GLPILIB2

glpilib2 is an easy to use python library for interfacing with GLPI's API.

Supports GLPI 10

Features

• Lots of tests for each API method
• A comprehensive documentation that dwelves into unusual behaviors of GLPI's
• Some obtuse errors that GLPI throws are wrapped with more meaningful ones

How to install

How to use

The usage is fairly straightforward:

1. Create an instance of the RequestHandler class.
2. Initialize a session.
3. Use the RequestHandler's methods.
4. kill the session.

Please see the individual methods documentation for more.

Example:

from glpilib2 import RequestHandler
handler = RequestHandler(
    host_url,
    app_token,
    user_api_token,
)
handler.init_session()
ticket = handler.get_item("Ticket", 1)
handler.delete_items("Software", [6])
handler.kill_session()

Wondering how to fill those variables called host_url, app_token or user_api_token? Read along!

Getting GLPI API to work

GLPI can be a bit tricky to get it to work with the API. In brief you will need a valid app_token a user_api_token and of course the host name/ip of the server that you will be accessing, for the sake of brevity we will refer to it as host_url.

Please note that the host_url must include the protocol http:// or https:// and the path to the GLPI itself if applicable. Ex: http://localhost:8000 or https://www.myhost.com/glpi .

Getting an app_token

First you need to create a API Client object. It's not documented anywhere when I last looked for it, so we will have to navigate through a few menus to find it.

1. Make sure that you are logged into GLPI and your user is using a super-admin profile.

2. Navigate to the General Setup screen that's usually in host_url/front/config.form.php .

3. On the left menu bar click API link.

4. Make sure that the Enable Rest API and Enable login with external token are set to Yes.

   Note: You could also enable login with credentials but it's neither safer, not easier to do, so I'm not giving you that option for the sake of simplicity (on my part).

5. Note the URL of the API field we will be using that to make our requests, however please also observe that up to GLPI 9.5.3 the method will always be HTTP even if you use HTTPS in your host_url.

6. If you are going to be using GLPI from localhost, the full access from localhost object should be enough for you (if it exists) and you can stop right here and skip to Getting an user_api_token.

7. If you are using the docker-compose.yml that I have included in this project or you are accessing a remote host (most likely). You will need to create a new API Client object. For that click the Add API client button.

8. Fill the obligatory fields and make sure the new API Client is set as Active. The IP range should be within the range of the client (whatever device that will be making calls to the GLPI API). Make sure that the Application Token's Regenerate checkbox is marked and click the Add button.

9. Go back to the previous screen and click the new API Client you just created. Take note of value of the app_token field.

Getting an user_api_token

Now you need an user_api_token. This is the key that informs the API which user is trying to access the API.

Getting this value is far more straightforward than the previous one.

1. As a super-admin profile use the Administration menu and select Users.
2. Find the user that you want to use to access the API and click on it.
3. Close to the bottom of the first page that opens you will see a Remote access keys section. Beneath it there is a API token label. If there's a field there, copy that value. That's our user_api_token.
4. If there's no value check the Regenerate box on the right side of it and click Save. After the page reloads there should be a field next to the label copy it as described in the last step.

Testing your settings

Now we should be almost done. You can test that the you can access the api with the parameters we just collected.

Example:

$ curl -X GET \
    -H 'Content-Type: application/json' \
    -H "Authorization: user_token <user_api_token>" \
    -H "App-Token: <app_token>" \
    '<host_url>/apirest.php/initSession?get_full_session=true'

< 200 OK
< {
<     "session_token": "83af7e620c83a50a18d3eac2f6ed05a3ca0bea62"
< }

Source: https://github.com/glpi-project/glpi/blob/master/apirest.md#init-session

If you got an answer 200 OK as in the previous example you are done and can plug the parameters you just collected to the library as mentioned on the How to use section.

Otherwise there are a few things that might have gone wrong. Check the documentation for common errors.

Now, if after making sure that every parameter is set correctly you are still getting ERROR_LOGIN_PARAMETERS_MISSING. There's the possibility that the application server that's hosting (usually Apache) GLPI is removing the headers with the authentication data. Check this bug report And this server configuration guide for more info.

DISCLAIMER

GLPI API is quirky, some options don't work, some things aren't documented and the documentation doesn't always describes what the software actually does. Besides that GLPI is known to be prone to break a few things between updates. While I've done my best to shield the user from all of this with this library, sometimes unexpected errors will leak to the user. Please bear with me as we travel along this bumpy road.

License

This project is licensed under the GPL-2.0 license.