mirror/rsync
1
0
Fork 0
mirror of https://github.com/RsyncProject/rsync.git synced 2026-03-10 18:37:40 -04:00
Code Issues Packages Projects Releases Wiki Activity
3,400 Commits 6 Branches 173 Tags
7e9059d60fcead9a5e0108032da35fada4587fbb
Commit Graph

3400 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Wayne Davison
7e9059d60f Fixed a just-introduced crash bug in the --fuzzy processing. 2005-03-14 22:22:42 +00:00
Wayne Davison
f3ebe1a77e A simple test to ensure that fuzzy processing is working. 2005-03-14 22:17:25 +00:00
Wayne Davison
301fb56ce9 Split the conditional-directory sending out of send_file_name() into 
a new function: send_if_directory().  This lets the code that is
recursively descending through the directories make its list of a
dir's contents and close the DIR handle before recursing into the
subdirs.  Also, the "recurse" var is just true/false once again.
 2005-03-14 17:30:15 +00:00
Wayne Davison
aa7a6e878b The "recurse" value is back to being just 1 or 0 (true or false). 2005-03-14 17:30:13 +00:00
Wayne Davison
0a39837a62 Got rid of "fudged_recurse". 2005-03-14 17:30:10 +00:00
Wayne Davison
b2e8a9b293 Got rid of an unused extern. 2005-03-14 17:06:08 +00:00
Wayne Davison
a98ad81760 Got rid of some code in f_name_cmp() that tried to make all the 
dirname pointers to equivalent strings have identical pointers.
 2005-03-14 03:36:56 +00:00
Wayne Davison
ccc51c8331 The --fuzzy code now handles a file->dirname that has an identical 
string as another file without being an identical pointer.
 2005-03-14 03:35:40 +00:00
Wayne Davison
ee171c6da9 Document the latest %i output. 2005-03-13 05:36:13 +00:00
Wayne Davison
d5609e969d Output a '*' at the start of the %i string when deleting. 2005-03-13 05:35:49 +00:00
Wayne Davison
927c806841 - Improved a couple error messages. 
- Improved a function name.
 2005-03-13 05:34:00 +00:00
Wayne Davison
2da9dda1c0 Some misc. improvements (I hope). 2005-03-12 23:54:05 +00:00
Wayne Davison
3117bc16a5 Improved two sentences. 2005-03-12 23:52:18 +00:00
Wayne Davison
717b04306a Tweaked the name of a variable. 2005-03-12 23:52:08 +00:00
Wayne Davison
271220c542 Mention --copy-dest. 2005-03-11 19:23:09 +00:00
Wayne Davison
566a874141 Re-enabled the --copy-dest part of the test. 2005-03-11 17:36:05 +00:00
Wayne Davison
967866d4df Added --copy-dest logic, and improved the updating of --compare-dest 
and --link-dest files that are up-to-date but have differing attributes.
 2005-03-11 17:36:03 +00:00
Wayne Davison
1de3e99bc5 Added --copy-dest logic. 2005-03-11 17:35:59 +00:00
Wayne Davison
3e13004b6b Tweaked the comment on copy_file(). 2005-03-11 17:35:57 +00:00
Wayne Davison
2f03ce67d6 Document --copy-dest. 2005-03-11 17:35:54 +00:00
Wayne Davison
b9232d45eb - Fixed the reading of the fuzzy xname from the socket. 
- Call read_item_attrs() with its new arg.
 2005-03-10 00:06:01 +00:00
Wayne Davison
6087ef2a84 - Changed read_item_attrs() to return the length of the xname string. 
- Tweaked the order of the args to write_item_attrs().
 2005-03-10 00:05:58 +00:00
Wayne Davison
1f1d368ad5 - Improved the error-checking for some delete_item() calls. 
- Move the non-regular-file delete-check above the alt-basis check
  where it belongs.
- Keep track of the real statret and real stat-struct for certain
  alt-basis scenarios (e.g. partial-dir and fuzzy) so that we send
  the right itemized change flags.
 2005-03-09 23:46:28 +00:00
Wayne Davison
dd18526e5b Mention the latest protocol-29 changes. 2005-03-09 18:55:09 +00:00
Wayne Davison
4d53c4dd46 We now handle the reading and writing of extra basis-file info: the 
fnamecmp_type byte, and the extra name (currently used for fuzzy
processing and hard-link status).
 2005-03-09 18:54:19 +00:00
Wayne Davison
9e4a8d29b5 Got rid of the name-pipe, so we now read the fnamecmp_type data over 
the socket for protocol >= 29, or handle it like the old days for
older protocol versions.  This means that we now validate this extra
data for safety (such as the fuzzy filename).
