5.4. Lesson: Actions

Now that you have seen a default action in the previous lesson, it is time to define your own actions.

An action is something that happens when you click on a feature. It can add a lot of extra functionality to your map, allowing you to retrieve additional information about an object, for example. Assigning actions can add a whole new dimension to your map!

Objectif de cette leçon : Apprendre à ajouter des actions personnalisées.

In this lesson you will use the school_property layer you created previously. The sample data include photos of each of the three properties you digitized. What we are going to do is to associate each property with its image. Then we will create an action that will open the image for a property when clicking on the property.

5.4.1. basic Follow Along: Ajouter un champ pour images

The school_property layer has no way to associate an image with a property yet. First we will create a field for this purpose.

  1. Ouvrez la fenêtre Propriétés de la couche.

  2. Cliquez sur l’onglet Champs.

  3. Basculez en mode édition :

  4. Ajoutez une nouvelle colonne :

  5. Entrez les valeurs ci-dessous :

  6. After the field has been created, move to the Attributes Form tab and select the image field.

  7. Set Widget Type to Attachment:

  8. Click OK in the Layer Properties dialog.

  9. Utilisez l’outil Identifier pour cliquer sur une des trois entités dans la couche school_property.

    Since you are still in edit mode, the dialog should be active and look like this:

  10. Cliquez sur le bouton parcourir (le à côté du champ image).

  11. Select the path for your image. The images are in exercise_data/school_property_photos/ and are named the same as the features they should be associated with.

  12. Cliquez OK.

  13. Associez toutes les images avec les bonnes entités en utilisant cette méthode :

  14. Sauvegardez vos modifications et sortez du mode d’édition.

5.4.2. basic Follow Along: Création d’une Action

  1. Open the Actions tab for the school_property layer, and click on the signPlus Add a new action button.

  2. In the Add New Action dialog, enter the words Show Image into the Description field:


    Ce qui faut ensuite faire varie selon votre système d’exploitation, ainsi choisissez la ligne de conduite appropriée à suivre :

    • Windows

      Cliquez sur le menu déroulant Type et choisissez Ouvrir.

    • Ubuntu Linux

      Under Action, write eog for the Gnome Image Viewer, or write display to use ImageMagick. Remember to put a space after the command!

    • MacOS

      1. Cliquez sur le menu déroulant Type et choisissez Mac.

      2. Under Action, write open. Remember to put a space after the command!

    Now you can continue writing the command.

    Vous souhaitez ouvrir l’image, et QGIS connaît sont emplacement. Il ne reste plus qu’à renseigner le Action où l’image est.

  3. Sélectionnez image dans la liste :

  4. Click the Insert field button. QGIS will add the phrase [% "image" %] in the Action Text field.

  5. Click the OK button to close the Add New Action dialog

  6. Click OK to close the Layer Properties dialog

Now it is time to test the new action:

  1. Click on the school_property layer in the Layers panel so that it is highlighted.

  2. Find the actionRun Run feature action button (in the Attributes Toolbar).

  3. Click on the down arrow to the right of this button. There is only one action defined for this layer so far, which is the one you just created.

  4. Cliquez sur le bouton pour activer l’outil.

  5. En utilisant cet outil, cliquez sur une des trois propriétés de l’école.

    The image for that property should open.

5.4.3. moderate Follow Along: Recherche sur Internet

Let’s say we are looking at the map and want to know more about the area that a farm is in. Suppose you know nothing of the area in question and want to find general information about it. Your first impulse, considering that you’re using a computer right now, would probably be to Google the name of the area. So let’s tell QGIS to do that automatically for us!

  1. Ouvrez la table d’attribut pour la couche landuse.

    We will be using the name field for each of our landuse areas to search Google.

  2. Fermez la table d’attribut.

  3. Retournez à Actions dans Propriétés des couches.

  4. Click on the Create Default Actions button to add a number of pre-defined actions.

  5. Remove all the actions but the Open URL action with the short name Search Web using the signMinus Remove the selected action button below.

  6. Double-click on the remaining action to edit it

  7. Change the Description to Google Search, and remove the content of the Short Name field.

  8. Make sure that Canvas is among the checked Action scopes.

    Ce qui faut ensuite faire varie selon votre système d’exploitation, ainsi choisissez la ligne de conduite appropriée à suivre :

    • Windows

      Sous Type, choisissez Ouvrir. Cela va dire à Windows d’ouvrir une adresse internet dans votre navigateur par défaut, tel qu’Internet Explorer.

    • Ubuntu Linux

      Under Action, write xdg-open. This will tell Ubuntu to open an Internet address in your default browser, such as Chrome or Firefox.

    • MacOS

      Under Action, write open. This will tell MacOS to open an Internet address in your default browser, such as Safari.

