798 2004-06-15 09:48:30 +0000 Error instead of warning for ampersand as data in XHTML 2006-11-08 01:14:37 +0000 1 1 1 Unclassified Validator Parser HEAD All All RESOLVED FIXED http://www1.ndr.de/ndr/ndr_pages_idx/0,2571,SPM880,00.html P2 normal --- 1 joern.hofer link bjoern www-validator-cvs oldest_to_newest 2015 0 joern.hofer 2004-06-15 09:48:30 +0000 Hi, i got a page with a link and title-Tag like this: <li><a href="/ndr/ndr_pages_idx/0,2571,SPM908,00.html" title="KONZERTE & EVENTS"><img src="/container/ndr_style_images_gen_default/0,2718,65996,00.jpg" height="19" alt="KONZERTE & EVENTS" /></a></li> and the Validtor still says, it is valid XHTML 1.0 Strict, even the amersand is used as Data. If i got a page with another Error, the Validtor correctly says: Ampersand used as Data. If i don't have any Error, except the Ampersand, he says, it ist Valid. Validating with the CSE HTML Validator Pro and the http://www.validome.org/: Both got an error with the Amerpsand (and they are right, i think). I think this must be a Bug with the Validator. The URL above is a dynamic-page, so it could be, that in 1 hour there isn't any ampersand, so i paste in here some source around of the error: <div id="leftNav" class="submenu"><ul> <li><a href="/ndr/ndr_pages_idx/0,2571,SPM902,00.html" title="KINO-TIPPS"><img src="/container/ndr_style_images_gen_default/0,2718,52182,00.jpg" height="19" alt="KINO-TIPPS" /></a></li> <li><a href="/ndr/ndr_pages_idx/0,2571,SPM1476,00.html" title="FILMFESTIVALS"><img src="/container/ndr_style_images_gen_default/0,2718,105874,00.jpg" height="19" alt="FILMFESTIVALS" /></a></li> <li><a href="/ndr/ndr_pages_idx/0,2571,SPM904,00.html" title="FILME IM FERNSEHEN"><img src="/container/ndr_style_images_gen_default/0,2718,52190,00.jpg" height="19" alt="FILME IM FERNSEHEN" /></a></li> <li><a href="/ndr/ndr_pages_idx/0,2571,SPM908,00.html" title="KONZERTE & EVENTS"><img src="/container/ndr_style_images_gen_default/0,2718,65996,00.jpg" height="19" alt="KONZERTE & EVENTS" /></a></li> <li><a href="/ndr/ndr_pages_idx/0,2571,SPM910,00.html" title="CD-TIPPS"><img src="/container/ndr_style_images_gen_default/0,2718,66000,00.jpg" height="19" alt="CD-TIPPS" /></a></li> <li><a href="/ndr/ndr_pages_idx/0,2571,SPM912,00.html" title="QUIZ & SPIELE"><img src="/container/ndr_style_images_gen_default/0,2718,66004,00.jpg" height="19" alt="QUIZ & SPIELE" /></a></li> <li><a href="/ndr/ndr_pages_idx/0,2571,SPM1102,00.html" title="LEUTE"><img src="/container/ndr_style_images_gen_default/0,2718,86754,00.jpg" height="19" alt="LEUTE" /></a></li> </ul></div> Am i right or am i stupid? ;-) 2016 1 ville.skytta 2004-06-15 13:11:30 +0000 I am pretty sure that you are right, and that this is why there is the "Note: The Validator XML support has some limitations." link as the very first thing in the XHTML validation results page... See http://openjade.sourceforge.net/doc/xml.htm 2025 2 bjoern 2004-06-17 13:35:51 +0000 You would get a warning, see http://www.websitedev.de/markup/validator/tests/ampersand-as-data.html,validate Even with Validome.org all you get is a warning. 4225 3 link 2005-02-05 04:28:47 +0000 This was a symptom of some warnings not getting displayed when the page was otherwise valid; the warning about the bare amperstand was always there, it was just being supressed. This issue has been resolved in current development code. 12683 4 ot 2006-10-31 06:39:46 +0000 checking a bunch of bugs today, I noticed that S:P:O seems to not give this "ampersand as data" warning, whereas openSP does. compare: http://qa-dev.w3.org/wmvs/0.7.3/check?uri=http%3A%2F%2Fwww1.ndr.de%2Fndr%2Fndr_pages_idx%2F0%2C2571%2CSPM880%2C00.html and http://qa-dev.w3.org/wmvs/HEAD/check?uri=http%3A%2F%2Fwww1.ndr.de%2Fndr%2Fndr_pages_idx%2F0%2C2571%2CSPM880%2C00.html Bjoern, is that a bug in S:P:O? Should I be sending it your way? 12687 5 bjoern 2006-10-31 07:44:47 +0000 I think I fixed this bug in SGML::Parser::OpenSP CVS in so far as it is the cause of this issue. A workaround in check would be to push() 'xml' twice into the options array, only the last item in ->warnings(...) will be ignored, as far as I could tell. 12846 6 ot 2006-11-07 06:22:10 +0000 Bjoern, (In reply to comment #5) > I think I fixed this bug in SGML::Parser::OpenSP CVS in so far as it is the > cause of this issue. Have you tested it successfully? I updated my version (on laptop computer) of s.p.o to the version in CVS late last week, and saw no change in behavior, at least from the point of view of check?uri=http%3A%2F%2Fwww.websitedev.de%2Fmarkup%2Fvalidator%2Ftests%2Fampersand-as-data.html > A workaround in check would be to push() 'xml' twice into > the options array, only the last item in ->warnings(...) will be ignored, as > far as I could tell. I did that in check, and unless I misunderstood your suggestion, it is not working, either. http://qa-dev.w3.org/wmvs/HEAD/check?uri=http%3A%2F%2Fwww.websitedev.de%2Fmarkup%2Fvalidator%2Ftests%2Fampersand-as-data.html 12850 7 bjoern 2006-11-07 11:20:42 +0000 Yes, I tested it and I've now also added a new test case t/22mwarnings.t to CVS that tests for this particular case twice. The test succeeds on my platform. If the Validator does not handle it properly, it might be that the branch where you push the option is never executed due to some XHTML detection issues, as we discussed on IRC. If t/22mwarnings.t succeeds on your platform, I don't see how this can be a problem in SGML::Parser::OpenSP. 12851 8 bjoern 2006-11-07 11:40:33 +0000 I think the current code just lacks $opensp->warnings(@spopt); 12857 9 ot 2006-11-08 01:14:37 +0000 (In reply to comment #8) > I think the current code just lacks > > $opensp->warnings(@spopt); Confirmed, and closed. Pushing to CVS in a moment. Thanks Bjoern.