Re: hg-git: push -d default seems not to work.

> On Fri, 05 May 2017 16:26:45 -0400, Uwe Brauer <o...@mat.ucm.es> wrote: > I'm not clear on some of the details of what you are doing. The > original question was how to push a merge without one of its ancestor > branches. That implies you effectively want to f

Re: hg-git: push -d default seems not to work.

> On Fri, May 5, 2017 at 5:10 AM, Uwe Brauer <o...@mat.ucm.es> wrote: > style. > You CANNOT push a changeset without its ancestors. It's impossible > no matter which extension you are using. A commit's ID depends on > the ID(s) of its parent(s)

Re: hg-git: push -d default seems not to work.

> be pushed. What I am most interested is in the graph. That is I would like to have a graph like this (after a merge from a named branch into (default/master) @changeset: 7058:b017880d8552 |\ tag: tip | | parent: 7057:41a79851cd40 | | parent: 7056:a740234025

Re: hg-git: push -d default seems not to work.

> Why do you merge it? Generally when you're submitting code upstream, > you'd submit just your branch. The maintainer of the project will > apply your patches or merge your branch. Because they want clean 'clutter free' commits and they don't want branches to be pushed. I could use

hg-git: push -d default seems not to work.

| | parent: 6934:b61106aa8e7b | | user:Uwe Brauer <o...@mat.ucm.es> | | date:Wed May 03 07:19:00 2017 + | | summary: Actualize question macro. | | | o changeset: 6934:b61106aa8e7b |/ branch: exam |user: Uwe Brauer <o...@mat.ucm.es&g

git filter-branch equivalent

Hi I hope this my last question concerning hg equivalents to git features, but is there something comparable to git filter-branches? Evolve, maybe? Thanks Uwe Brauer ___ Mercurial mailing list Mercurial@mercurial-scm.org https://www.mercurial

Re: git reflog hg equivalent?

> On Wed, Apr 19, 2017 at 4:40 PM, Uwe Brauer <o...@mat.ucm.es> wrote: > I believe the journal extension is storing extra data, and if it > wasn't enabled then it won't know about the history of how the > repository changed over time. Ok that makes sense, thank

Re: push content of a symbolic link to remote server.

> Uwe Brauer <o...@mat.ucm.es> writes: > It would be (with name->path as symbolic link) > shell/repo1/lib->../repo3/lib > repo2/lib->../repo3/lib > repo3/lib Now I am more confused. Shell is supposed to be a repo as well? And now rep

push content of a symbolic link to remote server.

ick to do that? Or any other possibility to share one file in various repos? Thanks Uwe Brauer ___ Mercurial mailing list Mercurial@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial

SOLVED (was: bitbucket subrepos and set passwords)

>>> "Uwe" == Uwe Brauer <o...@mat.ucm.es> writes: > Hi > Maybe I somebody knows the answer on this list. This can be solved by putting the information in the global .hgrc file ___ Mercurial mailing list Me

bitbucket subrepos and set passwords

Hi Maybe I somebody knows the answer on this list. On bitbucket I set in my hgrc file something like this [paths] default = https://kalt...@bitbucket.org/myuser/myrepo [ui] username=Uwe Brauer <o...@mat.ucm.es> [auth] repo.prefix = https://bitbucket.org/myuser/myrepo repo.username =

subrepos and the share extension

, but its history is included in the subdir1 (repo)? Thanks Uwe Brauer ___ Mercurial mailing list Mercurial@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial

bitbucket, introduce in a directory a subrepo

Hi I have a repo which sits in bitbucket. Now I decided that I would like to push one directory as repo itself. Like repomain/dir1 repomain/dir2 dir1 should be in a a repo of its own in bitbucket. I thought I could do this with subrepos, but I am not so sure whether this can be done for

SOLVED (was: patches, record almost)

>>> "Uwe" == Uwe Brauer <o...@mat.ucm.es> writes: >>> "Uwe" == Uwe Brauer <o...@mat.ucm.es> writes: > So the following does almost what I want > hg init > echo Main1 > main.txt > hg add main.txt > hg comm

hg record and merge does not work (was: anti-dummy merge cherry pick from one commit of different branch)

>>> "Uwe" == Uwe Brauer <o...@mat.ucm.es> writes: > Hi > The question is this: Could I force to pop up a merging tool (again say > meld) which just would show me all the changes of the collaborator's > branch, so that I could cherry pick wha

Re: Is hg-git still alive

>>> "David" == David Demelier writes: > Le 09/02/2017 à 04:16, Jesus Cea a écrit : >> No traffic in the mailing list and current hg-git doesn't work with >> recent versions of mercurial. > Currently using Mercurial 4.1 with hg-git here, no problem so far.

