-From rusty@rustcorp.com.au Wed Apr 3 17:18:42 2002
-Return-Path: <rusty@rustcorp.com.au>
-Delivered-To: mbp@samba.org
-Received: from wagner.rustcorp.com.au (sydney1.au.ibm.com [202.135.142.193])
- by lists.samba.org (Postfix) with ESMTP id EA7B849DC
- for <mbp@samba.org>; Tue, 2 Apr 2002 23:06:29 -0800 (PST)
-Received: from wagner.rustcorp.com.au ([127.0.0.1] helo=rustcorp.com.au)
- by wagner.rustcorp.com.au with esmtp (Exim 3.35 #1 (Debian))
- id 16set7-0000pL-00
- for <mbp@samba.org>; Wed, 03 Apr 2002 17:08:57 +1000
-From: Rusty Russell <rusty@rustcorp.com.au>
-To: Martin Pool <mbp@samba.org>
-Subject: Re: gzip patch
-In-reply-to: Your message of "Wed, 03 Apr 2002 12:04:59 +1000."
- <20020403020455.GC18851@samba.org>
-Date: Wed, 03 Apr 2002 17:08:57 +1000
-Sender: rusty@rustcorp.com.au
-Message-Id: <E16set7-0000pL-00@wagner.rustcorp.com.au>
-Status: RO
-X-Status: A
-Content-Length: 12810
-Lines: 461
+The changes to generator.c were greatly simplified, making the patch
+easier to maintain and fixing the failing test in the testsuite.
+Lightly tested.
-In message <20020403020455.GC18851@samba.org> you write:
-> Hi,
->
-> I think you said the other day that you had a working --rsyncable
-> patch for gzip. Could I have it please?
+Be sure to run "make proto" before "make".
-Hi Martin,
-
- Just got your mail, sorry for the delay. Found old patch on
-google, and updated it for 2.5.4 (I know, but that's what apt-get
-source gave me).
-
-Compiles, otherwise untested.
-Rusty.
---
- Anyone who quotes me in their sig is an idiot. -- Rusty Russell.
-
-diff -urN rsync-2.5.4/Makefile.in rsync-2.5.4-fuzzy/Makefile.in
---- rsync-2.5.4/Makefile.in Tue Feb 26 05:48:25 2002
-+++ rsync-2.5.4-fuzzy/Makefile.in Wed Apr 3 16:35:55 2002
-@@ -28,7 +28,7 @@
- ZLIBOBJ=zlib/deflate.o zlib/infblock.o zlib/infcodes.o zlib/inffast.o \
- zlib/inflate.o zlib/inftrees.o zlib/infutil.o zlib/trees.o \
- zlib/zutil.o zlib/adler32.o
--OBJS1=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
-+OBJS1=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 alternate.o
- OBJS2=options.o flist.o io.o compat.o hlink.o token.o uidlist.o socket.o fileio.o batch.o \
- clientname.o
- DAEMON_OBJ = params.o loadparm.o clientserver.o access.o connection.o authenticate.o
-diff -urN rsync-2.5.4/alternate.c rsync-2.5.4-fuzzy/alternate.c
---- rsync-2.5.4/alternate.c Thu Jan 1 10:00:00 1970
-+++ rsync-2.5.4-fuzzy/alternate.c Wed Apr 3 17:04:15 2002
-@@ -0,0 +1,117 @@
-+#include "rsync.h"
-+
-+extern char *compare_dest;
-+extern int verbose;
-+
-+/* Alternate methods for opening files, if local doesn't exist */
-+/* Sanity check that we are about to open regular file */
-+int do_open_regular(char *fname)
-+{
-+ STRUCT_STAT st;
-+
-+ if (do_stat(fname, &st) == 0 && S_ISREG(st.st_mode))
-+ return do_open(fname, O_RDONLY, 0);
-+
-+ return -1;
-+}
-+
-+static void split_names(char *fname, char **dirname, char **basename)
-+{
-+ char *slash;
-+
-+ slash = strrchr(fname, '/');
-+ if (slash) {
-+ *dirname = fname;
-+ *slash = '\0';
-+ *basename = slash+1;
-+ } else {
-+ *basename = fname;
-+ *dirname = ".";
-+ }
-+}
-+
-+static unsigned int measure_name(const char *name,
-+ const char *basename,
-+ const char *ext)
-+{
-+ int namelen = strlen(name);
-+ int extlen = strlen(ext);
-+ unsigned int score = 0;
-+
-+ /* Extensions must match */
-+ if (namelen <= extlen || strcmp(name+namelen-extlen, ext) != 0)
-+ return 0;
-+
-+ /* Now score depends on similarity of prefix */
-+ for (; *name==*basename && *name; name++, basename++)
-+ score++;
-+ return score;
-+}
-+
-+int open_alternate_base_fuzzy(const char *fname)
+--- orig/generator.c 2005-02-03 02:04:20
++++ generator.c 2005-02-03 02:11:10
+@@ -47,6 +47,7 @@ extern int size_only;
+ extern OFF_T max_size;
+ extern int io_timeout;
+ extern int protocol_version;
++extern int fuzzy_basis;
+ extern int always_checksum;
+ extern char *partial_dir;
+ extern char *basis_dir[];
+@@ -227,6 +228,88 @@ static void generate_and_send_sums(int f
+ unmap_file(mapbuf);
+ }
+
++/* Try to find a filename in the same dir as "fname" with a similar name.
