<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p>Hey Karl,</p>
    <p>Do you have a module and entry key value I can use for testing?</p>
    <p>Troy</p>
    <p><br>
    </p>
    <div class="moz-cite-prefix">On 4/18/20 2:10 PM, Karl Kleinpaste
      wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:062f1d61-46f5-bb1a-a669-5f18e1456373@kleinpaste.org">
      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
      <div class="moz-cite-prefix">On 4/18/20 2:48 PM, Greg Hellings
        wrote:<br>
      </div>
      <blockquote type="cite"
cite="mid:CAHxvOVJz-OLqpW5iGSfgp_t6XbtOrcxE+kR7NgFXDS1FBSd3yw@mail.gmail.com">
        <div dir="auto">This is the HTML header. Not the HTTP header.</div>
      </blockquote>
      <br>
      <font face="FreeSerif">I had thought that's what the </font><tt>&lt;meta
        http-equiv=\"content-type\" content=\"application/xhtml+xml;
        charset=utf-8\"&gt;</tt><font face="FreeSerif"> was supposed to
        give me, for the case of a directly-loaded text blob that didn't
        arrive via HTTP. Apparently not.<br>
        <br>
      </font>
      <blockquote type="cite"
cite="mid:CAHxvOVJz-OLqpW5iGSfgp_t6XbtOrcxE+kR7NgFXDS1FBSd3yw@mail.gmail.com">
        <div dir="auto">For manually loading the text you don't have an
          HTTP header, but you still tell the widget what content type
          when you call wk_html_open_stream or
          webkit_web_view_load_string or similar functions. It's this
          value that is telling Webkit what to do.</div>
      </blockquote>
      <br>
      &lt;headdesk&gt;<br>
      <br>
      It hadn't occurred to me that this was done other than in the
      header directives as I began to fill the widget.<br>
      <br>
      Unfortunately, still no joy. There are 6 occurrences of
      "text/html" in 3 files that do webkit widget loading:<br>
      src/editor/editor.c<br>
      src/editor/webkit_editor.c<br>
      src/webkit/wk-html.c<br>
      And the one that is immediately relevant is of course the last,
      which has 1 of the 6. Changing them all from "text/html" to
      "application/xhtml+xml" had precisely zero effect. The
      self-closing &lt;div&gt; still makes my .introMaterial bleed
      through to the end of the chapter.<br>
      <br>
      I can't win. Sorry. I think I'm done with this exercise in
      madness.<br>
      <br>
      For the record, this is the patch that restores
      &lt;div/&gt;-non-destructive behavior, with which I've been
      testing all these variants. The &lt;meta&gt; is gone, all the
      other &lt;?xml&gt; and DOCTYPE are gone, all that's added is the
      namespace spec, plus the code change proper that simply pastes the
      engine's returned content into the widget, rather than engaging
      the &lt;div/&gt; wiping.<br>
      <br>
      Ohwell. I tried.<br>
      <br>
      <div dir="auto"><tt>--- src/main/display.cc    2020-04-18
          16:39:59.743635867 -0400</tt><br>
      </div>
      <tt>+++ ../databits/display.cc.FAIL    2020-04-18
        16:44:35.607215035 -0400<br>
        @@ -89,7 +89,7 @@<br>
             N_("&lt;br/&gt;&lt;br/&gt;&lt;center&gt;&lt;i&gt;This
        module has no content at this point.&lt;/i&gt;&lt;/center&gt;");<br>
         <br>
         #define HTML_START \<br>
        -    "&lt;html&gt;&lt;head&gt;&lt;meta
        http-equiv=\"content-type\" content=\"text/html;
        charset=utf-8\"&gt; \<br>
        +    "&lt;html xmlns=\<a class="moz-txt-link-rfc2396E"
          href="http://www.w3.org/1999/xhtml\" moz-do-not-send="true">"http://www.w3.org/1999/xhtml\"</a>&gt;&lt;head&gt;
        \<br>
         &lt;style type=\"text/css\"&gt;&lt;!-- \<br>
         A { text-decoration:none } \<br>
         *[dir=rtl] { text-align: right; } \<br>
        @@ -1181,6 +1181,12 @@<br>
                 key-&gt;setChapter(i * chapter);<br>
                 key-&gt;setVerse(0);<br>
         <br>
        +#if 1<br>
        +        buf = g_strdup_printf("%s&lt;br /&gt;",<br>
        +                      (strongs_or_morph<br>
        +                       ?
        block_render(imodule.renderText().c_str())<br>
        +                       : imodule.renderText().c_str()));<br>
        +#else<br>
                 // begin gosh this is gross.<br>
                 // hunt down self-closing &lt;div ... /&gt; and stomp
        them with spaces.<br>
                 // this is disgustingly special-case, caused by
        self-closing &lt;div&gt;<br>
        @@ -1217,6 +1223,7 @@<br>
                 g_string_free(divBuf, TRUE);<br>
         <br>
             // end grossness<br>
        +#endif /* 0 */<br>
         <br>
                 swbuf.append(buf);<br>
                 g_free(buf);<br>
      </tt><br>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <pre class="moz-quote-pre" wrap="">_______________________________________________
sword-devel mailing list: <a class="moz-txt-link-abbreviated" href="mailto:sword-devel@crosswire.org">sword-devel@crosswire.org</a>
<a class="moz-txt-link-freetext" href="http://www.crosswire.org/mailman/listinfo/sword-devel">http://www.crosswire.org/mailman/listinfo/sword-devel</a>
Instructions to unsubscribe/change your settings at above page</pre>
    </blockquote>
  </body>
</html>