Re: merged two branches, and pushed, how to backout/strip?

> Uwe Brauer <o...@mat.ucm.es> writes: > This is because the merge does not exist yet on the vs-11.89 > branch. However when you now do some work on default and merge it into > vs-11.89, you will introduce the backout of vs-11.89 into vs-11.89, > und

Re: merged two branches, and pushed, how to backout/strip?

> Uwe Brauer writes: > What’s your current branch? default > What happens if you merge this into the other branch? You mean Hg merge vs-11.89 abort: merging with a working directory ancestor has no effect Or do you mean Hg up vs-11.89 Hg merge default? Result: @

Re: merged two branches, and pushed, how to backout/strip?

> Assuming you didn't leave out steps between your two reverts you are > now worse off because that last commit isn't backing out a bad merge > from vs-11.89, it's replacing everything on default with your last > revision from vs-11.89. Oops that is not good. But that changeset (4)

Re: branches, bookmarks, new heads

> On 12/30/2016 07:03 PM, Uwe Brauer wrote: > The best history is the one that does not have a head. I strongly disagree, I find the following graph changeset: 3:7a4a5f52960e | tag: tip | parent: 0:02c2a80b8472 | user: Uwe Brauer <o...@mat.ucm.es

Re: merged two branches, and pushed, how to backout/strip?

>>> "Arne" == Arne Babenhauserheide writes: >> >> That is correct? However I end up in the vs-11.89 branch. >> >> I don't really understand >> hg up vs-11.89 >> hg merge default >> hg ci -m "merge default with the backout" >> hg log -G >> # revert

Re: merged two branches, and pushed, how to backout/strip?

> Uwe Brauer <o...@mat.ucm.es> writes: > If the code being in the open is no problem, you can always revert > default to the commit before the merge, then commit, then merge default > into vs-11.89, revert vs-11.89 to the other commit before the merge and > c

merged two branches, and pushed, how to backout/strip?

: 138:7e1a64287b87 | | parent: 142:1dd731cf6a6a | | user:Uwe Brauer <o...@mat.ucm.es> | | date:Sun Jan 08 19:42:20 2017 + | | summary: Merged | | | o changeset: 142:1dd731cf6a6a | | branch: vs-11.89 | | user: Uwe Brauer <o...@mat.ucm.es&g

branches, bookmarks, new heads

; new.el hg commit -m feature Does not. When using a bookmark could a head optionally be created? Thanks Uwe Brauer ___ Mercurial mailing list Mercurial@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial

Re: clone, apply patch then pull, pulled «identical» changeset?

>>> "Benjamin" == Benjamin Fritz writes: > The changesets are not identical, they have different commit messages. Thus > the hash over the entire changeset differs, making the changeset ID > different. Oops you are right, my bad. Sorry

clone, apply patch then pull, pulled «identical» changeset?

ch.patch The patch gets applied on clone them in server clone is pulled (also all files are identical) and indeed the last changset is pulled resulting in the following graphs. Clone o changeset: 3:01f5c2db879d | tag: tip | user:Uwe Brauer <o...@mat.ucm.es> | date:

Re: [fold: different for branches and bookmarks]

> On 12/21/2016 08:40 PM, Uwe Brauer wrote: > It looks like your fold is empty and that fold unfriendly crash in > that case. Can you submit a bug? (and test latest evolve first if you > are not using it yet) I just updated to the latest evolve version and it still cras

Re: [fold: different for branches and bookmarks]

> On 12/19/2016 10:25 PM, Arne Babenhauserheide wrote: > The call `hg fold -r master -r feature` seems wrong. It will fold > '(master+feature)::. + .::(master+feature)' Most certainly not what > you want as master will be included in the resulting folded > changesets. > You

[fold: different for branches and bookmarks] (was: [git repos])

> Uwe Brauer writes: > This sounds like a typical case for rebasing — or more elegantly using > the evolve extension. > Upstream requires a single commit (one patch). So you’d ideally have > just that. > With the evolve extension you’d simply issue

[git repos] (was: elmentary (watson): merge (named) branches and/or bookmarks)