++ *
++ * TODO:
++ * - We should be using a cache of names for the current dir, not
++ * re-reading the destination directory for every file.
++ * - We must not return an rsync tempfile from the current transfer.
++ * - If the highest-rated name is not a normal file, we should fall-
++ * back to the next highest-rated file.
++ * - We must not return a destination file that is being updated
++ * during the current transfer, even if we already processed it
++ * (since the receiver may not be done with it yet).
++ * - We must weed out any names that a daemon's config has excluded.
++ */
++static int find_fuzzy(const char *fname, char *buf, STRUCT_STAT *st_ptr)
+{
+ DIR *d;
+ struct dirent *di;
-+ char *basename, *dirname;
-+ char mangled_name[MAXPATHLEN];
++ char *basename, *dirname, *slash;
+ char bestname[MAXPATHLEN];
-+ unsigned int bestscore = 0;
-+ const char *ext;
++ int suf_len, basename_len;
++ uint32 lowest_dist = 0x7FFFFFFF;
++ const char *suf;
+
-+ /* FIXME: can we assume fname fits here? */
-+ strcpy(mangled_name, fname);
++ strlcpy(buf, fname, MAXPATHLEN);
++ if ((slash = strrchr(buf, '/')) != NULL) {
++ dirname = buf;
++ *slash = '\0';
++ basename = slash + 1;
++ } else {
++ basename = buf;
++ dirname = ".";
++ }
++ basename_len = strlen(basename);
+
-+ split_names(mangled_name, &dirname, &basename);
-+ d = opendir(dirname);
-+ if (!d) {
-+ rprintf(FERROR,"recv_generator opendir(%s): %s\n",
-+ dirname,strerror(errno));
++ if (!(d = opendir(dirname))) {
++ rsyserr(FERROR, errno, "recv_generator opendir(%s)", dirname);
+ return -1;
+ }
++ if (slash)
++ *slash = '/';
+
-+ /* Get final extension, eg. .gz; never full basename though. */
-+ ext = strrchr(basename + 1, '.');
-+ if (!ext)
-+ ext = basename + strlen(basename); /* ext = "" */
++ suf = find_filename_suffix(basename, basename_len, &suf_len);
+
++ bestname[0] = '\0';
+ while ((di = readdir(d)) != NULL) {
-+ const char *dname = d_name(di);
-+ unsigned int score;
++ const char *dname_suf, *dname = d_name(di);
++ uint32 dist;
++ int dname_len, dname_suf_len;
+
-+ if (strcmp(dname,".")==0 ||
-+ strcmp(dname,"..")==0)
++ if (dname[0] == '.' && (dname[1] == '\0'
++ || (dname[1] == '.' && dname[2] == '\0')))
+ continue;
-+
-+ score = measure_name(dname, basename, ext);
-+ if (verbose > 4)
-+ rprintf(FINFO,"fuzzy score for %s = %u\n",
-+ dname, score);
-+ if (score > bestscore) {
-+ strcpy(bestname, dname);
-+ bestscore = score;
++
++ dname_len = strlen(dname);
++ dname_suf = find_filename_suffix(dname, dname_len, &dname_suf_len);
++
++ dist = fuzzy_distance(dname, dname_len, basename, basename_len);
++ /* Add some extra weight to how well the suffixes match. */
++ dist += fuzzy_distance(dname_suf, dname_suf_len, suf, suf_len) * 10;
++ if (verbose > 4) {
++ rprintf(FINFO, "fuzzy distance for %s = %d (%d)\n",
++ dname, (int)(dist>>16), (int)(dist&0xFFFF));
++ }
++ if (dist <= lowest_dist) {
++ strlcpy(bestname, dname, sizeof bestname);
++ lowest_dist = dist;
+ }
+ }
