Hi:

I tried http://w2popenid.appspot.com/oidconsumer

It is amazing to see the web2py Admin page up on the App Engine.

A few results:

1. dly...@gmail.com does NOT work.
   I (and my associates) want the standard email address to work. The
URL Open ID is too clunky for most people we deal with.

2. I tried a URL version of my OpenID, which is

       http://openid-provider.appspot.com/dlypka

   That got me to the Open ID page, but it asked me to first log into
Google, which I believe is wrong, since I am already logged in to this
web2py forum.

   So I conclude that the API is working great from web2py on GAE  to
OpenID, but it would be nice to have it recognize that I am already
logged into Google.
   However I went ahead and did the Google login that it was asking
for, and then I did the the Yes response and it then gave me that page
with
   the response codes:

resp    :       <openid.consumer.consumer.SuccessResponse id='http://openid-
provider.appspot.com/dlypka' signed=['openid.return_to',
'openid.mode', 'openid.identity', 'openid.sreg.nickname']>
admin request session response

That was great!

Then I closed the browser and went back in to try it all a second
time, and this time it did NOT ask me to log into Google.
So that was good, that it recognized my current Google login which it
had "seen' me do 'inside' OpenID
I assume it does not recognize my previous login to this forum,
because it was done 'outside' of OpenID.

I guess OpenID only recognizes logins which were done 'inside' its
API.
If so, that is a big negative for the immediate future. It means at
the moment, the regular user would have to find the 'special' OpenID-
enabled
login page for a given website in order to benefit from OpenID.


-------------------------------------------------------------------------

I looked at http://pastebin.co.za/22362
For that, it looks like one needs to also have the code for
openid.store
I assume the word 'store' means 'storage', rather an a 'store' where
one buys things...

Then I downloaded web2py.app.oidconsumer.w2p.gz
I extracted it and got the oidconsumer.w2p, but web2py failed to be
able to install it.
So I renamed the .w2p to be a .tar and I unpacked it with winrar.
That gave me the oidconsumer folder with all the source.
I copied that folder into my web2py v 1.65.5 Applications folder
and voila, it ran perfectly and still recognized my previous Google
login (which I had done 'inside' OpenID).
So I guess the code for openid.store is in there somewhere...

Next I will try it on App Engine. I assume it will work.

Thanks so much!


