On Fri, Sep 16, 2016 at 2:51 PM, Uwe Brauer <o...@mat.ucm.es> wrote:
>    > On Fri, Sep 16, 2016 at 1:55 PM, Uwe Brauer <o...@mat.ucm.es> wrote:
>    > Sorry, you're right. "hg import" does indeed handle multiple patches
>    > in the same file.
>    > The auctex repo isn't loading for me at the moment so I can't check,
>    > but has anyone modified auctex.texi since you made your changes? That
>    > would be one reason for the patch not applying.
> Not that I know of.
>    > Also, I noticed that the attachments on your previous email had CRLF
>    > line endings. Have you somehow ended up in a situation where your
>    > patch has one style of line endings but the files you are trying to
>    > apply them to have a different style?
> I will try to check this. The current problem is not urgent, I exported
> my recent changes cloned and merged that in, so that solved the problem
> somehow, I just wanted to know for the future whether I made something
> fundamental wrong?
> For me the most important questions are these.
>     -  it is not necessary to use rebase.

It's not necessary to use rebase, but it can be easier. When your
changes don't apply cleanly (because someone else has modified the
same files), rebase will launch the 3-way merge mechanism because it
knows what the common ancestor of the changes were. export/import
loses that information, so resolving conflicts can be more difficult.

For what it's worth (and ignoring any extra complications that might
be introduced by hg-git), the rebase workflow is normally as simple as
using "hg pull --rebase" instead of "hg pull".

>     -  In my workflow I alwayscreate a bookmark for my changes. When I
>       send a patch, I always have to send the whole branch (bookmark)
> That is hg export -r 1::3 -o mypatch.patch
> In my previous example. Is that  correct?
> BTW even if I used named branches instead of bookmarks there is no
> simple way to export a whole branch? As in  in my example
>  hg export --branch uwe -o mypatch.patch
> Does not exist?

This is where it's really worth learning the revset language again
(https://selenic.com/hg/help/revsets). In this case, you want to
export revisions that are ancestors of your current working directory,
and aren't ancestors of "master".

ie. "ancestors(.) - ancestors(master)"

You can use the "::" operator as shorthand for the ancestors() function:

::. - ::master

You can also use the only() function:

only(::., ::master)

(You *might* be able to leave out the "::master" in that one, I'm not certain)

If you've used a named branch, the branch() function would return all
the revisions on that branch.

You should test all of these with "hg log" to see what they return.

Mercurial mailing list

Reply via email to