+ closedir(d);
+
+ /* Found a candidate. */
-+ if (bestscore != 0) {
-+ char fuzzyname[MAXPATHLEN];
-+
-+ snprintf(fuzzyname,MAXPATHLEN,"%s/%s", dirname, bestname);
-+ if (verbose > 2)
-+ rprintf(FINFO,"fuzzy match %s->%s\n",
-+ fname, fuzzyname);
-+ return do_open_regular(fuzzyname);
++ if (bestname[0] != '\0') {
++ strlcpy(basename, bestname, MAXPATHLEN - (basename - buf));
++ if (verbose > 2) {
++ rprintf(FINFO, "fuzzy match %s->%s\n",
++ safe_fname(fname), buf);
++ }
++ return link_stat(buf, st_ptr, 0);
+ }
+ return -1;
+}
+
-+int open_alternate_base_comparedir(const char *fname)
-+{
-+ char fnamebuf[MAXPATHLEN];
-+ /* try the file at compare_dest instead */
-+ snprintf(fnamebuf,MAXPATHLEN,"%s/%s",compare_dest,fname);
-+
-+ /* FIXME: now follows symlinks... */
-+ return do_open_regular(fnamebuf);
-+}
-diff -urN rsync-2.5.4/generator.c rsync-2.5.4-fuzzy/generator.c
---- rsync-2.5.4/generator.c Fri Feb 8 03:36:12 2002
-+++ rsync-2.5.4-fuzzy/generator.c Wed Apr 3 17:00:06 2002
-@@ -42,11 +42,12 @@
- extern int always_checksum;
- extern int modify_window;
- extern char *compare_dest;
-+extern int fuzzy;
+ /* Acts on flist->file's ndx'th item, whose name is fname. If a directory,
+ * make sure it exists, and has the right permissions/timestamp info. For
+@@ -477,6 +560,15 @@ static void recv_generator(char *fname,
+ } else
+ partialptr = NULL;
- /* choose whether to skip a particular file */
- static int skip_file(char *fname,
-- struct file_struct *file, STRUCT_STAT *st)
-+ struct file_struct *file, const STRUCT_STAT *st)
- {
- if (st->st_size != file->length) {
- return 0;
-@@ -185,7 +186,61 @@
- return s;
- }
-
-+/* Returns -1 for can't open (null file), -2 for skip */
-+static int open_base_file(struct file_struct *file,
-+ char *fname,
-+ int statret,
-+ STRUCT_STAT *st)
-+{
-+ int fd = -1;
-+
-+ if (statret == 0) {
-+ if (S_ISREG(st->st_mode)) {
-+ if (update_only
-+ && cmp_modtime(st->st_mtime, file->modtime) > 0) {
-+ if (verbose > 1)
-+ rprintf(FINFO,"%s is newer\n",fname);
-+ return -2;
-+ }
-+ if (skip_file(fname, file, st)) {
-+ set_perms(fname, file, st, 1);
-+ return -2;
-+ }
-+ fd = do_open(fname, O_RDONLY, 0);
-+ if (fd == -1) {
-+ rprintf(FERROR,"failed to open %s, continuing : %s\n",fname,strerror(errno));
-+ return -1;
-+ } else
-+ return fd;
-+ } else {
-+ /* Try to use symlink contents */
-+ if (S_ISLNK(st->st_mode)) {
-+ fd = do_open_regular(fname);
-+ /* Don't delete yet; receiver will need it */
-+ } else {
-+ if (delete_file(fname) != 0) {
-+ if (fd != -1)
-+ close(fd);
-+ return -2;
-+ }
-+ }
++ if (statret == -1 && fuzzy_basis) {
++ if (find_fuzzy(fname, fnamecmpbuf, &st) == 0
++ && S_ISREG(st.st_mode)) {
++ statret = 0;
++ fnamecmp = fnamecmpbuf;
++ fnamecmp_type = FNAMECMP_FUZZY;
+ }
+ }
+
-+ if (fd == -1 && compare_dest != NULL)
-+ fd = open_alternate_base_comparedir(fname);
+ if (statret == -1) {
+ if (preserve_hard_links && hard_link_check(file, HL_SKIP))
+ return;
