[jhowell]

Originally Posted by DNSB

I suspect that this will block the automatic plugin update.

As far as I know the automatic plugin update is based on the information collected periodically by a separate program that also updates the Index of calibre plugins website. That gets its information from the "Version History:" spoiler in the first post of each plugin thread, which must be formatted exactly the right way. I suspect that the last update to the that post broke the formatting since it sees that latest version as "3.2.7 Released 17 May, 2020".

EDIT: It looks like it was changed from "Version History:" to "Version History".

Edit again: I was wrong it isn't the colon. It looks like the first plugin post is parsed looking for a specific sequence of HTML elements that should result if the proper BBCODE is used to format the Version History section.

Spoiler:

    def _read_version_history_html(self, forum_link):
        br = browser()
        br.set_handle_gzip(True)
        try:
            raw = br.open_novisit(forum_link).read()
            if not raw:
                return None
        except:
            traceback.print_exc()
            return None
        raw = raw.decode('utf-8', errors='replace')
        root = html.fromstring(raw)
        spoiler_nodes = root.xpath('//div[@class="smallfont" and strong="Spoiler"]')
        for spoiler_node in spoiler_nodes:
            try:
                if spoiler_node.getprevious() is None:
                    # This is a spoiler node that has been indented using[INDENT]
                    # Need to go up to parent div, then previous node to get header
                    heading_node = spoiler_node.getparent().getprevious()
                else:
                    # This is a spoiler node after a BR tag from the heading
                    heading_node = spoiler_node.getprevious().getprevious()
                if heading_node is None:
                    continue
                if heading_node.text_content().lower().find('version history') != -1:
                    div_node = spoiler_node.xpath('div')[0]
                    text = html.tostring(div_node, method='html', encoding='unicode')
                    return re.sub(r'<div\s.*?>', '<div>', text)
            except:
                if DEBUG:
                    prints('======= MobileRead Parse Error =======')
                    traceback.print_exc()
                    prints(html.tostring(spoiler_node))
        return None