Open GoogleCodeExporter opened 9 years ago
matt also points out that appid-switching can happen here.
Original comment by adam.sah
on 14 May 2009 at 9:08
I claim this is useful for launch so we can solve some other problems with
appids and secrets. Here's a start
(totally untested, and in bash). Note that it pulls straight from SVN so
there's no question of 'is this stuff
checked in right'?
# args
appid = $1
fbsecret = $2
# action
tempdir=push_$(date "+%Y%m%dT%H%M%S")
mkdir $tempdir
cd $tempdir
# Grab the files straight from svn. Exact command probably should
reference a tag/rev/etc.
svn export http://footprint2009dev.googlecode.com/svn/trunk/frontend
cd frontend
perl -p -i -e "s/footprint2009dev/$appid/;" app.yaml
perl -p -i -e "s/FACEBOOK_SECRET = '.*'/FACEBOOK_SECRET =
'$fbsecret'/" userinfo.py
appcfg update .
about memcache:
If we always want to flush every time, we can add a property to the keys which is the exact version# of the
app. But I don't think we usually want to flush it.
about admin actions:
appengine_rpc py will handle authentication.
Original comment by matthew....@google.com
on 14 May 2009 at 9:11
test comment-- pls ignore.
Original comment by adam.sah
on 20 May 2009 at 7:36
moving from 1.1, since we're not planning that many releases and can therefore
do
things by hand for a while.
Original comment by adam.sah
on 28 May 2009 at 5:04
I've been looking into this.
Original comment by jblockso...@gmail.com
on 5 Jun 2009 at 5:52
downgrading priority-- not strictly a launch blocker
Original comment by adam.sah
on 8 Jun 2009 at 3:37
Another version out for review, http://footprint2009reviews.appspot.com/29007.
Original comment by jblockso...@gmail.com
on 15 Jun 2009 at 3:34
Working on adding App Engine RPC calls to flush cache at Adam's request. Script
is usable and available in my
home directory, ~jblocksom/push_release.py.
Moving to low priority as I expect at this point Paul / Adam won't care to use
it for upcoming release.
Original comment by jblockso...@gmail.com
on 17 Jun 2009 at 6:12
as a matter of good discipline, since v1.5 is technically live,
re-targeting for v1.6
(since there's few issues in 1.6, they won't get "lost in the noise")
Original comment by adam.sah
on 19 Jun 2009 at 3:41
Original issue reported on code.google.com by
adam.sah
on 14 May 2009 at 7:18