Le 28/04/2017 à 11:36, Patrick Ohly a écrit :
On Fri, 2017-04-28 at 10:18 +0100, Graham Cobb wrote:
On 28/04/17 10:05, Patrick Ohly wrote:
On Fri, 2017-04-28 at 10:31 +0200, Vincent wrote:
Could you tell me more about synccompare?
In the syncevolution.org packages, it is under /usr/bin/synccompare.
It's a perl script that takes two database dumps and compares them,
similar to a diff between text files.
I find it to be of variable usefulness. It is great when it works, but
in my experience it scales horribly.
I just use it on a laptop and it works for me, but I agree that it's
mostly a hack originating in the automated testing. There's even a bug
open for rewriting it...
I think I always see "comparison was impossible" with refreshes. I
assumed that is because the databases are deleted or something (although
thinking about it further I am not sure that is a reasonable expectation).
There should be "before" and "after" dumps also for refreshes, so this
has to be something else.
So what is the goal of this command? Does it mean that something
went wrong but nothing important enought to break the system?