Some background on the large change:
This is cascading from the impact of line endings which triggered the use of common preferences i.e. to add UNIX line endings for new files, we took the opportunity to apply the Mylyn project preferences as a base.
Using the project preferences introduces many formatting changes each save, making it very difficult to review the actual functional changes as they are embedded among formatting changes.
So the large change applies the formatting changes in one shot so we can get back to see functional changes per commit..
This change is unusually large and very much needed as it's a big time saver, reverting the change would have side effects which translates to a big effort for a change that has to go in any how.
The good thing is that the large commit is related to formatting of new lines, spaces and tabs, and this formatting can be re-applied after the merge.
Git offers merge / re-base strategies that can be exercised to see which one gives a better result to make your updates have precedence over the formatting changes and once all is compiled and tested, the formatting changes can be re-applied before completing the re-base / merge.
We'll have to rebase / merge only the large commit first to avoid impact to commits ahead of the large commit.
e.g. git merge -s recursive -Xours bigChangeBranch
I think it's best if you file a bug on this so we can discuss the details and results in there
/Alvaro
On Thu, Oct 11, 2012 at 3:03 PM, Miles Parker
<miles.parker@xxxxxxxxxxx> wrote:
Hi guys,
We're going to have a big problem with changes like:
f20df9ab7dfcd645fb6a4122e34e593afe86e979 391541 Apply formatting preferences..
When I tried to rebase those against the major changes to the generated model in bug324327CommonModel, naturally they all had merge issues.
So let's cherry pick that commit out, ok? We can re-apply these once we've merged the common model changes in. In general for my own sanity, please avoid making any changes to the generated model classes or the classes consuming them that aren't essential.
cheers,
Miles
_______________________________________________
mylyn-reviews-dev mailing list
mylyn-reviews-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mylyn-reviews-dev
