This commit adds support for the Vimeo provider in the e-learning platform.
This includes managing Vimeo video links and showing the Vimeo embedded player
in the fullscreen and non-fullscreen viewers.
But also managing video events to:
- automatically set the slide as completed when the video reaches the last 30
seconds
- go to the next slide when the video ends (if there is one)
Task-2510174
Part-of: odoo/odoo#71477
PURPOSE
Globally refactor the way we handle external content (YouTube videos,
Google Drive documents, ...) in the course slides.
This includes both technical changes (new computed fields, rename, ...) as well
as visual changes on the form views that allow introducing new content, both in
the backend (slide.slide form view) and in the frontend (course main website
page).
SPECS
TECHNICAL REFACTOR
1. Rework the "url" field
This field was used without any distinction for videos, regular documents and
images, leading to a big confusion both technically and visually.
The urls are now split in separate fields (video_url, document_google_url and
image_google_url), based on the type of slide.
Those fields are implemented as 'related' to the stored 'url' field to avoid
extra storage.
This change globally increases the readability and usability in the code, and
also allows for clear labels on the various views.
1b. Rework the "binary_content" field
The same way as for the 'url' field, the 'binary_content' field has been split
into 2 related fields ('document_binary_content' and 'image_binary_content').
This allows for more readability and also helps in the form view to limit the
file selection to the supported file types.
2. Remove the "document_id" stored field
This field did not have to be stored since it can "easily" be inferred from the
video/document URL with some regex work.
3. Remove the "mime_type" field
This field's value was only determined by an external call to the Google Drive
API. It was inconsistent because not set for manually uploaded document and was
used for some strange checks that seemed to work but did not make much sense.
(e.g: used to make the difference between a Google Drive video and a YouTube
video??).
It has been replaced by a "document_type" field, which is always 'pdf' for
local files (only type we support) and inferred from the mimeType returned by
Google Drive metadata for external documents.
This allows, for instance, to show pretty icons next to every slide based on
their type, "sheet" files (Excel, Google Sheet, ...) share the same icons, the
same principle is applied to "doc" files (Word, OpenOffice, Google Doc, ...)
and finally "slides" files (PowerPoint, Google Slides, ...).
4. Introduce several computed fields
Such as:
- video_source_type
- youtube_id
- google_drive_id
These fields are mostly inferred from the slide type and the related URL.
They are used to make clear checks in the code and ease readability, for
example when constructing the slide "embed code" to insert it in the frontend
full screen viewer.
5. Remove the "presentation" slide_type
The slide type "presentation" was some kind of very confusing type between
image and pdf. It was mostly relying on the fact that the content had a bigger
width than height ('landscape' display).
This lead to a lot of confusion both in the code base and in the various forms
allowing to introduce new content ("what should I choose? document or
presentation?").
In order to simplify everything, we completely removed the slide type
"presentation" and converted all the data that had this type and the various
code checks to the type "document" instead.
Meaning that we end up with the following slide types:
- Infographic - for images, local files + external links
- Web Page - local input only
- Document local files + external links
- Video - external only, YouTube or Google Drive links
- Quiz
FUNCTIONAL
Following the technical changes here above, we can improve the form view of the
slide.slide model as well as the website form that allows adding new content.
1. Slide form view
To make it clear where the source of the document comes from, we introduced a
"source_type" field that is either "local_file", used when uploading a file
from your computer, or "external", used when linking content from Google Drive.
The source_type selection appears when adding content of type "document" or
"infographic" and conditions the display of the file upload button or the
document_url field.
2. Adding content from the website
When adding content from the website, the user is now invited to select from
ALL the available slide type, meaning we added "Infographic" to the selection.
Before this change, if you wanted to link an image, you had to select the type
"presentation" (that has a PDF file as icon...) and then input an image file.
The JS then processed the file type to determine the slide type.
As this was slightly confusing for the end user, we decided to use the same
approach as the form view, meaning you FIRST select the type of slide you want
and THEN input the content as a local file or external link.
Since we now handle all types of documents when using an external Google Drive
link, the various screen and text helpers have been reworked accordingly.
3. Handle more types of slides with the new "slide_type" field
This commit also introduces a new "slide_type" field.
(Don't get confused, the previous "slide_type" field has been renamed to
"slide_category" in an earlier rename commit).
This slide_type is a refined slide_category:
- Videos are split into 'youtube_video' and 'google_drive_video'
- Documents are refined based on the file mime_type if it's external content
For example, if you link a Excel file, the slide_type will be 'sheet'.
If you link a Google Doc, the slide_type will be 'doc'
etc...
(Local documents are only PDFs since it's the only type we support for now)
This small change allows to include minor nice details into views, such as a
refined icon for each slide_type on the website.
LINKS
Task-2510174
UPG PR odoo/upgrade#2498
Part-of: odoo/odoo#71477
Due to recent legal changes in EU, you cannot install tracking cookies in a user's browser without explicit consent.
Before this patch, Odoo embedded youtube videos were violating that law (AFAIK & IANAL).
With this simple patch, you benefit from YouTube's offering to embed videos without cookies. The UI remains the same, but no tracking is done.
@Tecnativa TT26153
closesodoo/odoo#61828
X-original-commit: a1a1169f28233b69cd70d01621d605b36517ac73
Signed-off-by: Thibault Delavallee (tde) <tde@openerp.com>
PURPOSE
This commit fixes the behavior of archiving and un-archiving slide.channels.
SPECS
When archiving a slide.channel, you could run into an issue where the
associated slide.channel.partners would be marked as "completed" even though
the related slides were NOT completed by the user.