+@@ -505,6 +597,8 @@ static void recv_generator(char *fname,
-+ if (fd == -1 && fuzzy)
-+ fd = open_alternate_base_fuzzy(fname);
-+
-+ /* Update stat to understand size */
-+ if (fd != -1) {
-+ if (do_fstat(fd, st) != 0)
-+ rprintf(FERROR,"fstat %s : %s\n",fname,strerror(errno));
+ if (!compare_dest && fnamecmp_type <= FNAMECMP_BASIS_DIR_HIGH)
+ ;
++ else if (fnamecmp_type == FNAMECMP_FUZZY)
++ ;
+ else if (unchanged_file(fnamecmp, file, &st)) {
+ if (fnamecmp_type == FNAMECMP_FNAME)
+ set_perms(fname, file, &st, PERMS_REPORT);
+@@ -579,8 +673,24 @@ notify_others:
+ write_int(f_out, ndx);
+ if (protocol_version >= 29 && inplace && !read_batch)
+ write_byte(f_out, fnamecmp_type);
+- if (f_out_name >= 0)
++ if (f_out_name >= 0) {
+ write_byte(f_out_name, fnamecmp_type);
++ if (fnamecmp_type == FNAMECMP_FUZZY) {
++ uchar lenbuf[3], *lb = lenbuf;
++ int len = strlen(fnamecmpbuf);
++ if (len > 0x7F) {
++#if MAXPATHLEN > 0x7FFF
++ *lb++ = len / 0x10000 + 0x80;
++ *lb++ = len / 0x100;
++#else
++ *lb++ = len / 0x100 + 0x80;
++#endif
++ }
++ *lb = len;
++ write_buf(f_out_name, lenbuf, lb - lenbuf + 1);
++ write_buf(f_out_name, fnamecmpbuf, len);
++ }
+ }
-+
-+ return fd;
-+}
-
- /*
- * Acts on file number I from FLIST, whose name is fname.
-@@ -203,9 +258,6 @@
- struct sum_struct *s;
- int statret;
- struct file_struct *file = flist->files[i];
-- char *fnamecmp;
-- char fnamecmpbuf[MAXPATHLEN];
-- extern char *compare_dest;
- extern int list_only;
- extern int preserve_perms;
- extern int only_existing;
-@@ -341,82 +393,29 @@
- return;
- }
-
-- fnamecmp = fname;
--
-- if ((statret == -1) && (compare_dest != NULL)) {
-- /* try the file at compare_dest instead */
-- int saveerrno = errno;
-- snprintf(fnamecmpbuf,MAXPATHLEN,"%s/%s",compare_dest,fname);
-- statret = link_stat(fnamecmpbuf,&st);
-- if (!S_ISREG(st.st_mode))
-- statret = -1;
-- if (statret == -1)
-- errno = saveerrno;
-- else
-- fnamecmp = fnamecmpbuf;
-- }
--
-- if (statret == -1) {
-- if (errno == ENOENT) {
-- write_int(f_out,i);
-- if (!dry_run) send_sums(NULL,f_out);
-- } else {
-- if (verbose > 1)
-- rprintf(FERROR, RSYNC_NAME
-- ": recv_generator failed to open \"%s\": %s\n",
-- fname, strerror(errno));
-- }
-- return;
-- }
--
-- if (!S_ISREG(st.st_mode)) {
-- if (delete_file(fname) != 0) {
-- return;
-- }
--
-- /* now pretend the file didn't exist */
-- write_int(f_out,i);
-- if (!dry_run) send_sums(NULL,f_out);
-- return;
-- }
--
-- if (opt_ignore_existing && fnamecmp == fname) {
-- if (verbose > 1)
-- rprintf(FINFO,"%s exists\n",fname);
-- return;
-- }
--
-- if (update_only && cmp_modtime(st.st_mtime,file->modtime)>0 && fnamecmp == fname) {
-+ /* Failed to stat for some other reason. */
-+ if (statret == -1 && errno != ENOENT) {
- if (verbose > 1)
-- rprintf(FINFO,"%s is newer\n",fname);
-+ rprintf(FERROR, RSYNC_NAME
-+ ": recv_generator failed to open \"%s\": %s\n",
-+ fname, strerror(errno));
- return;
- }
-
-- if (skip_file(fname, file, &st)) {
-- if (fnamecmp == fname)
