obp-image-manipulation

This script simplifies elements of html files exported from InDesign.

With our present configuration, InDesign outputs images and captions like so:

<div>
    <div id="important-id">
        <img />
    </div>
</div>
<p class="caption"></p>

Where #important-id describes important rules to display the image.

Whereas this works fine, our workflow can benefit from a more streamlined structure to embed tools (i.e. zoomify) more efficiently. This would be:

<figure>
    <img id="important-id" />
    <p class="caption"></p>
</figure>

Installation

Install the system packages. On Debian:

$ apt-get install python3.5-venv python3-pip

You might want to install third-party python libraries and run the script in a virtual environment. Create the environment first:

$ cd your-work-folder
$ pyvenv-3.5 .venv
$ source .venv/bin/activate

Then install the dependencies:

(.venv) $ pip3 -r requirements.txt 

Run the script

Simply:

(.venv) $ python3 main.py input-file.xhtml output-file.xhtml

If needed, get tips by:

(.venv) $ python3 main.py -h

Troubleshooting

Character encoding

The current version of the script outputs utf-8 files. If required a different encoding (i.e. utf-16), please change this in the last part of main.py, when beautifulsoup encodes the soup prior writing the output file.