I've tested the case you've provided.
As I understand the problem is that the branch is not actually "a
branch", but more like its just a folder with the similar
structure. So, you're right, many of the issues will be solved
with the "ignore ancestry" option. On the other hand if it were to
be a real branch there would be a reason to not use "ignore
ancestry" option. So, I guess the option should be present as a
checkbox in the dialog.
Regarding the file content difference It is true there is no easy
way and the way you suggest seems to be the proper one.
P.S.
Thank you very much for your help, I'll work on comparison code
improvements in order to integrate your suggestions.
15.11.2016 19:31, Florent Angebault пишет:
Le 04/11/2016 à 09:10, Florent Angebault a écrit :
Le 02/11/2016 à 14:25, Alexander Gurov a écrit :
- SameContent.java is shown as an incoming deletion while it should not
shown at all because there is no difference (this is bug [326694]
but this time without common ancestry)
- DifferentContent.java is shown as an incoming deletion while it
should be shown as an incoming change or a conflict
I did many tests to find out how to remove false positive (files with no
differences). It seems that the available API do not provide a simple
way to discard files with same content. The only solution I could find
is to actually perform a real diff (with ISVNConnector#diffTwo()) and
parse the output to catch the names of files that are really different.
Here is the (dirty) patch that solves both problems mentionned above.
_______________________________________________
subversive-dev mailing list
subversive-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/subversive-dev
--
Best regards,
Alexander Gurov,
Subversive Team.