-- set_perms(fname,file,&st,1);
-- return;
-- }
--
-- if (dry_run) {
-- write_int(f_out,i);
-+ fd = open_base_file(file, fname, statret, &st);
-+ if (fd == -2)
- return;
-- }
--
-- if (whole_file) {
-- write_int(f_out,i);
-- send_sums(NULL,f_out);
-- return;
-- }
--
-- /* open the file */
-- fd = do_open(fnamecmp, O_RDONLY, 0);
-- if (fd == -1) {
-- rprintf(FERROR,RSYNC_NAME": failed to open \"%s\", continuing : %s\n",fnamecmp,strerror(errno));
-- /* pretend the file didn't exist */
-+ if ((whole_file || dry_run) && fd != -1) {
-+ close(fd);
-+ fd = -1;
-+ }
-+
-+ if (fd == -1) {
-+ /* the file didn't exist, or we can pretend it doesn't */
- write_int(f_out,i);
-- send_sums(NULL,f_out);
-+ if (!dry_run)
-+ send_sums(NULL,f_out);
+ if (dry_run || read_batch)
return;
- }
+--- orig/main.c 2005-01-30 10:07:21
++++ main.c 2005-01-14 18:33:15
+@@ -48,6 +48,7 @@ extern int keep_dirlinks;
+ extern int preserve_hard_links;
+ extern int protocol_version;
+ extern int recurse;
++extern int fuzzy_basis;
+ extern int relative_paths;
+ extern int rsync_port;
+ extern int whole_file;
+@@ -491,7 +492,8 @@ static int do_recv(int f_in,int f_out,st
+ int pid;
+ int status = 0;
+ int error_pipe[2], name_pipe[2];
+- BOOL need_name_pipe = (basis_dir[0] || partial_dir) && !dry_run;
++ BOOL need_name_pipe = (basis_dir[0] || partial_dir || fuzzy_basis)
++ && !dry_run;
-@@ -427,7 +426,7 @@
- }
+ /* The receiving side mustn't obey this, or an existing symlink that
+ * points to an identical file won't be replaced by the referent. */
+--- orig/options.c 2005-02-01 10:39:22
++++ options.c 2005-01-28 19:31:20
+@@ -90,6 +90,7 @@ int copy_unsafe_links = 0;
+ int size_only = 0;
+ int daemon_bwlimit = 0;
+ int bwlimit = 0;
++int fuzzy_basis = 0;
+ size_t bwlimit_writemax = 0;
+ int only_existing = 0;
+ int opt_ignore_existing = 0;
+@@ -303,6 +304,7 @@ 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," --fuzzy find similar file for basis when no dest file\n");
+ rprintf(F," -z, --compress compress file data\n");
+ rprintf(F," -C, --cvs-exclude auto-ignore files the same way CVS does\n");
+ rprintf(F," -f, --filter=RULE add a file-filtering RULE\n");
+@@ -408,6 +410,7 @@ 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", 0, POPT_ARG_NONE, &fuzzy_basis, 0, 0, 0 },
+ /* TODO: Should this take an optional int giving the compression level? */
+ {"compress", 'z', POPT_ARG_NONE, &do_compression, 0, 0, 0 },
+ {"stats", 0, POPT_ARG_NONE, &do_stats, 0, 0, 0 },
+@@ -1321,6 +1324,9 @@ void server_options(char **args,int *arg
+ if (!implied_dirs && !am_sender)
+ args[ac++] = "--no-implied-dirs";
- if (verbose > 3)
-- rprintf(FINFO,"gen mapped %s of size %.0f\n",fnamecmp,(double)st.st_size);
-+ rprintf(FINFO,"gen mapped %s of size %.0f\n",fname,(double)st.st_size);
-
- s = generate_sums(buf,st.st_size,adapt_block_size(file, block_size));
-
-diff -urN rsync-2.5.4/options.c rsync-2.5.4-fuzzy/options.c
---- rsync-2.5.4/options.c Thu Feb 28 09:49:57 2002
