[bitbake-devel] [PATCH] lib/bb/fetch2: don't skip checksums when local mirror enabled

Gary Thomas gary at mlbassoc.com
Wed Apr 11 10:47:11 UTC 2012


On 2012-04-11 03:11, Paul Eggleton wrote:
> On Wednesday 11 April 2012 00:25:24 Richard Purdie wrote:
>> On Thu, 2012-04-05 at 18:07 +0100, Paul Eggleton wrote:
>>> If you have PREMIRRORS set up to point to a local directory, the local
>>> download function does nothing and always succeeds and the mirror code
>>> was writing a .done stamp regardless, thus the checksum verification was
>>> being skipped even if the file wasn't actually copied from the mirror.
>>>
>>> Signed-off-by: Paul Eggleton<paul.eggleton at linux.intel.com>
>>> ---
>>>
>>>   lib/bb/fetch2/__init__.py |    7 ++++---
>>>   1 files changed, 4 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/lib/bb/fetch2/__init__.py b/lib/bb/fetch2/__init__.py
>>> index e5c4b34..414cc2b 100644
>>> --- a/lib/bb/fetch2/__init__.py
>>> +++ b/lib/bb/fetch2/__init__.py
>>>
>>> @@ -476,9 +476,10 @@ def try_mirrors(d, origud, mirrors, check = False):
>>>               if not os.path.exists(ud.donestamp) or
> ud.method.need_update(newuri, ud, ld):
>>>                   ud.method.download(newuri, ud, ld)
>>>
>>> -                open(ud.donestamp, 'w').close()
>>> -                if hasattr(ud.method,"build_mirror_data"):
>>> -                    ud.method.build_mirror_data(newuri, ud, ld)
>>> +                if os.path.exists(ud.localpath):
>>> +                    open(ud.donestamp, 'w').close()
>>> +                    if hasattr(ud.method,"build_mirror_data"):
>>> +                        ud.method.build_mirror_data(newuri, ud, ld)
>>>
>>>               if not ud.localpath or not os.path.exists(ud.localpath):
>>>                   continue
>>
>> I messed up some commands and this ended up merged with the following
>> commit I manipulated:
>>
>> http://git.openembedded.org/bitbake/commit/?id=c9eed04c6275ef2c694f89e047f85
>> c7de76f89b6
>>
>> Since its pushed at this point I'm not sure what options I have to undo
>> it. Sorry :(
>
> No worries - but does this fix in any way help with the problem Gary was
> seeing?

No, it does not.  See https://lists.yoctoproject.org/pipermail/poky/2012-April/007752.html

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------




More information about the bitbake-devel mailing list