int unsort_ndx;
+ extern uid_t our_uid;
+ extern struct stats stats;
+ extern char *filesfrom_host;
++extern char *link_by_hash_dir;
+ extern char *usermap, *groupmap;
+
+ extern char curr_dir[MAXPATHLEN];
+@@ -910,7 +911,7 @@ static struct file_struct *recv_file_entry(int f, struct file_list *flist, int x
+ extra_len += EXTRA_LEN;
+ #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 SIZEOF_INT64 >= 8
+diff --git a/hashlink.c b/hashlink.c
+new file mode 100644
+--- /dev/null
++++ b/hashlink.c
@@ -0,0 +1,339 @@
+/*
+ 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');
+ }
+ *dst = 0;
+
-+ asprintf(&dst,"%s/%s",link_by_hash_dir,hash);
++ if (asprintf(&dst,"%s/%s",link_by_hash_dir,hash) < 0)
++ out_of_memory("make_hash_name");
+ return dst;
+}
+
+ *fnbr = this_fnbr;
+
+ hashfile = new_array(struct hashfile_struct, 1);
-+ asprintf(&hashfile->name,"%s/%s",hashname,
-+ di->d_name);
++ if (asprintf(&hashfile->name,"%s/%s",hashname, di->d_name) < 0)
++ out_of_memory("find_hashfiles");
+ if (do_stat(hashfile->name,&st) == -1) {
+ rsyserr(FERROR, errno, "stat failed: %s", hashfile->name);
+ kill_hashfile(hashfile);
+}
+
+
-+int link_by_hash(char *fnametmp,char *fname,struct file_struct *file)
++int link_by_hash(const char *fnametmp, const char *fname, struct file_struct *file)
+{
+ STRUCT_STAT st;
+ char *hashname = make_hash_name(file);
+ 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) {
+ }
+
+ first = 1;
-+ asprintf(&linkname,"%s/0",hashname);
++ if (asprintf(&linkname,"%s/0",hashname) < 0)
++ out_of_memory("link_by_hash");
+ rprintf(FINFO, "(1) linkname = %s\n", linkname);
+ } else {
+ struct hashfile_struct *hashfiles, *hashfile;
+
+ if (hashfiles == NULL) {
+ first = 1;
-+ asprintf(&linkname,"%s/0",hashname);
++ if (asprintf(&linkname,"%s/0",hashname) < 0)
++ out_of_memory("link_by_hash");
+ rprintf(FINFO, "(2) linkname = %s\n", linkname);
+ } else {
+ int fd;
+ kill_hashfile(hashfile);
+ } else {
+ first = 1;
-+ asprintf(&linkname, "%s/%ld", hashname,
-+ last_fnbr + 1);
++ if (asprintf(&linkname, "%s/%ld", hashname, last_fnbr + 1) < 0)
++ out_of_memory("link_by_hash");
+ rprintf(FINFO, "(4) linkname = %s\n", linkname);
+ }
+ }
+ if (errno == EMLINK) {
+ first = 1;
+ free(linkname);
-+ asprintf(&linkname,"%s/%ld",hashname,
-+ last_fnbr + 1);
++ if (asprintf(&linkname,"%s/%ld",hashname, last_fnbr + 1) < 0)
++ out_of_memory("link_by_hash");
+ rprintf(FINFO, "(5) linkname = %s\n", linkname);
+ rprintf(FINFO,"link-by-hash: max link count exceeded, starting new file \"%s\".\n", linkname);
+ } else {
+ free(hashname);
+ return rc;
+}
-+
+#endif
---- old/options.c
-+++ new/options.c
-@@ -144,6 +144,7 @@ char *backup_suffix = NULL;
+diff --git a/options.c b/options.c
+--- a/options.c
++++ b/options.c
+@@ -159,6 +159,7 @@ char *backup_suffix = NULL;
char *tmpdir = NULL;
char *partial_dir = NULL;
char *basis_dir[MAX_BASIS_DIRS+1];
+char *link_by_hash_dir = NULL;
char *config_file = NULL;
char *shell_cmd = NULL;
- char *log_format = NULL;
-@@ -337,6 +338,7 @@ void usage(enum logcode F)
+ char *logfile_name = NULL;
+@@ -758,6 +759,7 @@ 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," --link-by-hash=DIR create hardlinks by hash into DIR\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");
-@@ -383,7 +385,7 @@ enum {OPT_VERSION = 1000, OPT_DAEMON, OP
