michas's profile - activity

Sounds like you want some separate registration page, which creates the user in LDAP. Afterwards the user can use those credentials to access all your services. (Instead of LDAP you could also use OpenID but the registration issue will be the same.)

Well, LDAP is usually used in a company that manages all employees in a central LDAP server. If you do not have a fixed set of users, maybe another solution would be better. What is your situation? Why do you think LDAP would be good for you?

Just activate "Always display local login form" and deactivate all other check boxes.

This should leave you with a name/password entry. If you entered your LDAP settings correctly, askbot will validate all logins against your LDAP.

There will be no register button, because askbot is using LDAP read only and therefore cannot create users in your LDAP.

What "two versions" are you talking about? I installed askbot directly from the git repository. There is a correct translation available for "UNANSWERD", but for some reason it is not used.

I switched LANGUAGE_CODE from en to de. Now Askbot is half English and half German.

For example in the header the text

ALL UNANSWERED FOLLOWED

changed to

ALLE UNANSWERED BEOBACHTETE

Here ALL and FOLLOWED translate correctly, while UNANSWERED did not change at all.

The corresponding code in scope_nav.html looks like this:

    >{% trans %}ALL{% endtrans %}</a>
    >{{ settings.WORDS_UNANSWERED|escape }}</a>
    >{% trans %}FOLLOWED{% endtrans %}</a>

The definition of WORDS_UNANSWERED in conf/words.py reads:

settings.register(
    values.StringValue(
        WORDS,
        'WORDS_UNANSWERED',
        default = _('UNANSWERED'),
        description = _('UNANSWERED'),
    )
)

Question: Why is the UNANSWERED case handled different that the ALL and FOLLOWED case? Shouldn't the UNANSWERED be translated nevertheless?

UPDATE: Another local installation of Askbot translates perfectly fine. Could you think of any configuration error resulting in such a behavior?

Which version of Askbot do you run? Did you make any custom modifications on the code? Did you already try `DEBUG=True` as the message suggests?

Askbot is based on django, which features explicit CSRF protection.

Please read djangos documentation on that topic and have a look at the askbot source code corresponding to your installed version of askbot.

Also try to set DEBUG=True in your settings.py and try to trigger that situation from one of the affected browsers afterwards.

pull request sent.

Please give some more detail. What version of askbot did you install? How did you start it? Where is the error displayed?

In site_settings.py description for the title option is set as "Site title for the Q&A forum".

This string is later escaped by django.

And later again escaped by the templating engine. Therfore the final page shows "Site title for the Q&A forum".

It seems like it is actually a bug in django, which is fixed at version 1.6 and newer.

Is there any reason for insisting on 'django<=1.5' for Askbot?

• Go to the settings tab
• Select the "Content sharing" section.
• Uncheck the boxes

Sorry for the "htttp" part. I don't have enough karma to write it correctly.

• Go to the settings tab
• Select the "Login provider setings" section.
• Uncheck all boxes but the second.

This should deactivate all OpenID providers and always just shows the Login and password fields.

I am currently using the avatars from http://gravatar.com, but I would like to use the avatars served by our local Confluence. The URLs are constructed the same way. See the updated question.

We have a local Confluence with "Confluence Avatar Server" installed.

This works exactly like gravatar but serves the avatars configured in confluence: With gravatar you use an URL like "htttp://www.gravatar.com/avatar/SOMEHASH" - With Confluence you use a URL like "htttps://example.org/rest/cas/1.0/avatar/server/SOMEHASH". The hash is computed exactly the same way.

Is there an easy way to tell Askbot to use the avatars from confluence instead of gravatar?

A quick scan through the source shows multiple occurrences of "gravatar.com/avatar". Would it be possible to extract the ""htttp://www.gravatar.com/avatar/" part into a variable and make it configurable?

This way I could simply set it to "htttps://example.org/rest/cas/1.0/avatar/server/" to get the avatars from confluence.

I am running my own askbot instance.

Next to each question there are icons for sharing the question on twitter, facebook,etc.

Is there an easy way to disable those icons?

(Pointers to useful documentation are always welcome.)

I set up an Askbot-Instance at my company and connected it with LDAP for authentication.

The login page still shows the Options for OpenID.

Is there an easy way to disable OpenID completely and only allow LDAP logins?

(I am aware of question 2857, but do not see any solution there. - I use version 0.7.49 of askbot.)