On 2010-05-12 16:46, Philip Martin wrote: > I saw you talking about keep-local on IRC a couple of days ago, are > you working on keep-local in harvest_committables?
Yes, indeed I am ... was ... will. I've been kept busy by real life issues, but that's about to pass. Thanks for reminding me, btw. The bottom line is: simply remove that madly misguided entry->keep_local conditional and always call bail_on_tree_conflicted_children(). The rest of the patch is mending the test suite to be sensible about it. Will do. For the record: We (mostly Julian and me) discussed that tree-conflicts should not be ignored at commit time, no matter how they came about. ~Neels
signature.asc
Description: OpenPGP digital signature
