Ability to create users from command line

[Christopher Allan Webber]

A script like:

::

    ./bin/gmg adduser <username> <password> <email>

is clearly needed.

We could also use:

::

    ./bin/gmg makeadmin <username>
    ./bin/gmg changepw <username>

for making an existing user an admin or changing a user's
password.

[Alejandro Villanueva]

patch.tar.gz

[Alejandro Villanueva]

Just implement CLI support to add users, it does NOT validate the
email, and the connection is hardcoded i couldn't find the way to
create the conn from the app\_config.

Usage: ./bin/gmg adduser -u ialex -p secretword -m
`admin@ialex.org <mailto:admin@ialex.org>`_

The user doesn't need verification after created.

[Christopher Allan Webber]

Hey, thanks for this patch(es)!

Yeah the config stuff / setting up a connection is going to be
refactored soon and is currently a mess. I'll look at merging this
shortly.

I've added you to the hackers/reporters lists in redmine.

[Alejandro Villanueva]

0003-Add-changepw-and-makeadmin-to-gmg-commands.patch.tar.gz

[Alejandro Villanueva]

I added the another two commands mentioned on the feature, now we
can change passwords and make admin to already registeres users.

usage: ./bin/gmg makeadmin-u username
usage: ./bin/gmg changepw -u username -p new\_password

need some testing to see if everything works fine at least for me
it does, and change the database stuff mentioned earlier.

[Christopher Allan Webber]

First of all, sorry that it's taken me so long to comment on this.
Second, thanks for all the work you've done on this... you got all
three commands, nice!

That said, I'm hoping you can make a few adjustments before I merge
this.


-  Try and make sure your code is PEP-8 compliant.
   `http://www.python.org/dev/peps/pep-0008/ <http://www.python.org/dev/peps/pep-0008/>`_
   ... in pacticular:
   
   -  Avoid going over 80 characters (right now it's just your
      comments that have done this, so please wrap them)
   -  when doing an if statement, no need to do "if (foo bar):"... in
      python "if foo bar:" more appropriate
   -  Put two newlines between your functions

-  Pull the connection information out of the config. Yeah, okay,
   not your fault, and you already acknowledged this :). Here's how to
   do it:

Look at the way that gmg\_commands/shell.py does it:

::

    from mediagoblin import mg_globals
    from mediagoblin.gmg_commands import util as commands_util
    
    def shell_parser_setup(subparser):
        subparser.add_argument(
            '-cf', '--conf_file', default='mediagoblin.ini',
            help="Config file used to set up environment")
    
    [...]
    
    def shell(args):
        """ 
        Setup a shell for the user
        """ 
        mgoblin_app = commands_util.setup_app(args)

... After you do this, you'll have access to the database through
"mg\_globals.database".

Looking good though! Add this stuff in another patch and I think we
can probably merge this. If you don't have time / can't do this,
let us know. Otherwise looking forward to seeing that patch!

[Alejandro Villanueva]

It's Monday and its time to work, will work on that today i will
put everything on a branch since im tired of patches haha, i'll
send the corrections greetz

[Christopher Allan Webber]

Wonderful! I prefer patches also. I've assigned this to you and
targeting to 0.0.3 because that looks likely and reasonable now.

[Alejandro Villanueva]

0002-Adding-new-function-to-gmg_commands-adduser-makeadmi.patch.tar.gz

[Alejandro Villanueva]

here is the patch, made changes to be pep8 compatible, and change
the database stuff.

[Christopher Allan Webber]

Looking! Btw, I said "I prefer patches also" and meant "I prefer
branches also", so if you'd rather work in branches from now on
please, by all means ;)

[Christopher Allan Webber]

Merged! I also made a bit of a change and made the arguments
positional instead.

Thanks for your work on this Alejandro, wonderful job!

[Alejandro Villanueva]

Great! time to close this feature

[Will Kahn-Greene]

The original url for this bug was ​http://bugs.foocorp.net/issues/373 .
Relations:
#88: blocks