Should I crush a small, development is moving into larger ones?

There are many questions about how to execute squash commits for git and other DVCS, for example:

My question is: I want to compromise ? Do I have to contain a detailed sequence of commits showing how the function was created, or should I rather put them into one as soon as the function is finished in order to save the history cleaner?

+5
source
3

:

  • .
  • . .
  • "", , .
  • .

:

  • /, , , , , . ( A Git .)
  • Git bisect . , , 300 , , Git, , . ( A Git .)

, - , .

+7

, , . , , , , ?

diff, , . ( )? , - , - , ?

, ? - , , . , , .

. , , , , , .

+1

DVCS .

, , , , " - ", Mercurial. , , , , . M, , , , . , , Mercurial, , .

I am also intrigued by the potential Mercurial (currently) concept of marker obsolescence with matte bleeding . Saving the original changes, but allowing them to be replaced by crushed ones, it would seem that you can get your cake and eat it too: a clean story with the latest, reasonable sizes shown by default, but always with the option to unpack the function and carefully study the small efforts that entered into it.

+1
source

All Articles