summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* - Fixed a bug where the --*-dest options weren't being culled.Wayne Davison2005-06-171-25/+17
| | | | | - Simplified the output format by putting all the long-name options into a single hash.
* - Improved some comments (some taken from the new cull-options output).Wayne Davison2005-06-171-14/+23
| | | | | | | | - Made the user-config sections easier to find. - Default to running a fully-specified rsync executable so that there are not PATH surprises. - A few minor tweaks to the arg-checking code to make it clearer without actually changing its logic.
* Another readability improvement in the perl code we output.Wayne Davison2005-06-171-2/+5
|
* Improved some of the comments that are output into the perl code.Wayne Davison2005-06-171-9/+15
|
* - Make sure that the command contained the '.' arg at the end of theWayne Davison2005-06-171-21/+16
| | | | | options. - Fixed the expanding of wildcards for a restriction of '/'.
* - Be restrictive on matching the --sender option so that the userWayne Davison2005-06-171-10/+12
| | | | | | | can't fool us into thinking we're the sender by using a filename of --sender (or similar ploy). - Make sure that our '..' check can't be fooled by sly backslashing. - Tweaked some comments.
* Output a leading comment and surrounding blank lines.Wayne Davison2005-06-171-1/+6
|
* Mention rrsync changes.Wayne Davison2005-06-171-0/+5
|
* If SUBDIR is specified as a non-absolute path, make it absolute.Wayne Davison2005-06-171-1/+3
|
* - Made it easy to configure the path of the rsync executable.Wayne Davison2005-06-161-38/+160
| | | | | | | | | | | | | | | | | | | - Fully parse all legal options that might come from a stock rsync and validate/restrict any option's args (previously we rejected options with args). - Made it easy to disable specific options. - Properly handle a multi-arg copy's multiple args. - Make sure that the restricted dir actually exists and we can chdir to it. - Just reject any dangerous args that have .. in them instead of trying to munge them. - We don't treat single-or double-quotes specially, just backslash- escaped characters. - Turn on strict mode. - Use the array-arg version of exec() so that we don't spawn a shell. - Use File::Glob to glob wildcards in args. - Tweaked the log-file format so that it logs the modified command without any "OK" suffix.
* A developer-oriented script that outputs perl code for valid serverWayne Davison2005-06-161-0/+58
| | | | options.
* Mention latest fix.Wayne Davison2005-06-121-0/+5
|
* In clean_flist(), fixed the setting of the flist->high value for anWayne Davison2005-06-121-1/+5
| | | | | empty list so that it can't cause a crash if it gets passed to flist_find().
* - Added a comment to the filter_outfile function.Wayne Davison2005-06-101-4/+6
| | | | - Twiddled where the calls to filter_outfile happen.
* Simplified the regular expressions for sed due to problems on someWayne Davison2005-06-101-2/+5
| | | | OSes.
* - Put the output we're expecting into an rsync.chk file, for easierWayne Davison2005-06-101-15/+31
| | | | | | | | | | debugging on failures. - Added the function filter_outfile() and used it in place of the various output-filtering commands. - Use the sed rules for filtering the combination of --itemized output with double-verbose output so that we're sure there's no superfluous lines being output (which the old grep filtering would have ignored).
* Several expansions of $srcdir were not double-quoted.Wayne Davison2005-06-101-6/+6
|
* One expansion of $srcdir wasn't double-quoted.Wayne Davison2005-06-101-1/+1
|
* One more instance of the umask kluge was needed.Wayne Davison2005-06-101-0/+2
|
* Mention backup change.Wayne Davison2005-06-101-0/+4
|
* Fixed a problem in the backing up of symlinks and devices: weWayne Davison2005-06-101-20/+21
| | | | | | | | need to remove any old symlink/device/file to create the backup. Also fixed a couple minor logic errors in the handling of symlinks and devices. NOTE: the code still doesn't handle the changing of a name from a dir to a non-dir or visa versa (which is a very old deficiency in the code).