+ rprintf(F," --skip-compress=LIST skip compressing files with a suffix in LIST\n");
+@@ -810,7 +812,7 @@ 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_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_NO_D,
-+ OPT_NO_D, OPT_LINK_BY_HASH,
+- OPT_NO_D, OPT_APPEND, OPT_NO_ICONV, OPT_INFO, OPT_DEBUG,
++ OPT_NO_D, OPT_APPEND, OPT_NO_ICONV, OPT_INFO, OPT_DEBUG, OPT_LINK_BY_HASH,
+ OPT_USERMAP, OPT_GROUPMAP, OPT_CHOWN, OPT_BWLIMIT,
OPT_SERVER, OPT_REFUSED_BASE = 9000};
- static struct poptOption long_options[] = {
-@@ -477,6 +479,7 @@ static struct poptOption long_options[]
+@@ -954,6 +956,7 @@ 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 },
+ {"link-by-hash", 0, POPT_ARG_STRING, 0, OPT_LINK_BY_HASH, 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 },
-@@ -1062,6 +1065,21 @@ int parse_arguments(int *argc, const cha
- usage(FINFO);
- exit_cleanup(0);
+ {"fuzzy", 'y', POPT_ARG_VAL, &fuzzy_basis, 1, 0, 0 },
+ {"no-fuzzy", 0, POPT_ARG_VAL, &fuzzy_basis, 0, 0, 0 },
+ {"no-y", 0, POPT_ARG_VAL, &fuzzy_basis, 0, 0, 0 },
+@@ -1780,6 +1783,21 @@ int parse_arguments(int *argc_p, const char ***argv_p)
+ return 0;
+ #endif
+ case OPT_LINK_BY_HASH:
-+#if HAVE_LINK
++#ifdef HAVE_LINK
+ arg = poptGetOptArg(pc);
+ if (sanitize_paths)
-+ arg = sanitize_path(NULL, arg, NULL, 0);
++ arg = sanitize_path(NULL, arg, NULL, 0, SP_DEFAULT);
+ link_by_hash_dir = (char *)arg;
+ break;
+#else
default:
/* A large opt value means that set_refuse_options()
* turned this option off. */
-@@ -1710,6 +1728,11 @@ void server_options(char **args,int *arg
- }
- }
+@@ -2662,6 +2680,11 @@ void server_options(char **args, int *argc_p)
+ } else if (inplace)
+ args[ac++] = "--inplace";
+ if (link_by_hash_dir && am_sender) {
+ args[ac++] = "--link-by-hash";
if (files_from && (!am_sender || filesfrom_host)) {
if (filesfrom_host) {
args[ac++] = "--files-from";
---- old/receiver.c
-+++ new/receiver.c
-@@ -53,6 +53,7 @@ extern int delay_updates;
- extern struct stats stats;
- extern char *log_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
-
+diff --git a/receiver.c b/receiver.c
+--- a/receiver.c
++++ b/receiver.c
+@@ -217,11 +217,13 @@ int open_tmpfile(char *fnametmp, const char *fname, struct file_struct *file)
+ }
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 *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 char file_sum1[MAX_DIGEST_LEN];
struct map_struct *mapbuf;
struct sum_struct sum;
-+ struct mdfour mdfour_data;
++ md_context mdfour_data;
int32 len;
OFF_T offset = 0;
OFF_T offset2;
-@@ -149,6 +151,9 @@ static int receive_data(int f_in, char *
+@@ -257,6 +259,9 @@ static int receive_data(int f_in, char *fname_r, int fd_r, OFF_T size_r,
} else
mapbuf = NULL;
+
sum_init(checksum_seed);
- if (append_mode) {
-@@ -191,6 +196,8 @@ static int receive_data(int f_in, char *
+ if (append_mode > 0) {
+@@ -304,6 +309,8 @@ static int receive_data(int f_in, char *fname_r, int fd_r, OFF_T size_r,
cleanup_got_literal = 1;
sum_update(data, i);
if (fd != -1 && write_file(fd,data,i) != i)
goto report_write_error;
-@@ -217,6 +224,8 @@ static int receive_data(int f_in, char *
+@@ -331,6 +338,8 @@ static int receive_data(int f_in, char *fname_r, int fd_r, OFF_T size_r,
see_token(map, len);
sum_update(map, len);
+ mdfour_update(&mdfour_data, (uchar*)map, len);
}
- if (inplace) {
-@@ -257,6 +266,8 @@ static int receive_data(int f_in, char *
