Re: [PATCH] Add failing test: "fsck survives inflate errors"

2014-08-08 Thread Duy Nguyen
On Mon, Jul 21, 2014 at 3:43 AM, Samuel Bronson wrote: > So, given that parse_object()'s documentation is: > > --8<---cut here---start->8--- > /* > * Returns the object, having parsed it to find out what it is. > * > * Returns NULL if the object is missing or

Re: [PATCH] Add failing test: "fsck survives inflate errors"

2014-07-20 Thread Samuel Bronson
The following message is a courtesy copy of an article that has been posted to gmane.comp.version-control.git as well. Oh, I forgot to provide any analysis of the problem. Oops. It may be just as well, though; I was tired enough that I might have botched it in any case. So, have an analysis: W

[PATCH] Add failing test: "fsck survives inflate errors"

2014-07-19 Thread Samuel Bronson
While inflate errors are obviously NOT GOOD, and should perhaps be fatal for most commands, git fsck is something of a special case because it is useful to have *it* report as many corrupt objects as possible in one run. Signed-off-by: Samuel Bronson --- t/t1450-fsck.sh | 17 + 1