This happened because when computing the completion, it's searching for
slide.slide.partners marked as completed, and then comparing it with the count
of slide.slides in the channel, which would always return 0 since all slides
are archived when the channel is archived.
Marking channels as completed when they should not triggers undesired
side-effects (karma increase, completion email sent, ...).
Now, when the slide.channel is not active, we don't mark associated
slide.channel.partners as completed anymore.
The ordering in which we archive channel and slides was also altered, see
method docstring of 'slide.channel#toggle_active' for more info.
A new test ensures the correct behavior of archiving.
LINKS
Task 2170708
closesodoo/odoo#61773
X-original-commit: f7795f85ed66fcebaf8a96bf4e550b069fcf1e83
Signed-off-by: Thibault Delavallee (tde) <tde@openerp.com>
Signed-off-by: awa-odoo <awa-odoo@users.noreply.github.com>
PURPOSE
Fix the 'multi' behavior when sending completion emails for slide.channels.
SPECS
There was a typo in the 'slide.channel#_send_completed_mail' method that
prevented it from correctly working in "multi" mode.
A new test was introduced to ensure the fix is working.
LINKS
Task 2170708
closesodoo/odoo#61004
Signed-off-by: Thibault Delavallee (tde) <tde@openerp.com>
PURPOSE
Archiving a course should also archive its content, and unpublish it as it
should not be accessible from outside. Improve archive / publish management
between a course and its slides.
SPECIFICATIONS
Show the content of an archived course having archived slides. When slides and
course are archived its content should still be displayed. This is done using
an active test in context of field.
Propagate channel archive to content. Also unpublish archived slides.
LINKS
Continuity/IMP of task-2170708
Task-2225598
closesodoo/odoo#52233
Signed-off-by: Thibault Delavallee (tde) <tde@openerp.com>
This commit fixes the reordering of slides on upload or drag and drop
in frontend. This issue was due to the fact that in a previous commit
(c4eba6f) uncategorized slides were moved to the bottom of the list,
which broke the reordering after drag and drop.
Indeed we use the dom order itself to reorder the slides with a simple rpc
call to dataset/resequence. Instead we revert part of the previous fix
to display uncategorized slides at the top. It allows to be coherent with
resequence and with backend behavior that displays uncategorized at the
top of the list.
Another issue was that, on slide upload, in the controller, we used
the category_id on the slide (which seems to not always be correct at
this stage) instead of the one given by the javascript. This caused the slide
to not be uploaded in the right section.
This commit also adds some margin to the categories in frontend for training
courses. Otherwise uncategorized slides are too close to the first category.
Task ID 2090927
X-original-commit: 7bd042d6e20c44558b90641409a6e3f8fa8000ed
When new slides are added through frontend a resequencing of slides and
categories sequence is done by calling _resequence_slides on the channel.
However this method is not working as expected, and resequencing should
take a category in order to better know where to start its resequencing.
Otherwise it simply reorder slides in the same order as before, or in an
unpredicted or prodicted but fasly order if new slide category has just been
created.
Task ID 2090927
X-original-commit: f038946ae2e3f174bee086281a541c6bc770e354
PURPOSE
Have user groups specific to eLearning to separate them from generic website
editors. Indeed this is a specific work area and require more than pure design
skills.
SPECIFICATIONS
Two user groups are added in this commit
* group_website_slides_officer
+ can read any course/slide
+ can create a course (will be responsible)
+ can write on its courses
+ can upload slides to the courses he is responsible
+ can publish slides to the courses he is responsible
- cannot unlink courses/slides (you should archive them)
* implies group_website_publisher in order to see the top navbar in
frontend with the publish toggle button and the new content button;
* group_website_slides_manager: has full rights
+ can read any course/slide;
+ can create a course;
+ can add slide to any courses;
+ can publish any courses/slides;
+ can unlink any courses/slides;
+ has access to reporting menu entries;
Rights for other groups are untouched for courses / slides. As a reminder
we still support responsibles being outside of eLearning groups, for example
if an employee is a teacher that handles a course:
* people in upload_group_ids of a channel can upload slides, even if they
don't belong to any eLearning group;
* people responsible of a course can upload and publish slides even if they
don't belong to any eLearning group (done through controller as they
don't have real rights on underlying models);
We choose to keep the survey rights separated from eLearning ones. Indeed
it is easy in odoo to update the imply_ids field of groups if necessary
to link survey / eLearning groups; default user template can also be updated
easily. Standard Odoo should therefore correctly separate those groups. The
only specific rights is that eLearning officers can read all surveys, allowing
to choose them for certification-based courses.
LINKS
Task ID 2062462
PR #39485
Signed-off-by: Thibault Delavallee (tde) <tde@openerp.com>
PURPOSE
Like already done for sale order, invoice of survey, purpose of this commit
is to remove category model and replace by a flagged line (slide). It allows
to easily reorder slides in an embedded list view.
SPECIFICATIONS
Instead of having a fully fledged slide.category model, slide.slide will serve
that purpose with a is_category flag. This will allow to drag and drop slides
and sections in the channel form view.
This change had an impact on the way slides were added/sorted on the front-end.
In fact, whenever a slide is added from the front-end, a resequencing of all
the slides in the course has to be triggered.
Category of a slide is now a computed field based on the sequence. Order
of slides is based on sequence, with categories splitting the slide list based
on is_category flag.
In this commit tests are added. Some cleaning in tests is also performed to
speedup a bit tests (savepointcase) and some cleaning / renaming to ease
their understanding.
Future commit will add JS necessary to manage slides in the section list view.
LINKS
TaskID: 1978731
PR: #33255
Aurélien Warnon