[PATCHv11 02/11] receive-pack.c: die instead of error in case of possible future bug

Discussion on the previous patch revealed we rather want to err on the safe side. To do so we need to stop receive-pack in case of the possible future bug when connectivity is not checked on a shallow push.

Also while touching that code we considered that removing the reported refs may be harmful in some situations. Sound the message more like a "This Cannot Happen, Please Investigate!" instead of giving advice to remove refs. Running 'git fsck' should not cause any problems though.

Suggested-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Stefan Beller <sbeller@google.com>
---

Notes:
    v11:
    * only die at the end so the loop works out for all refs.
    * Remove the advice to delete refs.
    
    > If the message says
    >        fatal: BUG: connectivity check skipped???
    > then it has exactly the right amount of information to tell me what to
    > do.  Now I have
    > - a short string to grep for in the source code (or on the web) to
    >    find out what happened
    
    And do a git blame to see previous versions?
    
    I am not so sure of this patch any more as it actually stops people
    doing work if they want to do so. (They may deliberately choose to
    ignore the BUG:... message, because of a deadline in 2 hours.)
    
    So I do think this helps on getting people to report the bug in the
    future if it arises faster, but on the other hand if we assume the
    faulty hardware scenario and the deadline we actually stop people
    from getting their desired work done.
    
    This patch doesn't actually relate to the topic of the series
    (atomic pushes), but is a cleanup-as-we-go patch. If we need to
    have further discussion on this, I'd rather want to delay this patch
    and have a follow up on top of the atomic series.
    
    Thanks,
    Stefan
    
    v10:
    * new in v10.

 builtin/receive-pack.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)