The redirect infinite loop error that was previously in the bug tracker (not sure where it went) was not fixed with the new release - the error is unchanged from how it was before.
Logged In: YES
I have the same error, but only on one account. 5 other gmail accounts don't produce the error although they gave the same error before version 1.8. Maybe gmail is pushing even more changes and again not to all accounts?
The strangest thing happened on that one account where it produced the same 302 error. IMAP support is gone and so is the beta version, that account seems to have reverted to the old version for some reason, there's no link anymore on top that says "older version" or "new version".
I retract that last comment, that was because I switched languages from us-english to uk-english. I have my imap back as well as a choice to return to the old version. But the error is still there.
Logged In: NO
I have the same issue too.
The last version fixed some redirection errors, but not all of them.
Logged In: YES
So... this doesn't happen to me, so I can't really reproduce this to test it. If one of you has an account you're willing to turn over to me for a little, I might be able to play around and figure something out.
Or, of course, a patch is always welcome :)
Logged In: YES
Ok, I send you and example.
wdaher - is there anything we may need to change in our function calling code since the most recent update? I can't send you my account since it contains personal and sensitive information. I am using the same code that I have been using for the past year to send daily emails, and it has been broken, no matter which libgmail version I use, for the past 3 weeks or so due to this 302 redirect infinite loop error. I posted this bug (I should really open an account, my apologies). I have tried changing the account's language settings, but it did not change anything for me - the error remains the same. Thanks for all your efforts, by the way, this has been an incrediblely useful library for me.
Has there been any news on this? I don't mean to rush and I really appreciate all the work done by the generous volunteers. Just curious if anyone has been able to reproduce and/or fix this since I still cannot send any emails out with the latest version (it broke in November). Thanks.
Yeah, I'm not sure a lot is being done here.
Of course, I'd still love a patch, and I promise to apply it speedily :-P
I appreciate you looking into things, as much as I wish this was an easy fix :). Do you know of any other packages that allow simple Gmail email sending with Python, or any other relatively easy ways to do this? Thanks again.
I mean, now that gmail supports IMAP, you can probably do ~most of the things you'd want to do with libgmail over that interface.
Thanks, I've never played with Python's IMAP interface, but I'll look into that.
I think I've come to a solution: using "ClientCookie" library instead of libgmail's CookieJar class, and making some changes in function "_retrievepage", I can connect to problematic accounts, so it seems there was a problem extracting or setting cookies.
Though I have to do more tests.
@rhauer: Great work! If you're willing to send me a patch (even a really hackish one), I'm happy to help test it out a bit, too.
Ok, in one or two days I will send you a patch, "my" libgmail has a lot of personal modifications.
It definitely works ;).
Oh, I wrote the last comment, I forgot to log in xD.
Special thanks to rhauer for a version that actually solved this problem!
I'd really appreciate if you guys would test the latest cvs version against your accounts to make sure new bugs haven't been introduced/this doesn't break anyone/etc, etc.
One thing to note is that this change introduces a new dependency on ClientCookie (which I've just sucked into CVS, but packagers should take note, I guess)
Fixed in CVS. Please check out the latest version; if it still doesn't work for you there, let us know.