Add --payload to devserver to allow the update payload to be specified by the devserver user.

Add --payload to devserver to allow the update payload to be specified by the devserver user.

Should have no effect if --payload is not specified, but should cause update.gz and stateful.tgz to be symlinked to
the specified payload files if it is.

Right now, it expects you to specify the file update.gz, and for stateful.tgz to be in the same directory. That could easily be changed to only specify the directory.

BUG=chromium-os:8207
TEST=Ran by hand.

Committed: http://chrome-svn/viewvc/chromeos?view=rev&revision=0c880e2

[sosa, 2010-11-18 01:47:43]

Why that change from symlinks to copys?

http://codereview.chromium.org/5109004/diff/8001/autoupdate.py
File autoupdate.py (right):

http://codereview.chromium.org/5109004/diff/8001/autoupdate.py#newcode523
autoupdate.py:523: if self.forced_payload:
No indention for multiline comments

http://codereview.chromium.org/5109004/diff/8001/autoupdate.py#newcode528
autoupdate.py:528: self._Copy(self.forced_payload,
os.path.join(static_image_dir,
Why not just symlink like the caches?

http://codereview.chromium.org/5109004/diff/8001/devserver.py
File devserver.py (right):

http://codereview.chromium.org/5109004/diff/8001/devserver.py#newcode213
devserver.py:213: if options.exit:
if not options.exit start dev server

[dgarrett, 2010-11-18 01:53:44]

I started trying to manipulate these files from cros_au_test_harnes.py, and
the symlinks would be broken from outside the chroot. Also, if I modified
the files without making copies, I would end up modifying the files inside
the cache directly.

On Wed, Nov 17, 2010 at 5:47 PM, <sosa@chromium.org> wrote:

> Why that change from symlinks to copys?
>
>
> http://codereview.chromium.org/5109004/diff/8001/autoupdate.py
> File autoupdate.py (right):
>
> http://codereview.chromium.org/5109004/diff/8001/autoupdate.py#newcode523
> autoupdate.py:523: if self.forced_payload:
> No indention for multiline comments
>
> http://codereview.chromium.org/5109004/diff/8001/autoupdate.py#newcode528
> autoupdate.py:528: self._Copy(self.forced_payload,
> os.path.join(static_image_dir,
> Why not just symlink like the caches?
>
> http://codereview.chromium.org/5109004/diff/8001/devserver.py
> File devserver.py (right):
>
> http://codereview.chromium.org/5109004/diff/8001/devserver.py#newcode213
> devserver.py:213: if options.exit:
> if not options.exit start dev server
>
>
> http://codereview.chromium.org/5109004/
>



-- 
Don

[sosa, 2010-11-18 01:57:43]

LGTM then w/ those other nits

On Wed, Nov 17, 2010 at 5:53 PM, Don Garrett <dgarrett@chromium.org> wrote:
> I started trying to manipulate these files from cros_au_test_harnes.py, and
> the symlinks would be broken from outside the chroot. Also, if I modified
> the files without making copies, I would end up modifying the files inside
> the cache directly.
> On Wed, Nov 17, 2010 at 5:47 PM, <sosa@chromium.org> wrote:
>>
>> Why that change from symlinks to copys?
>>
>>
>> http://codereview.chromium.org/5109004/diff/8001/autoupdate.py
>> File autoupdate.py (right):
>>
>> http://codereview.chromium.org/5109004/diff/8001/autoupdate.py#newcode523
>> autoupdate.py:523: if self.forced_payload:
>> No indention for multiline comments
>>
>> http://codereview.chromium.org/5109004/diff/8001/autoupdate.py#newcode528
>> autoupdate.py:528: self._Copy(self.forced_payload,
>> os.path.join(static_image_dir,
>> Why not just symlink like the caches?
>>
>> http://codereview.chromium.org/5109004/diff/8001/devserver.py
>> File devserver.py (right):
>>
>> http://codereview.chromium.org/5109004/diff/8001/devserver.py#newcode213
>> devserver.py:213: if options.exit:
>> if not options.exit start dev server
>>
>> http://codereview.chromium.org/5109004/
>
>
>
> --
> Don
>

[dgarrett, 2013-05-30 21:02:28]

https://codereview.chromium.org/5109004/diff/8001/autoupdate.py
File autoupdate.py (right):

https://codereview.chromium.org/5109004/diff/8001/autoupdate.py#newcode523
autoupdate.py:523: if self.forced_payload:
On 2010/11/18 01:47:43, sosa wrote:
> No indention for multiline comments

Done.