Directives

likec4-view

Embeds one view as an interactive iframe. Dynamic views (sequences) share the same id space and embed identically.

.. likec4-view:: cloud-to-amazon
   :height: 420px
   :title: Upload flow

MyST form:

```{likec4-view} cloud-to-amazon
:height: 420px
```

Options: height (default 460px), title (iframe title, for accessibility), render (see Static images), and mode — diagram (viewer default) or sequence, which opens a dynamic view directly in its sequence rendering (appends the viewer’s ?dynamic= parameter):

.. likec4-view:: cloud-to-amazon
   :mode: sequence

An unknown view id fails the build (works with -W).

Static images

render picks how the view is embedded: iframe (HTML default), png, jpg, or text. It is a preference, not a demand — the same source builds for every output format:

.. likec4-view:: cloud
   :render: png
   :width: 80%
   :alt: Cloud, structural view

  • On non-HTML builders (latex/PDF, epub, …) views are images by default; :render: iframe there falls back to the builder’s default.

  • On builders that can’t embed images at all (text, man, linkcheck) everything is plain text and no image export runs.

  • :render: jpg needs "jpg" somewhere in likec4_render (only PNG is exported unless asked).

  • In image mode width, height, alt, align, scale pass through to the image with the standard docutils validators; alt defaults to title.

  • :mode: sequence works in image mode too: dynamic views are exported a second time in sequence layout (--seq), and the directive picks that rendering. On non-dynamic views the option is ignored, as in the viewer.

likec4-model

Embeds the whole viewer gallery, or just links it:

.. likec4-model::
   :height: 600px

.. likec4-model::
   :link-only: