<fig-group> not working as expected in Janeway

[pgoussy]

Describe the bug Our JATS vendor has tagged multiple <fig> images within <fig-group>, seeming following the JATS spec correctly, but Janeway is not rendering them as expected. In the PDF/manuscript versions of these figures, the first element that displays is a single "blanket" label/caption that describes a collection of images (e.g., "Figure 1"). Then, the sub-figures appear in sequence, each followed by its own dedicated sub-caption (e.g., "Figure 1a", "Figure 1b", etc.). However, in Janeway's HTML galley, there are a handful if issues with the way it renders:

• the "blanket" caption is accompanied by a broken-image icon, which implies that Janeway is looking for a dedicated image to associate with it (even though no such image exists)
• the "blanket" caption is not styled/boxed in a way that matches other figure captions
• next to the broken-image icon, there's a random link to the first sub-figure (which doesn't exist in the JATS)
• the sub-figure images appear all squished together, with no space in between them
• the sub-captions for these individual images do not display at all
• links to the figure within the article body do not seem to work

Janeway version 1.4.1

To Reproduce Steps to reproduce the behavior:

1. Go to https://journals.publishing.umich.edu/mp/plugins/typesetting/preview_galley/article/3110/galley/1038/#F2a
2. This link goes directly to Example 2a, the first sub-image in Example 2. Scroll up slightly to see the "blanket" caption for Example 2, and the broken-image icon and mysterious link.
3. Scroll up to the preceding paragraph and click the link for "Example 2"; note that this link doesn't function.

Expected behavior Main/"blanket" label and caption should display, followed by an image for each sub-figure, each of which should be followed by its own dedicated label and/or caption. All of these captions should be styled/boxed like any other figure caption.

Screenshots

Here's the JATS tagging used for this example:

<p id="P53">Yet the sincerity is just an act, as Tom is not the guarded, sincere protagonist of the song but a womanizer throughout the film. Before the performance, he dedicates it to &#x201C;someone special who just might be here tonight.&#x201D; The camera cuts between three women Tom has slept with&#x2014;bandmate Mary (Cristina Raines), Opal, L.A. Joan (Shelley Duvall)&#x2014;and Linnea (Lily Tomlin), a married woman Tom has doggedly pursued and is attempting to bed through this performance (<xref rid="F2" ref-type="fig">Example 2</xref>). In a later scene, after Tom and Linnea have slept together, Altman shows a close-up of a tape player that plays a recording of &#x201C;I&#x2019;m Easy&#x201D; as Linnea and Tom talk in bed. Once the song stops, emphasized with a shot of the tape player stopping, Linnea puts on her clothes and leaves (<xref rid="F3" ref-type="fig">Example 3</xref>). The implication is that Tom&#x2019;s song is a manufactured, literally mechanical, siren song, undercutting the musical affect of singer-songwriter authenticity. The left&#x2019;s music is no less hypocritical and manufactured than the right&#x2019;s.</p>
<fig-group id="F2" position="anchor">
<label>Example 2:</label>
<caption><p id="P167">(A) Tom Frank (Keith Carradine) sings &#x201C;I&#x2019;m Easy&#x201D; as (B) Mary (Cristina Raines), (C) Opal (Geraldine Chaplin), (D) L.A. Joan (Shelley Duvall), and (E) Linnea Reese (Lily Tomlin) watch.</p></caption>
<fig id="F2a">
<label>Example 2a</label>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="mp.3110-f0002a.jpg"/>
</fig>
<fig id="F2b">
<label>Example 2b</label>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="mp.3110-f0002b.jpg"/>
</fig>
<fig id="F2c">
<label>Example 2c</label>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="mp.3110-f0002c.jpg"/>
</fig>
<fig id="F2d">
<label>Example 2d</label>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="mp.3110-f0002d.jpg"/>
</fig>
<fig id="F2e">
<label>Example 2e</label>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="mp.3110-f0002e.jpg"/>
</fig>
</fig-group>

[pgoussy]

In retrospect, I realized that the missing sub-captions are unrelated to <fig-group>! They're just a manifestation of the same <label> bug reported here: #2657

When I replaced the sub-caption tags with tags, they started rendering in the HTML. In addition, the broken image icon no longer appeared before the main caption. However, the other issues with <fig-group> still need addressing: the unstyled main caption, the lack of spacing between each sub-figure (it's not clear at a glance whether the sub-caption is supposed to go with the image above it or below it), and the non-functional figure links in the article body. See screenshot:

[pgoussy]

Oh, and there isn't any space between the end of the and the start of the following

, which doesn't look great.