> Uwe Brauer writes: > Same for me. And I wish I could tell hg-git to use named branches — but > I know that this doesn’t work. > The trouble you run into is what git calls fast-forward merge, by > the way. It’s not an actual merge, just a pointer movement. A

Re: elmentary (watson): merge (named) branches and/or bookmarks

>>> "Arne" == Arne Babenhauserheide <arne_...@web.de> writes: > Uwe Brauer writes: >> Sigh, for me this is another reason to concentrate on named branches, >> since they behave much more as I expect them to behave. (Besides the >> e

elmentary (watson): merge (named) branches and/or bookmarks

m.el hg add exam.el hg commit -m exam echo exam2 >> exam.el hg commit -m exam2 hg update master hg merge exam But I obtain abort: nothing to merge (use 'hg update' or check 'hg heads') Sigh what is the problem here? Uwe Brauer ___ Mercuria

Re: bitbucket, using the «wrong» branch

> On 16 Dec 2016, at 21:26, Arne Babenhauserheide wrote: > > Is it possible that he updated with `hg update tip` > instead of `hg update`? Yes you are absolutely right. My bad. Thanks. smime.p7s Description: S/MIME cryptographic signature

Re: parent -1

> On 11/29/2016 11:29 PM, Uwe Brauer wrote: > The warning here means that for some reason some Mercurial internal > state got "corrupt" and was recording a current "checked out" revision > now unknown to Mercurial. This is most likely the reason for t

Re: a changeset belongs to two branches?

>>> "Uwe" == Uwe Brauer <o...@mat.ucm.es> writes: > Hi > In my repo I have to branches, > - the default > - the branch Uwe. > However I just realized that > hg log -G > Gave me > o change

a changeset belongs to two branches?

Hi In my repo I have to branches, - the default - the branch Uwe. However I just realized that hg log -G Gave me o changeset: 68:1ec5162467a7 | | branch: Uwe | | branch: default/Uwe | | parent: 66:ce46517fd6e4 | | user:Uwe Brauer <o...@mat.ucm

Re: hard and soft links

> On 11/24/2016 04:56 AM, Uwe Brauer wrote: > I'm unclear on what you want to have happen. dir1 and dir2 are > checkouts of the same repo? If you want their contents to be > identical, why not symlink dir2 to dir1? If you want the contents to > be different

parent -1

John Foo <j...@foo.com> date:Tue Nov 29 21:35:40 2016 +0200 summary: Msg I told him to move out the modified files, to strip, to pull again and then try to commit again. Anybody has an explanation? What does parent: -1:00000000 M

Re: feature proposal: named branches without named branches

> Uwe Brauer writes: > Yes. > You’d have to change the phase on the server to draft, then it would > work. That’s possible in many settings. Aha, it seems that bitbucket has that feature: activating Phases This is a non-publishing repository Seems to do th

Re: feature proposal: named branches without named branches

> Arne Babenhauserheide writes: > A more canonical version uses > hg ci --amend -m stable The first VC I used for years was RCS and I always hated ci and co, since I never was 100 % sure what was what, so for mercurial I use always hg commit -m "Some message" And hg update

Re: bit off topic: bitbucket fingerprint problem

So that looks as if the fingerprint were missing. Any ideas? Thanks Uwe Brauer ___ Mercurial mailing list Mercurial@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial

Re: two branches, cherrypicking when merging.

>>> "Steve" == Steve Fink writes: > It sounds like this does not match your preferred workflow, but > just as a comparison point, I would use a bookmark for Uwe instead > of a named branch, then: > - record -i -m WIP # select the work in progress What is the -i

Re: two branches, cherrypicking when merging.

>>> "Steve" == Steve Fink writes: > It sounds like this does not match your preferred workflow, but just as a > comparison point, I would use a bookmark for Uwe instead of a named branch, > then: well I know a lot of people prefer bookmarks over named brances, since

Re: two branches, cherrypicking when merging.

> Assuming that the experimental changes don't immediately need the fix, > I tend to do it the other way around. i.e. Shelve the fix and commit > the experimental changes first. Then update to default before > unshelving and committing the fix. Oh wait, can I unshelve *after* I

Re: two branches, cherrypicking when merging.

> On Wed, Nov 2, 2016 at 11:31 AM Uwe Brauer <o...@mat.ucm.es> wrote: > If it's a bug that exists in default, I'd fix it in default, commit, then > merge to Uwe Branch. Yes I agree that would be ideal, however it occurred now several times that I was a bit ca

Re: two branches, cherrypicking when merging.

