>>> "UB" == Uwe Brauer <o...@mat.ucm.es> writes:

   > Hi

   > I very use backout rarely. (Either I strip or solve the problem in a
   > different way).

   > I realised that I don't really understand its logic.

   > Take the following script 
   > hg init
   > echo "First" > test.org
   > hg add test.org
   > hg commit  -m "0: First"
   > echo "Second" >> test.org
   > hg commit  -m "1: Second"
   > echo "Third" >> test.org
   > hg commit  -m "2: Third"
   > echo "Fourth" >> test.org
   > hg commit  -m "3: BUG Fourth"
   > echo "Fifth" >> test.org
   > hg commit  -m "4: Fifth"
   > echo "Six" >> test.org
   > hg commit  -m "5: Six"

   > The file test.org contains six lines, each line correspond to one commit

   > First
   > Second
   > Third
   > Fourth
   > Fifth
   > Six

   > I want to backout the 4th commit with the log message 
   > 3: BUG Fourth".

   > If I stripped that changeset all children would also be stripped, but I
   > thought a backout would only strip the Fourth line resulting in 

   > First
   > Second
   > Third
   > Fifth
   > Six


   > However when I run 
   > hg backout -r 3

   > A merging tool pops up asking me either to use 


   > First
   > Second
   > Third
   > Fourth
   > Fifth
   > Six

   > Or 

   > First
   > Second
   > Third
   > Fifth
   > Six

   > Or 

   > First
   > Second
   > Third


   > Which I find very odd. What do I miss?

   > Uwe Brauer 

Ok I think I know what's up:

If I run 
 hg up 3
 hg backout 3

I create a new head, which is the backout changeset. This head I shall
merge with the former tip. Ok makes sense the behavior. But is not that
intuitive.

Why I have to use hg up 3 is not obvious.

Uwe Brauer 

Attachment: smime.p7s
Description: S/MIME cryptographic signature

_______________________________________________
Mercurial mailing list
Mercurial@mercurial-scm.org
https://www.mercurial-scm.org/mailman/listinfo/mercurial

Reply via email to