- }
+ if (updating_basis_or_equiv) {
+@@ -384,6 +393,9 @@ static int receive_data(int f_in, char *fname_r, int fd_r, OFF_T size_r,
+ if (sum_end(file_sum1) != checksum_len)
+ overflow_exit("checksum_len"); /* Impossible... */
- sum_end(file_sum1);
+ if (md4)
-+ mdfour_result(&mdfour_data, (unsigned char*)md4);
-
++ mdfour_result(&mdfour_data, (uchar*)md4);
++
if (mapbuf)
unmap_file(mapbuf);
-@@ -272,7 +283,7 @@ static int receive_data(int f_in, char *
+
+@@ -398,7 +410,7 @@ static int receive_data(int f_in, char *fname_r, int fd_r, OFF_T size_r,
static void discard_receive_data(int f_in, OFF_T length)
{
+ receive_data(f_in, NULL, -1, 0, NULL, -1, length, NULL);
}
- static void handle_delayed_updates(struct file_list *flist, char *local_name)
-@@ -604,8 +615,12 @@ int recv_files(int f_in, struct file_lis
- rprintf(FINFO, "%s\n", fname);
+ static void handle_delayed_updates(char *local_name)
+@@ -828,7 +840,7 @@ int recv_files(int f_in, int f_out, char *local_name)
/* 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, iflags, NULL);
- if (!log_before_transfer)
- log_item(file, &initial_stats, iflags, NULL);
---- old/rsync.c
-+++ new/rsync.c
-@@ -49,6 +49,7 @@ extern int inplace;
+diff --git a/rsync.c b/rsync.c
+--- a/rsync.c
++++ b/rsync.c
+@@ -47,6 +47,7 @@ extern int flist_eof;
+ extern int file_old_total;
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 *cur_flist, *first_flist, *dir_flist;
extern struct chmod_mode_struct *daemon_chmod_modes;
-
-@@ -269,8 +270,15 @@ void finish_transfer(char *fname, char *
+ #ifdef ICONV_OPTION
+@@ -648,7 +649,12 @@ int finish_transfer(const char *fname, const char *fnametmp,
/* move tmp file over real file */
- if (verbose > 2)
+ if (DEBUG_GTE(RECV, 1))
rprintf(FINFO, "renaming %s to %s\n", fnametmp, fname);
-- ret = robust_rename(fnametmp, fname, partialptr,
-- file->mode & INITACCESSPERMS);
-+#if HAVE_LINK
+- ret = robust_rename(fnametmp, fname, temp_copy_name, file->mode);
++#ifdef HAVE_LINK
+ if (link_by_hash_dir)
+ ret = link_by_hash(fnametmp, fname, file);
+ else
+#endif
-+ {
-+ ret = robust_rename(fnametmp, fname, partialptr,
-+ file->mode & INITACCESSPERMS);
-+ }
++ ret = robust_rename(fnametmp, fname, temp_copy_name, file->mode);
if (ret < 0) {
- rsyserr(FERROR, errno, "%s %s -> \"%s\"",
+ rsyserr(FERROR_XFER, errno, "%s %s -> \"%s\"",
ret == -2 ? "copy" : "rename",
---- old/rsync.h
-+++ new/rsync.h
-@@ -640,6 +640,14 @@ struct stats {
- int current_file_index;
+diff --git a/rsync.h b/rsync.h
+--- a/rsync.h
++++ b/rsync.h
+@@ -899,6 +899,14 @@ struct stats {
+ int xferred_files;
};
+struct hashfile_struct {
+
struct chmod_mode_struct;
- #include "byteorder.h"
---- old/rsync.yo
-+++ new/rsync.yo
-@@ -363,6 +363,7 @@ to the detailed description below for a
+ struct flist_ndx_item {
+diff --git a/rsync.yo b/rsync.yo
+--- a/rsync.yo
++++ b/rsync.yo
+@@ -402,6 +402,7 @@ to the detailed description below for a complete description. verb(
--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
+ --link-by-hash=DIR create hardlinks by hash into DIR
-z, --compress compress file data during the transfer
--compress-level=NUM explicitly set compression level
- -C, --cvs-exclude auto-ignore files in the same way CVS does
+ --skip-compress=LIST skip compressing files with suffix in LIST
git.samba.org / rsync-patches.git / blobdiff