-+++ rsync-2.5.4-fuzzy/options.c Wed Apr 3 16:43:54 2002
-@@ -73,6 +73,7 @@
- #else
- int modify_window=0;
- #endif
-+int fuzzy=0;
- int blocking_io=-1;
-
- /** Network address family. **/
-@@ -245,6 +246,7 @@
- rprintf(F," --bwlimit=KBPS limit I/O bandwidth, KBytes per second\n");
- rprintf(F," --write-batch=PREFIX write batch fileset starting with PREFIX\n");
- rprintf(F," --read-batch=PREFIX read batch fileset starting with PREFIX\n");
-+ rprintf(F," --fuzzy use similar file as basis if it does't exist\n");
- rprintf(F," -h, --help show this help screen\n");
- #ifdef INET6
- rprintf(F," -4 prefer IPv4\n");
-@@ -340,6 +342,7 @@
- {"hard-links", 'H', POPT_ARG_NONE, &preserve_hard_links},
- {"read-batch", 0, POPT_ARG_STRING, &batch_prefix, OPT_READ_BATCH},
- {"write-batch", 0, POPT_ARG_STRING, &batch_prefix, OPT_WRITE_BATCH},
-+ {"fuzzy", 0, POPT_ARG_NONE, &fuzzy},
- #ifdef INET6
- {0, '4', POPT_ARG_VAL, &default_af_hint, AF_INET },
- {0, '6', POPT_ARG_VAL, &default_af_hint, AF_INET6 },
-@@ -757,7 +760,9 @@
- args[ac++] = "--compare-dest";
- args[ac++] = compare_dest;
- }
--
-+
-+ if (fuzzy && am_sender)
++ if (fuzzy_basis && am_sender)
+ args[ac++] = "--fuzzy";
-
++
*argc = ac;
- }
-diff -urN rsync-2.5.4/proto.h rsync-2.5.4-fuzzy/proto.h
---- rsync-2.5.4/proto.h Sat Feb 23 11:05:06 2002
-+++ rsync-2.5.4-fuzzy/proto.h Wed Apr 3 16:35:25 2002
-@@ -256,3 +256,6 @@
- int cmp_modtime(time_t file1, time_t file2);
- int _Insure_trap_error(int a1, int a2, int a3, int a4, int a5, int a6);
- int sys_gettimeofday(struct timeval *tv);
-+int do_open_regular(char *fname);
-+int open_alternate_base_fuzzy(const char *fname);
-+int open_alternate_base_comparedir(const char *fname);
-diff -urN rsync-2.5.4/receiver.c rsync-2.5.4-fuzzy/receiver.c
---- rsync-2.5.4/receiver.c Thu Feb 14 05:42:20 2002
-+++ rsync-2.5.4-fuzzy/receiver.c Wed Apr 3 16:46:46 2002
-@@ -36,6 +36,7 @@
- extern char *compare_dest;
- extern int make_backups;
- extern char *backup_suffix;
-+extern int fuzzy;
+ return;
- static struct delete_list {
- DEV64_T dev;
-@@ -307,8 +308,6 @@
- char *fname;
- char template[MAXPATHLEN];
- char fnametmp[MAXPATHLEN];
-- char *fnamecmp;
-- char fnamecmpbuf[MAXPATHLEN];
- struct map_struct *buf;
- int i;
- struct file_struct *file;
-@@ -366,28 +365,24 @@
- if (verbose > 2)
- rprintf(FINFO,"recv_files(%s)\n",fname);
+--- orig/receiver.c 2005-02-03 02:04:20
++++ receiver.c 2005-01-15 21:21:02
+@@ -256,6 +256,27 @@ static int receive_data(int f_in, char *
+ }
-- fnamecmp = fname;
--
- /* open the file */
-- fd1 = do_open(fnamecmp, O_RDONLY, 0);
-+ fd1 = do_open(fname, O_RDONLY, 0);
-- if ((fd1 == -1) && (compare_dest != NULL)) {
-- /* try the file at compare_dest instead */
-- snprintf(fnamecmpbuf,MAXPATHLEN,"%s/%s",
-- compare_dest,fname);
-- fnamecmp = fnamecmpbuf;
-- fd1 = do_open(fnamecmp, O_RDONLY, 0);
-- }
-+ if (fd1 == -1 && compare_dest != NULL)
-+ fd1 = open_alternate_base_comparedir(fname);
++static void read_gen_name(int fd, char *buf)
++{
