Markup Style Guide
This page covers the conventions for writing and use of the reStructuredText (RST) markup syntax.
Conventions
Three space indentation.
Lines should be less than 120 characters long.
Use italics for button/menu names.
Other loose conventions:
Avoid Unicode characters.
Avoid heavily wrapped text (i.e. sentences can have their own lines).
Headings
#################
Document Part
#################
****************
Document Chapter
****************
Document Section
================
Document Subsection
-------------------
Document Subsubsection
^^^^^^^^^^^^^^^^^^^^^^
Document Paragraph
""""""""""""""""""
Note
Parts should only be used for contents or index pages.
Note
Each .rst
file should only have one chapter heading (*
) per file.
Text Styling
See the overview on ReStructuredText for more information on how to style the various elements of the documentation and on how to add lists, tables, pictures and code blocks. The Sphinx reference provides more insight additional constructs.
The following are useful markups for text styling:
*italic*
**bold**
``literal``
Interface Elements
:kbd:`LMB`
– keyboard and mouse shortcuts.*Mirror*
– interface labels.:menuselection:`3D Viewport --> Add --> Mesh --> Monkey`
– menus.
Code Samples
There is support for syntax highlighting if the programming language is provided, and line numbers can be optionally shown with the :linenos:
option:
.. code-block:: python
:linenos:
import bpy
def some_function():
...
Images
Figures should be used to place images:
.. figure:: /images/interface_window-system_splash_current.png
Image caption.
For consistency, and since it would be good to ensure screenshots are all a similar size when floated next to text, writers should take screenshots in the following manner:
Prepare the area you would like to capture making sure to use the default theme and setting. (In some cases you may not want to use the default settings e.g. if some options are hidden behind a checkbox.)
Zoom to the maximum zoom level (hold NumpadPlus or Ctrl-MMB or similar).
Zoom out eight zoom levels (NumpadMinus – eight times).
In some cases you will want to leave a small margin around the thing you are trying to capture. This should be around 30px but does not have to be exact.
This can be applied to several parts of the interface but might not work for all cases.
Files
No Caps, No Gaps
Lower case filenames underscore between words.
Sort Usefully
Order naming with specific identifiers at the end.
Format
Use .png
for images that have solid colors such as screenshots of the Blender interface, and .jpg
for images with a high amount of color variance, such as sample renders and photographs.
Do not use animated .gif
files, these are hard to maintain, can be distracting and are usually large in file size. Instead use a video if needed (see Videos below).
Location
Place the image in the manual/images
folder. Use no other subfolders.
Naming
For naming files use underscores to separate chapters and sections, and use dashes to separate sections that are two or more words. So for image files should look like: chapter_subsection_sub-subsection_id.png
, e.g:
interface_splash_current.png
interface_undo-redo_last.png
interface_undo-redo_repeat-history-menu.png
Do not use special characters or spaces!
Usage Guides
Avoid specifying the resolution of the image, so that the theme can handle the images consistently and provide the best layout across different screen sizes.
When documenting a panel or section of the UI, it is better to use a single image that shows all of the relevant areas (rather than multiple images for each icon or button) placed at the top of the section you are writing, and then explain the features in the order that they appear in the image.
Note
It is important that the manual can be maintained long term, UI and tool options change so try to avoid having a lot of images (when they are not especially necessary). Otherwise, this becomes too much of a maintenance burden.
Videos
Videos can be embedded from Blender’s self-hosted PeerTube instance which can be found at video.blender.org. To embed a video using the following directive:
.. peertube:: ID
The ID
is found in the video’s URL, e.g:
The ID for https://video.blender.org/videos/watch/47448bc1-0cc0-4bd1-b6c8-9115d8f7e08c
is 47448bc1-0cc0-4bd1-b6c8-9115d8f7e08c
.
To get a new video uploaded, contact a Documentation Project Administrator or include the uploaded video in your Pull Request description.
Usage Guides
Avoid adding videos that rely on voice or words, as this is difficult to translate.
Do not embed video tutorials as a means of explaining a feature, the writing itself should explain it adequately. (Though you may include a link to the video at the bottom of the page under the heading
Tutorials
).
Useful Constructs
|BLENDER_VERSION|
– Resolves to the current Blender version.:abbr:`SSAO (Screen Space Ambient Occlusion)`
– Abbreviations display the full text as a tooltip for the reader.:term:`Manifold`
– Links to an entry in the Glossary.
Cross References and Linkage
You can link to another document in the manual with:
:doc:`The Title </section/path/to/file>`
To link to a specific section in another document (or the same one), explicit labels are available:
.. _sample-label:
[section or image to reference]
Some text :ref:`Optional Title <sample-label>`
Linking to a title in the same file:
Titles are Targets
==================
Body text.
Implicit references, like `Titles are Targets`_
Linking to the outside world:
`Blender Website <https://www.blender.org>`__
Context Sensitive Manual Access
It is possible to link to a specific part of the manual from in Blender by opening the context menu (right click) of a property or operator and selecting Online Manual. In order for this to work, this needs to be accounted for in the documentation. To link a property or operator to a specific part of the manual you need to add an external reference link tag whose ID matches Blender’s RNA tag. The easiest way to find out what the tag for a property is to open the context menu of the property/operator and select Online Python Reference to extract the tag from the URL. Some examples of how this looks in the RST document are given below:
.. _bpy.types.FluidDomainSettings.use_fractions:
Fractional Obstacles
Enables finer resolution in fluid / obstacle regions (second order obstacles)...
.. _bpy.types.FluidDomainSettings.fractions_distance:
Obstacle Distance
Determines how far apart fluid and obstacles are...
For an operator:
.. _bpy.ops.curve.subdivide:
Subdivide
=========
Further Reading
To learn more about reStructuredText, see:
Good basic introduction.
Docutils reStructuredText Reference
Links to reference and user documentation.