Closed (won't fix)
Project:
Drupal.org security advisory coverage applications
Component:
module
Priority:
Normal
Category:
Task
Assigned:
Unassigned
Reporter:
Created:
24 Mar 2011 at 12:22 UTC
Updated:
11 Jun 2012 at 17:27 UTC
Easy to use Facebook login using Open Graph API.
- no Javascript
- no external libraries
- easy configurable, only set up App ID and App Secret
- easy to use for user - one click for login, one click for access permissions on Facebook
- no other features, only easy registering and login for your website by Facebook
Comments
Comment #1
mrf commentedPlease have a look at http://drupal.org/project/fbconnect and http://drupal.org/project/fb and explain how your project approaches Facebook authentication in a fundamentally different manner.
If your project is just a simplified approach of the above it would be better to work with the maintainers of one of those modules to improve usability than to create a new project.
Comment #2
gajdusek commentedFacebook Connect: This module allows users to login on a Drupal website through the Facebook Connect API
Facebook Connect API is old and without documentation. Facebook now prefer OAuth connect.
And Facebook Connect require download external library.
Drupal for Facebook is in my opinion overkill for simply login.
Look for http://developers.facebook.com/connect.php linked from Drupal for Facebook module page. There was documentation before Facebook Connect was marked as obsolete.
Comment #3
gajdusek commentedComment #4
ralt commentedHello,
Here is a quick review :
Otherwise, looks good.
Have a nice day!
Comment #5
mrf commentedStill have concerns about duplication here.
If FB Connect is still using an outdated API, then better to help them upgrade than create a new project designed to meet the exact same need. Last time I checked all the current Drupal solutions were moving to support Facebook's latest and greatest APIs and would welcome any help getting there.
Currently there are at least two projects that provide this functionality to Drupal sites and the only difference I can see here is in how the API's are implemented.
Comment #6
gajdusek commentedHello,
thanks for your review.
One problem. When I remove function fboauth_settings_form_submit where it call variable_set, set new variable don't work.
Thanks.
Comment #7
quicksketchI ran across this sandbox project while looking for a reasonable Facebook Connect solution. I'll definitely vouch for the fact that all the existing Facebook Connect solutions are not adequate for the following reasons:
- No Facebook solution properly requests adequate permissions from Facebook (through the OAuth system). This makes it so that they are unable to request extensive information from Facebook. This includes e-mail (see next)...
- No solution seems to be able to pull in a user's e-mail address. This is because the OAuth login system is the only way you can get access to a user's e-mail currently. The old APIs used to provide e-mails but due to API changes (likely caused by legal and privacy issues), the support for e-mail information was abruptly yanked. This makes it so no solution can adequately create a Drupal user account that functions properly.
- No solution seems to use the "username" property when creating Drupal user names, despite that Facebook has supported this for some time, they all create accounts with real names or Facebook numeric IDs.
- Some solutions out there have drastically overreached (FBConnect), providing functionality that is obscurely related to their original tasks ("Invite your friends"?) The FB project has done a much better job of segregating functionality, but unfortunately is so behemoth overall that it would take a significant amount of work to properly use the new APIs throughout the whole system.
Additionally there are just a few major differences in approach:
- This module does not utilize XFBML, which means less dependency on Facebook's JavaScript APIs and an easier installation. It also means that incorporating this approach with other projects will be more difficult.
- This module does not use the Facebook PHP SDK, which means fewer incompatibilities with updated versions of the SDK. And again, would make this module harder to integrate with other modules that do use the SDK.
So all told, I think there's a strong case for a new module here.
Unfortunately I'm at a loss as to whether this project should actually be created because of a lack of history by the author. We have absolutely no history of contributions, no patches, no maintenance record, etc. I have a bad feeling that this project would almost immediately suffer the same fate as every other Facebook module and be left pretty much abandoned after a short time.
Just to further derail in things, this is the best implementation I've seen thus far of the current APIs and I needed this functionality now, so I forked gajdusek's sandbox into a separate sandbox that I've set up. The changes are pretty drastic, changing the module from a "connect-only" system to one that can be extended, profile module integration, proper account creation, API documentation, an extensive README and user experience improvements. I have no idea exactly what we should do with this code though, but I know it'd at least be useful to others so I threw it up as a sandbox. http://drupal.org/sandbox/quicksketch/1131300
I at least agree with the fact that current FB modules for Drupal are not very good. I'm not sure that creating a new project is the right way, but I know it's the only way that's going to be implemented in any reasonable amount of time. All the existing projects have taken a completely different approach to Facebook (through the SDK) making this approach incompatible. However even if this new module were accepted, I'm at least concerned about gajdusek's lack of contributor history. The last thing we need is yet another abandoned module.
Comment #8
mrf commentedHaving implemented all the available modules I fully agree with the assessment about the current state of the Drupal and facebook authentication, and after reading quicksketch's explanation I think there definitely is a case for another module here.
Would it be agreeable to everyone to have quicksketch bring on gajduseek as a co-maintainer for his version rather than leave this entirely up to gajduseek to maintain? Having a good contribution history to point to will likely make a future full project application go a lot more smoothly.
Comment #9
mlncn commentedHi gajdusek,
You do not need to be granted approval on this project to be granted vetted maintainer status with the ability to promote new projects. Co-maintain a project with quicksketch and show yourself to be a good maintainer there, and i'll definitely grant you vetted status. (He can also edit the project node to have it in your name if it's clear you are the primary maintainer.
As Nate Haug (quicksketch) is one of the very top contrib maintainers an all of Drupal, i recommend you take advantage of this opportunity.
If you want to complete the usual review process, make sure your code sparkles and you are as responsive here as we hope you will be in your issue queues, and we can approve it-- we'll be watching the module for signs of abandonment and be very quick to change ownership, given quicksketch's implicit co-maintainer offer above.
Note: Another module to check for overlap, i approved another module, Social Connect, recently.
Comment #10
sreynen commentedHi gajdusek,
Are you still interested in pursuing this application or collaborating on an existing project?
Comment #11
sreynen commentedI'm closing this due to no response for months. Please re-open if this was a mistake.
Comment #12
elandirayan commentedHI,
I am using the fboauth plugin in drupal6. will it support user role for registration and also is there any way to redirect to custom page if they uses fboauth for registration ? Please advice.
Thanks
Elan
Comment #13
mrf commentedelandirayan, this isn't the place to get support, please direct support requests to the module's issue queue