>>> "Uwe" == Uwe Brauer <o...@mat.ucm.es> writes: > Hi > The only solution seems to shelve, to add again the fix lines, to commit > to merge, to go back and to unshelve. The following seems to work. - hg record (cherry pick) - hg shelve

two branches, cherrypicking when merging.

the fix lines, to commit to merge, to go back and to unshelve. But that looks cumbersome. Does somebody has an idea? Thanks Uwe Brauer ___ Mercurial mailing list Mercurial@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial

Re: configure external editor a script

> On Wed, Oct 19, 2016 at 2:49 PM, Uwe Brauer <o...@mat.ucm.es> wrote: > emacs and emacsclient probably use a similar set of environment > variables for their control socket location. Try the other 2 > environment variables (TMP and TEMP), and if none

Re: extract one single file and its history from a repo

> On Thu, Sep 29, 2016 at 8:30 PM, Uwe Brauer <o...@mat.ucm.es> wrote: > It looks like export only exports whole changesets. You probably want > to use "hg convert" with a filemap that just specifies your single > file. > https://www.mercuria

convert (was: bitbucket and the evil largefile extension)

>>> "Uwe" == Uwe Brauer <o...@mat.ucm.es> writes: > Hi > I have a repo which uses the largefile extension. I wanted to push it to > bitbucket but it was refused with the message > repo > does not appear to be a largefile store I tried

git-hg plugin, problems with the patch.

: 6851:7a326e5bfa12 | bookmark:pretty | tag: tip | parent: 6847:c54ae6c3004a | user:Uwe Brauer <o...@mat.ucm.es> | date:Thu Sep 15 15:07:10 2016 + | summary: auctex.texi: Modify text for WYSIWYG paragraph, shorten reference to x-symbol. C

Re: problems with the hg-git plugin and bookmarks

> On 09/07/2016 04:41 AM, Uwe Brauer wrote: Thanks for that long and detailed answer, that helped me to understand things. > Your complexity seems to arise from this desire to have your original > commits. I think things would be easier if you didn't try to do that.

rebase: linearize and the inverse operation

e are now in master | echo uwe-four >> main.txt | hg commit -m "Uwe four" | hg log -G ` compare the graphs. Does my inverse operation make sense? Thanks Uwe Brauer ___ Mercurial mailing list Mercurial@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial

Re: change, transform convert branches to bookmarks

> On Fri, Sep 2, 2016 at 12:24 PM, Uwe Brauer <o...@mat.ucm.es> wrote: > One option is to use "hg convert" with the "--branchmap" option to > produce a new repository without any named branches. If you've only > got a small number of branch

Re: hg-git plugin, problem with local cloning: bookmarks

> On Thu, Sep 1, 2016 at 9:06 AM, Uwe Brauer <o...@mat.ucm.es> wrote: > I'm not much of an hg-git user, so I may be wrong here, but I would > guess that you need to get the git repository that is being managed by > hg-git (which I think is stored at

Re: hg-git plugin, problem with local cloning: bookmarks

> Git only clones the master branch (or whatever the remote says its > HEAD is), so your local copy actually doesn't have the other Git > branches in it. You can get them by running `git fetch` in the Git > clone, then re-clone locally with hg. > Also, hg-git has its own mailing

Re: hg-git plugin, problem with local cloning: bookmarks

> Git only clones the master branch (or whatever the remote says its > HEAD is), so your local copy actually doesn't have the other Git > branches in it. You can get them by running `git fetch` in the Git > clone, then re-clone locally with hg. Thanks but git fetch Didn't do

hg-git plugin, problem with local cloning: bookmarks

ing because I already cloned via git GNU emacs and don't want to run hg clone git://git.savannah.gnu.org/emacs.git because it would need the whole day. Regards Uwe Brauer Footnotes: [1] a nifty addon for writing Latex files using GNU em

Re: dummy merge (two branches) for dummies

> If you want to keep more of his files than yours then you would use > internal:other and revert your few files back to uwe. If you want to > keep more of your files than his, use internal:local and revert his > files back to foo. For details about the various merge tools, see hg

several rev, one patch exclude branch

Hello Please have a lock at the following log -G @ changeset: 6831:51db987e1c4b | bookmark:master | tag: tip | parent: 6828:812891d581e1 | user:Uwe Brauer <o...@mat.ucm.es> | date:Sat Aug 13 19:25:53 2016 + | summary: * exam.el (LaTeX-exam

<    3   4   5   6   7   8