* Use umask kluge from rsync.fns to try to get rid of a potentialWayne Davison2005-06-101-0/+2
| | | | permissions difference when copying symlinks.
* Make the sed commands compatible with Solaris (hopefully).Wayne Davison2005-06-101-2/+2
|
* Starting NEWS for 2.6.6.Wayne Davison2005-06-101-170/+15
|
* Aged news from NEWS file.Wayne Davison2005-06-101-0/+185
|
* Got rid of some potentially unhandled permissions differences.Wayne Davison2005-06-101-2/+6
|
* Changed the "max verbosity" setting from global to per-module.Wayne Davison2005-06-102-6/+6
|
* Use the new NOBODY_USER and NOBODY_GROUP configure defines.Wayne Davison2005-06-101-11/+2
|
* Added a simple check to see if the /etc/group file has theWayne Davison2005-06-101-3/+12
| | | | group "nobody" or "nogroup".
* Test that our outputting of itemized messages (as well asWayne Davison2005-06-091-0/+149
| | | | double-verbose output) is happening correctly.
* - Fixed the overzealous outputting of unchanged hard-linked itemsWayne Davison2005-06-091-4/+7
| | | | | | when verbose > 1. - An unchanged hard-link with -i (log-format w/%i) now changes the trailing dots to spaces.
* Don't set ITEM_REPORT_TIME when a hard-linked file is alreadyWayne Davison2005-06-091-1/+2
| | | | linked up as it should be and --times was not enabled.
* - Never output the '"FOO" is a hard link' message when -i (%i)Wayne Davison2005-06-091-1/+2
| | | | | | output is enabled. - Go back to outputting '"FOO" is a hard link' message when -i (%i) is not enabled and verbose > 1.
* Updated version to 2.6.6cvs.Wayne Davison2005-06-091-1/+1
|
* Preparing for release of 2.6.5v2.6.5Wayne Davison2005-06-026-8/+8
|
* Improved a few of the items.Wayne Davison2005-06-021-25/+31
|
* One more configure item.Wayne Davison2005-06-011-0/+3
|
* Mention HAVE_REMSH fix.Wayne Davison2005-05-311-0/+2
|
* Simplified the last change a bit.Wayne Davison2005-05-271-3/+2
|
* Test for various past problems with --relative.Wayne Davison2005-05-271-0/+39
|
* If we unduplicate '.', make sure we also copy FLAG_DEL_HERE (inWayne Davison2005-05-271-2/+3
| | | | | addition to FLAG_TOP_DIR), so that the unduplication of a real dir and an implied dir (implied by --relative) works right.
* - Only define HAVE_REMSH if it is going to have a non-0 value.Wayne Davison2005-05-271-16/+3
| | | | - Got rid of ssh warning that was added for 2.6.0.
* Improved the opening comment.Wayne Davison2005-05-261-3/+3
|
* Mention the latest bug-fix.Wayne Davison2005-05-251-0/+5
|
* Handle nested "." dirs (caused by --relative and a trailing slash orWayne Davison2005-05-251-2/+12
| | | | | a trailing "/.") in f_name_cmp() when dirname is not an identical pointer.
* Mention the output change for a mount-dir when -x is used.Wayne Davison2005-05-221-1/+2
|
* - Don't stop deleting in a dir if one item fails.Wayne Davison2005-05-221-7/+3
| | | | | - Don't skip deleting in a dir if the initial rmdir fails (2.6.4 only continued if it was a non-empty error).
* Must check FLAG_MOUNT_POINT in delete_in_dir() to avoid starting aWayne Davison2005-05-221-1/+1
| | | | descent into a mounted directory when -x is specified.
* Improved the SYMBOLIC LINKS section, as suggested by John.Wayne Davison2005-05-201-1/+20
|
* Improved a --files-from example (as suggested by Paul).Wayne Davison2005-05-201-3/+8
|
View Lorry Controller Status