Closed
Bug 570360
Opened 16 years ago
Closed 16 years ago
FlightDeck - AMO dependency
Categories
(Mozilla Labs Graveyard :: FlightDeck, defect)
Tracking
(Not tracked)
RESOLVED
FIXED
1.0
People
(Reporter: zalun, Unassigned)
Details
FD is scraping the AMO website to validate users and to get their user ids.
FD is using mechanize to fill in the following inputs
* data[login][email]
* data[login][password]
on the
https://addons.mozilla.org/en-US/firefox/users/login?to=en-US
FD asserts that authentication was successful if submit redirects to
https://addons.mozilla.org/en-US/firefox/ or https://addons.mozilla.org/en-US/firefox
Then HTML is browsed for a link with the text "View Profile"
The last part of the link (split by "/") is taken as AMO's (and FD's) user ID
Then the system is following the "Edit profile" link. On this page FD is reading the form to fill profile data on the FD system.
Especially data[User][firstname], data[User][lastname], data[User][nickname] are crucial.
It also uses the img element with class "avatar".
If any of the above data changes (like recently the "/" was removed from the "View profile" link FlightDeck is not functioning properly.
There is a need to receive the information about the changes before they're will be live on AMO
| Reporter | ||
Comment 1•16 years ago
|
||
Just as a side note - the code is on http://github.com/zalun/FlightDeck/blob/master/flightdeck/amo/authentication.py
Comment 2•16 years ago
|
||
Yup, good point. In the long run, we'll figure out something better than screen scraping to get access to the authentication database. In the short run, I'll talk to the AMO team about getting notified when they plan to release a new version of AMO.
Comment 3•16 years ago
|
||
I talked to the AMO team about getting notified when they plan to release a new version of AMO, and they agreed to do so, which resolves the issue raised by this bug.
Status: NEW → RESOLVED
Closed: 16 years ago
Resolution: --- → FIXED
Comment 4•15 years ago
|
||
Login was having issues when I tried it today :(
Comment 5•15 years ago
|
||
Daniel: if you are still having problems, can you file a new bug so we can track and address the specific issue you are encountering?
Comment 6•15 years ago
|
||
(In reply to comment #4)
> Login was having issues when I tried it today :(
I sent Piotr credentials to a staging database yesterday so he can auth against the users table directly and avoid all this scraping stuff.
| Reporter | ||
Comment 7•15 years ago
|
||
I'll check the login - it should be applied to production as a quick fix IMO
| Reporter | ||
Comment 8•15 years ago
|
||
It's working for me.
I'm working on the credentials based solution
I couldn't log in today, with an error pointing me to this bug.
In a different profile that has my credentials remembered, it worked fine, though.
Updated•15 years ago
|
Target Milestone: -- → 1.0
You need to log in
before you can comment on or make changes to this bug.
Description
•