    Now you can continue writing the command

    Quelle que soit la commande que vous avez utilisée en-dessus, vous devez dire quelle adresse internet ouvrir après. Vous voulez visiter Google, et rechercher automatiquement une phrase.

    Usually when you use Google, you enter your search phrase into the Google Search bar. But in this case, you want your computer to do this for you. The way you tell Google to search for something (if you don’t want to use its search bar directly) is by giving your Internet browser the address https://www.google.com/search?q=SEARCH_PHRASE, where SEARCH_PHRASE` is what you want to search for. Since we don’t know what phrase to search for yet, we will just enter the first part (without the search phrase).

  9. In the Action field, write https://www.google.com/search?q=. Remember to add a space after your initial command before writing this in!

    Now you want QGIS to tell the browser to tell Google to search for the value of name for any feature that you could click on.

  10. Sélectionnez le champ nom.

  11. Click Insert button:


    What this means is that QGIS is going to open the browser and send it to the address https://www.google.com/search?q=[% "name" %]. [% "name" %] tells QGIS to use the contents of the name field as the phrase to search for.

    So if, for example, the landuse area you click on is named Marloth Nature Reserve, QGIS is going to send the browser to https://www.google.com/search?q=Marloth%20Nature%20Reserve, which will cause your browser to visit Google, which will in turn search for « Marloth Nature Reserve ».

  12. If you have not done so already, set everything up as explained above.

  13. Click the OK button to close the Add New Action dialog

  14. Click OK to close the Layer Properties dialog

Il est maintenant temps de tester la nouvelle action.

  1. With the landuse layer active in the Layers panel, click on the down arrow to the right of the actionRun Run feature action button, and select the only action (Google Search) defined for this layer.

  2. Click on any landuse area you can see on the map. Your browser will now open, and will start a Google search for the place that is recorded as that area’s name value.


Si votre action ne fonctionne pas, vérifiez que tout a été correctement entré ; les fautes de frappe sont communes avec ce genre de travail !

5.4.4. hard Follow Along: Ouvrir une page web directement dans QGIS

Above, you’ve seen how to open a webpage in an external browser. There are some shortcomings with this approach in that it adds an unknowable dependency – will the end-user have the software required to execute the action on their system? As you’ve seen, they don’t necessarily even have the same kind of base command for the same kind of action, if you don’t know which OS they will be using. With some OS versions, the above commands to open the browser might not work at all. This could be an insurmountable problem.

However, QGIS sits on top of the incredibly powerful and versatile Qt library. Also, QGIS actions can be arbitrary, tokenized (i.e. using variable information based on the contents of a field attribute) Python commands!

Now you will see how to use a python action to show a web page. It is the same general idea as opening a site in an external browser, but it requires no browser on the user’s system since it uses the Qt QWebView class (which is a webkit based html widget) to display the content in a pop-up window.

Let us use Wikipedia this time. So the URL you request will look like this:


Créer l’action associée à la couche:

  1. Ouvrez la boîte de dialogue Propriétés de la couche et rendez-vous sur l’onglet Actions.

  2. Paramétrez une nouvelle action en utilisant les propriétés suivantes:

    • Type: Python

    • Description: Wikipedia

    • Action Text (all on one line):

      from qgis.PyQt.QtCore import QUrl; from qgis.PyQt.QtWebKitWidgets import QWebView; myWV = QWebView(None); myWV.load(QUrl('https://wikipedia.org/wiki/[%name%]')); myWV.show()

    Il y a quelques choses qui se passent par ici :

    • Tout le code python est sur une seule ligne avec des points virgules séparant les commandes (à la place de nouvelles lignes, la manière habituelle de séparer des commandes python).

    • [%name%] will be replaced by the actual attribute value when the action is invoked (as before).

    • The code simply creates a new QWebView instance, sets its URL, and then calls show() on it to make it visible as a window on the user’s desktop.

    Notez que c’est un exemple un peu farfelu. Python fonctionne avec une indentation qui a un sens, par conséquent si l’on sépare les choses par des points-virgules ce n’est pas la meilleure manière de faire. Ainsi, dans la vraie vie, il est plus fréquent d’importer les éléments d’un module en Python, puis d’appeler une fonction avec comme paramètre un attribut de champ.

    You could also use this approach to display an image without requiring that the users have a particular image viewer on their system.

  3. Try to use the methods described above to load a Wikipedia page using the Wikipedia action you just created.

5.4.5. In Conclusion

Actions allow you to give your map extra functionality, useful to the end-user who views the same map in QGIS. Due to the fact that you can use shell commands for any operating system, as well as Python, the sky is the limit in terms of the functions you could incorporate!

5.4.6. What’s Next?

Now that you’ve done all kinds of vector data creation, you will learn how to analyze the data to solve problems. That is the topic of the next module.