On Tue, 22 Jan 2019 at 14:34, Martin Ågren <martin.agren@xxxxxxxxx> wrote: > > On Tue, 22 Jan 2019 at 08:07, Jeff King <peff@xxxxxxxx> wrote: > > > For the record, I can live with it either way. There are so many funky > > little setup corner cases in the code already, and we don't even really > > have a real-world case to dissect at this point. So the right thing may > > also just be to finish this patch series as quickly as possible and move > > on to something more useful. :) > > I rebased the "something like this?" into this series yesterday and I > think the end result is better, but also that the way there is clearer, > mostly because this patch is then gone. I wanted to double-check it > tonight and submit it. I'll do that tonight. Here's that reroll. I now reset the entire struct in the error path of `clear_...()`. Thus, the user that is reading `repo_fmt.hash_algo` despite not being supposed to, can keep reading it, now knowing that the value has a default value. I also expanded on the documentation a little to point out that we'll reset to the default struct state if we don't find any "core.repositoryformatversion". Martin Martin Ågren (2): setup: free old value before setting `work_tree` setup: fix memory leaks with `struct repository_format` cache.h | 27 ++++++++++++++++++++++++--- builtin/init-db.c | 3 ++- repository.c | 3 ++- setup.c | 33 +++++++++++++++++++++------------ worktree.c | 4 +++- 5 files changed, 52 insertions(+), 18 deletions(-) -- 2.20.1.98.gecbdaf0899
