Comments on: Mercurial Backouts http://blog.vlad1.com/2009/01/25/mercurial-backouts/ Words Mon, 15 Mar 2010 23:28:54 +0000 hourly 1 http://wordpress.org/?v=3.0 By: Neil Rashbrook http://blog.vlad1.com/2009/01/25/mercurial-backouts/comment-page-1/#comment-3137 Neil Rashbrook Tue, 27 Jan 2009 10:07:23 +0000 http://blog.vlad1.com/?p=104#comment-3137 hg doesn't provide bidirectional binary git diffs, so git-apply --reverse won't help you there, as I found out the hard way. You could always try combining backout, export and rollback to generate a reverse diff. hg doesn’t provide bidirectional binary git diffs, so git-apply –reverse won’t help you there, as I found out the hard way. You could always try combining backout, export and rollback to generate a reverse diff.

]]> By: Wladimir Palant http://blog.vlad1.com/2009/01/25/mercurial-backouts/comment-page-1/#comment-3136 Wladimir Palant Mon, 26 Jan 2009 08:33:15 +0000 http://blog.vlad1.com/?p=104#comment-3136 I guess writing a script that will take a patch and reverse all the operations in it (e.g. replace by -) shouldn't be too hard - and the resulting patch can be applied via "hg import" then. Using "hg export|patch -R" is really too risky for my taste. I guess writing a script that will take a patch and reverse all the operations in it (e.g. replace by -) shouldn’t be too hard – and the resulting patch can be applied via “hg import” then. Using “hg export|patch -R” is really too risky for my taste.

]]> By: Robert O'Callahan http://blog.vlad1.com/2009/01/25/mercurial-backouts/comment-page-1/#comment-3135 Robert O'Callahan Mon, 26 Jan 2009 05:16:41 +0000 http://blog.vlad1.com/?p=104#comment-3135 -R as a parameter for hg import would make this a piece of cake. I started off doing hg backouts using hg export|patch -Rp1, but after the first time I screwed up a backout that involved adding/removing files, I decided it wasn't worth the risk. -R as a parameter for hg import would make this a piece of cake.

I started off doing hg backouts using hg export|patch -Rp1, but after the first time I screwed up a backout that involved adding/removing files, I decided it wasn’t worth the risk.

]]> By: Justin Dolske http://blog.vlad1.com/2009/01/25/mercurial-backouts/comment-page-1/#comment-3134 Justin Dolske Mon, 26 Jan 2009 03:37:57 +0000 http://blog.vlad1.com/?p=104#comment-3134 Yeah, 1 to getting this built into Mercurial. I hate the merge turds, though it is nice to be able to use the same command in any situation. Yeah, 1 to getting this built into Mercurial. I hate the merge turds, though it is nice to be able to use the same command in any situation.

]]> By: Screwtape http://blog.vlad1.com/2009/01/25/mercurial-backouts/comment-page-1/#comment-3133 Screwtape Mon, 26 Jan 2009 01:15:20 +0000 http://blog.vlad1.com/?p=104#comment-3133 Well, obviously git-apply can apply git-format patches in reverse. Actually, git-apply is better than patch in many ways; not least is that it won't modify any files until it's verified that the patch will apply cleanly. If Mercurial is using the Git patch-interchange format, I'd be very surprised if it doesn't come with the basic tools for working with that format. Well, obviously git-apply can apply git-format patches in reverse. Actually, git-apply is better than patch in many ways; not least is that it won’t modify any files until it’s verified that the patch will apply cleanly. If Mercurial is using the Git patch-interchange format, I’d be very surprised if it doesn’t come with the basic tools for working with that format.

]]> By: bholley http://blog.vlad1.com/2009/01/25/mercurial-backouts/comment-page-1/#comment-3132 bholley Mon, 26 Jan 2009 01:01:37 +0000 http://blog.vlad1.com/?p=104#comment-3132 I support this. Doesn't moco pay various people to work on hg? It seems like me like this would be trivial to implement as a command-line flag for hg backout (fixing the complex-patch issue along the way), and it would save people from remembering what to type. I support this.

Doesn’t moco pay various people to work on hg? It seems like me like this would be trivial to implement as a command-line flag for hg backout (fixing the complex-patch issue along the way), and it would save people from remembering what to type.

]]> By: Seth Heeren http://blog.vlad1.com/2009/01/25/mercurial-backouts/comment-page-1/#comment-3130 Seth Heeren Mon, 26 Jan 2009 00:21:01 +0000 http://blog.vlad1.com/?p=104#comment-3130 long live 'bzr uncommit' (try 'bzr rocks' once) long live ‘bzr uncommit’

(try ‘bzr rocks’ once)

]]>