 2005-03-09 18:54:16 +00:00
Wayne Davison
ef20efcbb6 Made itemize() output the fnamecmp_type and the fuzzy name based on 
the new ITEM_BASIS_TYPE_FOLLOWS and ITEM_XNAME_FOLLOWS flags.  Got
rid of the name-pipe to the receiver.
 2005-03-09 18:54:12 +00:00
Wayne Davison
c70e07d9ac Got rid of the name-pipe from the generator to the receiver. 2005-03-09 18:54:09 +00:00
Wayne Davison
b7d4d28bb3 When itemizing, we now set ITEM_LOCAL_CHANGE and ITEM_XNAME_FOLLOWS. 2005-03-09 18:54:06 +00:00
Wayne Davison
9a6ed83f2c - Made an overflow in read_vstring() return an error instead of dying. 
- Got rid of a flush kluge that was needed for the name-pipe.
 2005-03-09 18:54:02 +00:00
Wayne Davison
fd84673e54 Handle the new way that 'c' and 'h' get output by "%i". 2005-03-09 18:53:58 +00:00
Wayne Davison
57b12568e6 Complain if a feature that requires protocol 29 doesn't get it. 2005-03-09 18:53:55 +00:00
Wayne Davison
f9a9f54720 Made the dest_option string non-static. 2005-03-09 18:53:53 +00:00
Wayne Davison
3019a9bafd Changed some of the ITEM_* defines. 2005-03-09 18:53:49 +00:00
Wayne Davison
278e3d4f6e Mention the latest bug-fix. 2005-03-09 04:00:45 +00:00
Wayne Davison
9b9dd06894 We need to mention any change to a directory, not just a time change. 
Yeah, this isn't very consistent with how files are treated, but it's
backward compatible.
 2005-03-09 04:00:20 +00:00
Wayne Davison
1f7e29b99c Fixed the change-report output for a directory that has no 
write permissions.
 2005-03-09 03:49:07 +00:00
Wayne Davison
c2f0e6e5e3 Backward compatibility fix in read_iflags() (for protocols < 29). 2005-03-09 02:25:34 +00:00
Wayne Davison
c2b11ba017 Backed out the hack that reversed ITEM_REPORT_XATTRS with 
ITEM_TRANSFER.  Yes, it allowed some kludge code that made backward
compatibility seamless, but it made it impossible to remove the hack
in the future.  This way, the backward compatibility is just
slightly inaccurate in the display of the first letter in the %i
output, and the only hack can be safely removed without causing
problems.
 2005-03-06 23:37:42 +00:00
Wayne Davison
f75a53e71b - When --max-delete is exceeded, we now count how many deletions 
would have happend, warn about the number skipped, and set
  io_error to IOERR_DEL_LIMIT.
- When dry_run > 1 (which indicates that the destinationdir is
  missing), skip deletions in that dir.  This fixes a bug in a
  copy that is creating the destination dir w/--delete enabled.
 2005-03-05 18:58:42 +00:00
Wayne Davison
054abde25f Handle new IOERR_DEL_LIMIT bit in io_error. 2005-03-05 18:58:38 +00:00
Wayne Davison
24cecf1365 Define the message for RERR_DEL_LIMIT. 2005-03-05 18:58:35 +00:00
Wayne Davison
821ff7f49a Added RERR_DEL_LIMIT. 2005-03-05 18:58:32 +00:00
Wayne Davison
ff3d3c32d5 Added IOERR_DEL_LIMIT. 2005-03-05 18:58:29 +00:00
Wayne Davison
3b2ef5b11c Mention that --max-delete must be non-zero. 2005-03-05 18:58:26 +00:00
Wayne Davison
0394e34a69 Moved the end_progress() call from match.c to sender.c so that we 
report progress on 0-length files when pushing files (the receiver
already called it, so we already produced progress on a 0-length
file when pulling).
 2005-03-05 17:51:23 +00:00
Wayne Davison
56efa56474 Fixed the elapsed time reported for 0-length files. 2005-03-05 17:49:46 +00:00
Wayne Davison
ed7e79553e Don't try to determine the phase we're in by looking at the value 
of csum_length -- it might have been computed to be SUM_LENGTH.
Instead, look at the "phase" variable directly.
 2005-03-05 16:42:52 +00:00
Wayne Davison
dec71e94f3 Added a hack that uses the ITEM_REPORT_XATTRS bit (which is the old 
ITEM_UPDATING bit) to make us compatible when sending/receiving bits
to/from an earlier pre-release.
 2005-03-05 04:34:06 +00:00
Wayne Davison
ac4f91a5ee Added a hack that sets the ITEM_REPORT_XATTRS bit (which is the old 
ITEM_UPDATING bit) when ITEM_TRANSFER or ITEM_LOCAL_CHANGE is set.
This lets us interact compatibly when sending itemized bits to an
earlier pre-release.
 2005-03-05 04:34:04 +00:00