++ int len = read_byte(fd);
++ if (len & 0x80) {
++#if MAXPATHLEN > 32767
++ uchar lenbuf[2];
++ read_buf(fd, (char *)lenbuf, 2);
++ len = (len & ~0x80) * 0x10000 + lenbuf[0] * 0x100 + lenbuf[1];
++#else
++ len = (len & ~0x80) * 0x100 + read_byte(fd);
++#endif
++ }
++ if (len >= MAXPATHLEN) {
++ rprintf(FERROR, "bogus data on generator name pipe\n");
++ exit_cleanup(RERR_PROTOCOL);
++ }
++
++ read_sbuf(fd, buf, len);
++}
+
-+ if (fd1 == -1 && fuzzy)
-+ fd1 = open_alternate_base_fuzzy(fname);
++
+ static void discard_receive_data(int f_in, OFF_T length)
+ {
+ receive_data(f_in, NULL, -1, 0, NULL, -1, length);
+@@ -395,6 +416,10 @@ int recv_files(int f_in, struct file_lis
+ case FNAMECMP_BACKUP:
+ fnamecmp = get_backup_name(fname);
+ break;
++ case FNAMECMP_FUZZY:
++ read_gen_name(f_in_name, fnamecmpbuf);
++ fnamecmp = fnamecmpbuf;
++ break;
+ default:
+ if (j >= basis_dir_cnt) {
+ rprintf(FERROR,
+--- orig/rsync.h 2005-02-03 02:04:20
++++ rsync.h 2005-01-19 18:36:47
+@@ -131,6 +131,7 @@
+ #define FNAMECMP_FNAME 0x80
+ #define FNAMECMP_PARTIAL_DIR 0x81
+ #define FNAMECMP_BACKUP 0x82
++#define FNAMECMP_FUZZY 0x83
- if (fd1 != -1 && do_fstat(fd1,&st) != 0) {
-- rprintf(FERROR,"fstat %s : %s\n",fnamecmp,strerror(errno));
-+ rprintf(FERROR,"fstat %s : %s\n",fname,strerror(errno));
- receive_data(f_in,NULL,-1,NULL,file->length);
- close(fd1);
- continue;
- }
+ /* For calling delete_file() */
+ #define DEL_DIR (1<<0)
+--- orig/rsync.yo 2005-02-01 10:39:23
++++ rsync.yo 2005-01-28 19:31:36
+@@ -354,6 +354,7 @@ 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
++ --fuzzy find similar file for basis when no dest
+ -z, --compress compress file data
+ -C, --cvs-exclude auto-ignore files in the same way CVS does
+ -f, --filter=RULE add a file-filtering RULE
+@@ -937,6 +938,14 @@ bf(--link-dest) from working properly fo
+ (or implied by bf(-a)). You can work-around this bug by avoiding the bf(-o) option
+ when sending to an old rsync.
- if (fd1 != -1 && !S_ISREG(st.st_mode)) {
-- rprintf(FERROR,"%s : not a regular file (recv_files)\n",fnamecmp);
-+ rprintf(FERROR,"%s : not a regular file (recv_files)\n",fname);
- receive_data(f_in,NULL,-1,NULL,file->length);
- close(fd1);
- continue;
-@@ -403,7 +398,7 @@
- if (fd1 != -1 && st.st_size > 0) {
- buf = map_file(fd1,st.st_size);
- if (verbose > 2)
-- rprintf(FINFO,"recv mapped %s of size %.0f\n",fnamecmp,(double)st.st_size);
-+ rprintf(FINFO,"recv mapped %s of size %.0f\n",fname,(double)st.st_size);
- } else {
- buf = NULL;
- }
-
++dit(bf(--fuzzy)) This option tells rsync that it should look around for a
++basis file for any destination file that is missing. The current algorithm
++looks for a similarly-named file in the same directory as the destination
++file, and, if found, uses that to try to speed up the transfer. Note that
++the use of the --delete option might get rid of any potential fuzzy-match
++files, so either use --delete-after or filename exclusions if you need to
++prevent this.
