<html><head></head><body><div class="ydpc1333a3ayahoo-style-wrap" style="font-family:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:16px;"><div></div>
<div><span style="color: rgb(38, 40, 42); font-size: 13px;">On Tuesday 17 September 2024 at 18:27:41 BST, Werner LEMBERG via mpeg-otspec <mpeg-otspec@lists.aau.at> wrote:</span><br></div><div id="ydpc1333a3ayahoo_quoted_7300416510" class="ydpc1333a3ayahoo_quoted"><div style="font-family:'Helvetica Neue', Helvetica, Arial, sans-serif;font-size:13px;color:#26282a;">
<div><br></div>
<div>> >> I have not yet seen a single XML document processed with the<br></div><div><div dir="ltr">> >> standard XML tools to create a PDF file that really looks good from<br clear="none">> >> a typographical point of view.<br clear="none">> > <br clear="none">> > That's possibly because the good ones generally don't mention how<br clear="none">> > they were produced.<br clear="none"><br clear="none">> This might be indeed the case.<br clear="none"><br clear="none">> > E.g. a good proportion of best-selling fiction is done from XML<br clear="none">> > using CSS for print and AntennaHouse Formatter or PDFReactor or<br clear="none">> > RenderX or PrinceXML, as are a good many scientific and technical<br clear="none">> > journal articles.<br clear="none"><br clear="none">> Well, the question how much manual editing was necessary after<br clear="none">> converting the input XML...<br clear="none"><br clear="none">> > Note that since Word and Open/Libre Office both use XML, the<br clear="none">> > document is ALREADY being produced from XML.<br clear="none"><br clear="none">> I was imprecise, sorry. There is a big difference between using XML<br clear="none">> as a document storage format, with zillions of proprietary and/or<br clear="none">> program-specific extensions, and using XML for document interchange.<br clear="none">> I meant the latter.<div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987"><br></div><div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987">RenderX is proprietary, I think; but I think it was considered one of the best (or only?), 2 decades ago, since I last looked. Apache FOP was just beginning 20 years ago but I remember it was quite okay - FOP stands for "formatting object processor" and from the look of it looks quite like description of pdf's data structure in XML syntax (and a special name space). FO (the namespace) is probably a w3c standard now, and FOP should be somewhat better in the last 20 years?</div><div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987"><br></div><div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987">Granted ,</div><div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987"><br></div><div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987">XML (some general namespace) -> (XSLT) -> FO -> (FOP / RenderX) -> PDF</div><div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987"><br></div><div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987">Is going to depend a bit on how clever the XSLT (specific to this particular xml format) is, and how clever FOP / RenderX does things. When I looked at FOP 20 years ago, it couldn't do automatic hyphenations (LaTeX style, sorry I am a LaTeX fan...) and spacing algorithms etc were a bit lacking, but it must have improved by now.</div><div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987"><br></div><div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987">Although, strictly speaking, automatic hyphenation LaTeX babel style is probably a bit against the XML philosophy of preserving the content....</div><div class="ydpc1333a3ayqt5858027926" id="ydpc1333a3ayqtfd15987"><br clear="none"></div></div></div>
</div>
</div></div></body></html>