shape_patch(1) [xfree86 man page]

SHAPE_PATCH(1)						      General Commands Manual						    SHAPE_PATCH(1)

NAME

       shape_patch - shapeTools RMS generate patch file

SYNOPSIS

       shape patch OLDRELEASE=<name1> NEWRELEASE=<name2> [PATCHFILE=<filename>]

DESCRIPTION

       Shape  patch  generates	a patch file for updating $(OLDRELEASE) to $(NEWRELEASE). Both releases are identified by release names associated
       with the macros OLD-/NEWRELEASE on the command line. Valid release names are those generated by any of the shape_RMS (pre-)release building
       procedures  (see  shape_releas(1)).  Performing 'vl -all' with the release identification file as argument usually gives a good overview of
       existing release names.	Patch generation happens recursively over all subsystems being part of the current node. Hence,  triggering  shape
       patch from the top node of the central source repository creates a patch file for the whole system.

       The  output  of	shape patch is stored in a file named <name1>+2+<name2>.pat. When the PATCHFILE macro is set on the command line, $(PATCH-
       FILE) is taken as output file name instead.  Defining PATCHFILE=- on the command line causes the patch to be written  to  standard  output.
       Note:  For patches invoking multiple subsystems, $(PATCHFILE) should be set to an absolute pathname. If not, the patch generation procedure
       will leave an equally named patch file for each visited subsystem.

       The patch is constructed using the vdiff(1) command and can be applied to any installation of $(OLDRELEASE) by invoking patch(1).

INCONVENIENCES

       On System V machines, the generated patch file name will almost certainly exceed the 14 characters filename length limit.

SEE ALSO

       shape_releas(1), vdiff(1), patch(1)

FILES

       <name1>+2+<name2>.pat

2.10.119															    SHAPE_PATCH(1)

SVK::Command::Patch(3)					User Contributed Perl Documentation				    SVK::Command::Patch(3)

NAME

       SVK::Command::Patch - Manage patches

SYNOPSIS

	patch --ls    [--list]
	patch --cat   [--view]	     PATCHNAME
	patch --regen [--regenerate] PATCHNAME
	patch --up    [--update]     PATCHNAME
	patch --apply		     PATCHNAME [DEPOTPATH | PATH] [-- MERGEOPTIONS]
	patch --rm    [--delete]     PATCHNAME

OPTIONS

	--depot DEPOTNAME      : operate on a depot other than the default one

DESCRIPTION

       To create a patch, use "commit -P" or "smerge -P".  To import a patch that's sent to you by someone else, just drop it into the "patch"
       directory in your local svk repository. (That's usually "~/.svk/".)

       svk patches are compatible with GNU patch. Extra svk-specific metadata is stored in an encoded chunk at the end of the file.

       A patch name of "-" refers to the standard input and output.

INTRODUCTION

       "svk patch" command can help out on the situation where you want to maintain your patchset to a given project.  It is used under the
       situation that you have no direct write access to remote repository, thus "svk push" cannot be used.

       Suppose you mirror project "foo" to "//mirror/foo", create a local copy on "//local/foo", and check out to "~/dev/foo". After you've done
       some work, you type:

	   svk commit -m "Add my new feature"

       to commit changes from "~/dev/foo" to "//local/foo". If you have commit access to the upstream repository, you can submit your changes
       directly like this:

	   svk push //local/foo

       Sometimes, it's useful to send a patch, rather than submit changes directly, either because you don't have permission to commit to the
       upstream repository or because you don't think your changes are ready to be committed.

       To create a patch containing the differences between "//local/foo" and "//mirror/foo", use this command:

	   svk push -P Foo //local/foo

       The "-P" flag tells svk that you want to create a patch rather than push the changes to the upstream repository.  "-P" takes a single flag:
       a patch name.  It probably makes sense to name it after the feature implemented or bug fixed by the patch. Patch files you generate will be
       created in the "patch" subdirectory of your local svk repository.

       Over time, other developers will make changes to project "foo". From time to time, you may need to update your patch so that it still
       applies cleanly.

       First, make sure your local branch is up to date with any changes made upstream:

	   svk pull //local/foo

       Next, update your patch so that it will apply cleanly to the newest version of the upstream repository:

	   svk patch --update Foo

       Finally, regenerate your patch to include other changes you've made on your local branch since you created or last regenerated the patch:

	   svk patch --regen Foo

       To get a list of all patches your svk knows about, run:

	   svk patch --list

       To see the current version of a specific patch, run:

	   svk patch --view Foo

       When you're done with a patch and don't want it hanging around anymore, run:
	   svk patch --delete Foo

       To apply a patch to the repository that someone else has sent you, run:

	   svk patch --apply - < contributed_feature.patch

perl v5.10.0							    2008-08-04						    SVK::Command::Patch(3)