Here's an example (this is where it works):
A -> B -> C -> D
\
\
-> CA -> CB
All the commits are either all jalopied or all not-jalopied. CA/CB are
a branch.
In this case, I can compare A & CB and see the differences. I can also
compare C & CB and see whats changed on the branch. I can also select
D & CB and see whats the diffence between the branch and trunk. I'm
happy .
(This is pretty much whats happening now in the archive)
Here's another example:
a -> b -> c -> D -> e -> f -> g
\
\
-> ca -> cb -> cC -> cd
Where "D" is a jalopy commit and the 'c->ca' is a branch (with "cC"
being a jalopy commit). The other commits arent jalopied.
I cannot select a&g and do a diff because I'll get 100 (maybe even
1000s) of changes between the files (because of D's jalopy). To do
this, you have to look at the changes between a & c, then manually
(THIS CAN TAKE A LONG TIME) go through all the changes in D, then diff
D and g.
You cannot diff c&cd because there's a jalopy (cC).
You, can, however, diff g & cd but you'll get more changes than you
expect because only parts of the commits have been jalopied.
I've had to spend a lot of time going through autoformatting changes.
One single file cost me more time that the sum total of all time that
autoformatting will save me in reading time (if any). Thats just one
file!
Autoformating existing files puts a black pit in your history. It can
easily take hours to sort through a single file.
I'm -1 except for new files.
dave
----------------------------------------------------------
This mail sent through IMP: https://webmail.limegroup.com/