++
+ dit(bf(-z, --compress)) With this option, rsync compresses any data from
+ the files that it sends to the destination machine. This
+ option is useful on slow connections. The compression method used is the
+--- orig/util.c 2005-01-28 19:08:20
++++ util.c 2005-01-19 17:30:51
+@@ -1213,3 +1213,108 @@ void *_realloc_array(void *ptr, unsigned
+ return malloc(size * num);
+ return realloc(ptr, size * num);
+ }
++
++/* Take a filename and filename length and return the most significant
++ * filename suffix we can find. This ignores suffixes such as "~",
++ * ".bak", ".orig", ".~1~", etc. */
++const char *find_filename_suffix(const char *fn, int fn_len, int *len_ptr)
++{
++ const char *suf, *s;
++ BOOL had_tilde;
++ int s_len;
++
++ /* One or more dots at the start aren't a suffix. */
++ while (fn_len && *fn == '.') fn++, fn_len--;
++
++ /* Ignore the ~ in a "foo~" filename. */
++ if (fn_len > 1 && fn[fn_len-1] == '~')
++ fn_len--, had_tilde = True;
++ else
++ had_tilde = False;
++
++ /* Assume we don't find an suffix. */
++ suf = "";
++ *len_ptr = 0;
++
++ /* Find the last significant suffix. */
++ for (s = fn + fn_len - 1; fn_len > 1; ) {
++ while (*s != '.' && s != fn) s--;
++ if (s == fn)
++ break;
++ s_len = fn_len - (s - fn);
++ fn_len = s - fn;
++ if (s_len == 3) {
++ if (strcmp(s+1, "bak") == 0
++ || strcmp(s+1, "old") == 0)
++ continue;
++ } else if (s_len == 4) {
++ if (strcmp(s+1, "orig") == 0)
++ continue;
++ } else if (s_len > 2 && had_tilde
++ && s[1] == '~' && isdigit(s[2]))
++ continue;
++ *len_ptr = s_len;
++ suf = s;
++ /* Determine if the suffix is all digits. */
++ for (s++, s_len--; s_len > 0; s++, s_len--) {
++ if (!isdigit(*s))
++ return suf;
++ }
++ /* An all-digit suffix may not be that signficant. */
++ continue;
++ }
++
++ return suf;
++}
++
++/* This is an implementation of the Levenshtein distance algorithm. It
++ * was implemented to avoid needing a two-dimensional matrix (to save
++ * memory). It was also tweaked to try to factor in the ASCII distance
++ * between changed characters as a minor distance quantity. The normal
++ * Levenshtein units of distance (each signifying a single change between
++ * the two strings) are defined as a "UNIT". */
++
++#define UNIT (1 << 16)
++
++uint32 fuzzy_distance(const char *s1, int len1, const char *s2, int len2)
++{
++ uint32 a[MAXPATHLEN], diag, above, left, diag_inc, above_inc, left_inc;
++ int32 cost;
++ int i1, i2;
++
++ if (!len1 || !len2) {
++ if (!len1) {
++ s1 = s2;
++ len1 = len2;
++ }
++ for (i1 = 0, cost = 0; i1 < len1; i1++)
++ cost += s1[i1];
++ return (int32)len1 * UNIT + cost;
++ }
++
++ for (i2 = 0; i2 < len2; i2++)
++ a[i2] = (i2+1) * UNIT;
++
++ for (i1 = 0; i1 < len1; i1++) {
++ diag = i1 * UNIT;
++ above = (i1+1) * UNIT;
++ for (i2 = 0; i2 < len2; i2++) {
++ left = a[i2];
++ if ((cost = *((uchar*)s1+i1) - *((uchar*)s2+i2)) != 0) {
++ if (cost < 0)
++ cost = UNIT - cost;
++ else
++ cost = UNIT + cost;
++ }
++ diag_inc = diag + cost;
++ left_inc = left + UNIT + *((uchar*)s1+i1);
++ above_inc = above + UNIT + *((uchar*)s2+i2);
++ a[i2] = above = left < above
++ ? (left_inc < diag_inc ? left_inc : diag_inc)
++ : (above_inc < diag_inc ? above_inc : diag_inc);
++ diag = left;
++ }
++ }
++
++ return a[len2-1];
++}
Matt McCutchen's Web Site / rsync / rsync-patches.git / blobdiff