On Jul 24, 8:21 am, hcvst <hcv...@googlemail.com> wrote:
> Hi,
>
> I've uploaded a working openid consumer 
> tohttp://w2popenid.appspot.com/oidconsumer
> It uses the Web2pyStore I posted above (http://pastebin.co.za/22362)
> and a custom fetcher (Web2pyFetcher) that permits headers etc. to be
> send.
>
> Please see whether it works for you. My battery is about to die, so I
> don't have time to
> post the source. Will post it later.
>
> Cheers,
> HC
>
> You can download the app fromhttp://w2popenid.appspot.com/
>
> On Jul 23, 4:16 pm, Hans Donner <hans.don...@pobox.com> wrote:
>
>
>
> > Don't know about the content like the name of the nick, you should be
> > able to tell.
> > But the fact you get shown a nick indicates everything went well.
>
> > Remember, it's now only a proof of concept app, so nothing usefull is
> > yet done besides getting your openid info
>
> > On Thu, Jul 23, 2009 at 12:05, dlypka<dly...@gmail.com> wrote:
>
> > > Thanks so much for this work.
> > > I tried it.
>
> > > It correctly took me to the Open ID confirmation 
> > > athttp://openid-provider.appspot.com
> > > and then  I clicked Yes.
> > > This is the result I got back:
>
> > > message
> > > :
> > > something heppened:{'url': 'http://openid-provider.appspot.com/
> > > dlypka', 'pape': <openid.extensions.draft.pape5.Response object at
> > > 0x017CDD70>, 'sreg': [('nickname', 'dlypka')]}
> > > admin request session response
>
> > > Is this the response to be expected?
>
> > > I am not sure what the expected result should be.
>
> > > Thanks.
>
> > > On Jul 21, 7:57 pm, Bottiger <bottig...@gmail.com> wrote:
> > >> I've uploaded to my website the minimal version of the working OpenID
> > >> implementation application. Just unzip in your application directory
> > >> and go tohttp://127.0.0.1:8000/openid/client/startOpenIDtostart.
>
> > >> I couldn't create a w2p file from it because Web2Py kept complaining
> > >> about an internal error.
>
> > >>http://www.codexon.com/temp/openid.zip
>
> > >> On Jul 21, 4:27 pm, Bottiger <bottig...@gmail.com> wrote:
>
> > >> > Yes, I accidentally missed your 2nd message and fixed it on my own. I
> > >> > also found another error.
>
> > >> > So as a canonical reference, here are the 3 things that need to be
> > >> > edited to get Massimo's OpenID to work.
>
> > >> > 1.
>
> > >> > trust_root = 'http://127.0.0.1:8000/openid/client/startOpenID
> > >> > change this to
> > >> > trust_root = 'http://127.0.0.1:8000/openid/client/
>
> > >> > 2.
>
> > >> > remove this extraneous line at 105. It creates a URL html tag when it
> > >> > is supposed to be a simple URL.
>
> > >> > return_to = URL(r=request,f='finishOpenID')
>
> > >> > 3.
>
> > >> > change this at line 130. sreg_response was None for me. I believe sreg
> > >> > is an optional argument that some OpenID providers may not provide.
>
> > >> > 'sreg': None if sreg_response is None else sreg_response.items(),
>
> > >> > So now we finally have a long overdue working OpenID implementation
> > >> > for Web2Py.
>
> > >> > On Jul 21, 1:14 pm, Hans Donner <hans.don...@pobox.com> wrote:
>
> > >> > > commenting out the return_to in finishOpenID did the trick (it will
> > >> > > take the return to as defined in the constant. The URL only provides 
> > >> > > a
> > >> > > path without the servername.
>
> > >> > > So two minor changes made it work...
>
> > >> > > On Tue, Jul 21, 2009 at 22:07, Hans Donner<hans.don...@pobox.com> 
> > >> > > wrote:
> > >> > > > I've used Massimo's code, installed it in a new app and executed it
> > >> > > > (using myopenid).
>
> > >> > > > I'd had to change
> > >> > > > #trust_root = 'http://127.0.0.1:8000/openid/client/startOpenID'
> > >> > > > trust_root = 'http://127.0.0.1:8000/openid/client'
> > >> > > > return_to = 'http://127.0.0.1:8000/openid/client/finishOpenID'
>
> > >> > > > to give me a proper reponse
>
> > >> > > > Otherwise I got:
> > >> > > > u'http://127.0.0.1:8000/openid/client/finishOpenID?janrain_nonce=2009-0...
> > >> > > > not under trust_root
> > >> > > > u'http://127.0.0.1:8000/openid/client/startOpenID'
>
> > >> > > > Does that help you? (the error message you posted is not helping 
> > >> > > > here)
>
> > >> > > > The next stop is web2py's response:
> > >> > > > message
> > >> > > > :
> > >> > > > something heppened:{'failure_reason': "return_to does not match 
> > >> > > > return
> > >> > > > URL. Expected '/openid/client/finishOpenID', got
> > >> > > > 'http://127.0.0.1:8000/openid/client/finishOpenID?janrain_nonce=2009-0...";,
> > >> > > > 'error': 'OpenID authentication failed.'}
>
> > >> > > > On Tue, Jul 21, 2009 at 10:46, Bottiger<bottig...@gmail.com> wrote:
>
> > >> > > >> I don't need wireshark. Its in plain sight. The problem is I have 
> > >> > > >> no
> > >> > > >> idea what is wrong with it.
>
> > >> > > >>https://open.login.yahooapis.com/openid/op/auth?openid.assoc_handle=c...
>
> > >> > > >> where test.domain.com is an actual subdomain I own that is 
> > >> > > >> mod_proxied
> > >> > > >> to web2py.
>
> > >> > > >> On Jul 21, 1:18 am, Yarko Tymciurak <yark...@gmail.com> wrote:
> > >> > > >>> you might try looking at what gets sent back and forth to try to 
> > >> > > >>> discover
> > >> > > >>> what's wrong;  wireshark or LiveHTTPHeaders for Firefox might 
> > >> > > >>> help...
>
> > >> > > >>> On Tue, Jul 21, 2009 at 3:10 AM, Bottiger <bottig...@gmail.com> 
> > >> > > >>> wrote:
>
> > >> > > >>> > I've been trying to get Massimo's OpenID implementation to 
> > >> > > >>> > work, which
> > >> > > >>> > seems to be the only OpenID implementation for Web2Py.
>
> > >> > > >>> > It doesn't seem to work with either Yahoo or Google, each time 
> > >> > > >>> > saying:
>
> > >> > > >>> > "Sorry! Something is not quite right with the request we 
> > >> > > >>> > received from
> > >> > > >>> > the website you are trying to use. Please try again in a few 
> > >> > > >>> > minutes."
>
> > >> > > >>> > I've hosted it on a public domain and changed trust_root and 
> > >> > > >>> > return_to
> > >> > > >>> > variables to be valid, so I know its not a localhost problem. 
> > >> > > >>> > So far I
> > >> > > >>> > can't figure out what's wrong, as it seems to look exactly 
> > >> > > >>> > like the
> > >> > > >>> > Django example on the python-openid site.
>
> > >> > > >>> > Does anyone have similar problems or suggestions? I already 
> > >> > > >>> > know about
> > >> > > >>> > CAS and I don't want to use it.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"web2py-users" group.
To post to this group, send email to web2py@googlegroups.com
To unsubscribe from this group, send email to 
web2py+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/web2py?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to