Strip HTML tags and character references from field and truncate

Liberty Create - Build Advice
1

Hullo,

I’d like to truncate a Text, multi-line formatted field, so I can display a snippet of the field in a list of search results.

I started by using the substring function in a composite, but unfortunately it’s not quite cutting the mustard, as the HTML tags from the formatted text (like <br/>) are returned as plaintext in the output.

image

So I’ve set up a rule to copy the value of the Text, multi-line formatted field to a plain Text, multi-line field, using the Strip HTML tags field processor, with the idea of then using this field in the composite.

image
image909×236 9.66 KB

But, although it’s correctly stripping out the HTML tags, it’s not stripping out the HTML character encodings (like &nbsp; for non-breaking spaces and &#39; for apostrophes)

image

Is there a straightforward way to take multi-line formatted text and the strip out the formatting? I could chain together a sequence of rules and composites to strip out the &nbsp;, and then another to strip out the &#39;… But this way lies madness! Surely there’s a field processor that could take care of the whole thing?

2

The “Strip HTML tags” processor is a very simple line of Javascript which you can find in the Code Studio area under Processors > Field Processors.

If you have a little JS knowledge you could duplicate it and create a version that strips the tags and html entities in one processor.

3

Assuming you just want to remove them rather than replace them with anything, then something like this (untested)…

return params.input ? String(params.input).replace(/<[^>]+>/g, '').replace(/&[a-zA-Z]+;/g, '').replace(/&#[0-9]{2,4};/g, '') : params.input;

This should remove both named (e.g. &) and numbered (e.g. &) entities.

4

Exactly what I was hoping for. Thanks Bob - I’ll give it a try!