reZonator user manual is built using Sphinx Documentation Builder.

The virtual environment shared between different project parts, e.g., calc and help, so it's better to make it on the project top level.

On Ubuntu patchelf package is required:

sudo apt-get install patchelf

Linux, macOS:

python3 -m venv .venv
source .venv/bin/activate

Windows (there is no python3 command on Windows so be sure Python 3 is in the PATH):

python -m venv .venv
.venv\Scripts\activate

For building help we only need

python -m pip install sphinx

Linux, macOS:

./make.sh

Windows:

make.bat

Target documentation is in ../out directory.

Main font size in the Assistant looks a bit small and ugly by default. It depends on the platform, so it's better to use the Assistant's View menu to adjust font size in runtime rather then settings it bigger in CSS files. Leave the body font-size undefined in styles. As for font family, the Assistant seems can't work with arbitrary fonts, its zoom command starts glitching and displayed font size can be even different for different runs. So probably it worth to stay the font face by default as well.

As the documentation contains a notable amount of physical formulas, they should be inserted as pictures. Qt Assistant uses QTextDocument to display help pages, and it has fewer abilities comparing to real browsers. Also, Assistant can't run JavaScript code. So pictures for formulas is the only possibility to display them.

Use the render_formula.html page as a simple formula editor to play with the TeX syntax. Here we have a short handbook on TeX symbols. Once formula is finished in the editor and proved to be a valid TeX, it can be pasted into a RST document after the tex tag followed by respective image tag, e.g.:

Some text some text some text

  .. tex:
    \cfrac{1}{F} = (n - 1)\bigg[\, 
        \cfrac{1}{R_1} - 
        \cfrac{1}{R_2} + 
        \cfrac{(n -1)T}{n R_1 R_2}
    \bigg]\,

  .. image:: img/calc_lens_f.png

Some text some text some text

tex is not a standard tag so sphinx just treats it as a comment. There is a Node.js script that reads these tags and uses well known MathJax library for rendering TeX syntax into SVG format. Then the script converts each SVG formula into PNG image and saves it into a file specified in the image tag.

Note that the indentation in the above code snippet is important. The image tag has to be indented to be a separate paragraph otherwise the image will be inserted inline. Since the image is indented, the tex should be indented as well otherwise the image becomes a part of comment and will be ignored.

Usage examples:

# Download MaxJax and additional libraries
npm install

# Generate PNG images for a single file
node render_formulas.js calc_lens.rst

# Update all formula images in all files
node render_formulas.js --overwrite *

# Run against several files and show formulas (no PNG generated)
node render_formulas.js --dry-run calc_lens.rst calc_grin.rst

There also a number of ODF files that are LibreOffice Math formulas. They should be considered as deprecated and gradually replaced. Reasons are Math uses its own syntax instead of conventional TeX, and its rendered formulas do not have such nice appearance as those rendered by MathJax.

For screenshots: #FF494F for buttons and arrows, #FF7C7F for window frames, e.g.

See the full syntax in the sphinx docs.

The first level for chapters lister in index.rst, see schema.rst, elements.rst:

******
Schema
******

The second level is for individual topic files e.g. params_window.rst:

Gaussian Beam Calculator
========================

The third level is for subtopics inside of topic files:

Calculation Algorithm
---------------------

The fourth and fifth levels are for most inner subtopics:

Change waist
^^^^^^^^^^^^

Lock waist
~~~~~~~~~~
When locking waist then bla-bla-bla...

Lock front
~~~~~~~~~~
When locking front then bla-bla-bla...

.. index:: single: global parameter

Set an anchor in the text you can navigate to:

.. _pump_mode_vector:

Ray Vector 
----------

section text bla-bla-bla...

this creates an id in the target html:

<div class="section" id="ray-vector">
<span id="pump-mode-vector"></span><h2>Ray Vector</h2>

It's seen that the section title also creates an anchor: Ray Vector --> id="ray-vector".

Then it can be used in code to navigate to the help topic, something like:

QString PumpParamsDialog::helpTopic() const
{
    return "pump_mode.html#" + _params->modeName().toLower();
}

Manually named reference to an anchor:

:ref:`Waist <pump_mode_waist>`

Automatically named reference to a doc:

:doc:`input_beam`

Manually named reference to a doc:

:doc:`Input beam<input_beam>`

External link

This is the Lua `original documentation <https://www.lua.org/manual/5.3/>`_ for the reference.