Re: Can/Does/Should the FeedValidator catch improperly escaped XHTML?
On Fri, 01 Sep 2006 01:19:35 +0200, James Holderness [EMAIL PROTECTED] wrote: Just because a feed appears to contain well-formed xhtml content today, that doesn't mean it's going to be well-formed tomorrow. Encouraging people to use xhtml when they don't know enough to have made that decision themselves is just asking for trouble. Sooner or later they're going to end up with broken xml which will be completely unreadable in many aggregators. Well, that completely depends on how we present this information to the user. I was not thinking of a simple message saying This is valid XHTML, change @type to 'xhtml' ASAP, you newb!, but more along the lines of informing the user that the content looks like wellformed XHTML, that he can read more about it here and there, why he should try to keep it wellformed and why he should not change it if he's not sure of its welformedness. Something like that. Also, escaped html tends to be better supported by aggregators anyway. That's today. I would hope and assume that this changes in the future. -- Asbjørn Ulsberg -=|=-http://virtuelvis.com/quark/ «He's a loathsome offensive brute, yet I can't look away»
Re: Can/Does/Should the FeedValidator catch improperly escaped XHTML?
That's today. I would hope and assume that this changes in the future.+1On 9/2/06, Asbjørn Ulsberg [EMAIL PROTECTED] wrote:On Fri, 01 Sep 2006 01:19:35 +0200, James Holderness [EMAIL PROTECTED] wrote: Just because a feed appears to contain well-formed xhtml content today, that doesn't mean it's going to be well-formed tomorrow. Encouraging people to use xhtml when they don't know enough to have made that decision themselves is just asking for trouble. Sooner or later they're going to end up with broken xml which will be completely unreadable in many aggregators.Well, that completely depends on how we present this information to theuser. I was not thinking of a simple message saying This is valid XHTML,change @type to 'xhtml' ASAP, you newb!, but more along the lines of informing the user that the content looks like wellformed XHTML, that hecan read more about it here and there, why he should try to keep itwellformed and why he should not change it if he's not sure of its welformedness.Something like that. Also, escaped html tends to be better supported by aggregators anyway.That's today. I would hope and assume that this changes in the future.--Asbjørn Ulsberg -=|=- http://virtuelvis.com/quark/«He's a loathsome offensive brute, yet I can't look away»-- /M:DM. David Peterson http://mdavid.name | http://www.oreillynet.com/pub/au/2354
Re: Can/Does/Should the FeedValidator catch improperly escaped XHTML?
M. David Peterson wrote: via http://www.oreillynet.com/xml/blog/2006/08/and_the_winner_of_the_best_ind_1.html#comment-75533 There's one other small problem though: they put XHTML as CDATA in html text constructs, while they're supposed to contain HTML 4. And since it's XHTML, they should embed it directly in xhtml constructs... Anthony brings out a good point http://www.oreillynet.com/xml/blog/2006/08/and_the_winner_of_the_best_ind_1.html#comment-75822 , Odd that the validator isn't saying anything about this. Should it, or is this an edge case that can be difficult, at best, to catch? At the moment, the HTML content is passed through the following: http://docs.python.org/lib/module-HTMLParser.html Note that this parser includes a handle_startendtag method, which is not a part of the HTML standard. Given the rather loose nature of HTML, this only tends to catch things like unmatched angle brackets and quotes. Also, there are a number of tools that attempt to produce well-formed XHTML, but don't do so consistently enough to drop the content into an Atom feed in such a manner. - Sam Ruby
Re: Can/Does/Should the FeedValidator catch improperly escaped XHTML?
Thanks for the response/info Sam!On 8/31/06, Sam Ruby [EMAIL PROTECTED] wrote: Also, there are a number of tools that attempt to produce well-formedXHTML, but don't do so consistently enough to drop the content into anAtom feed in such a manner.I assume this is from an XHTML standpoint (meaning the result isn't a one-to-one HTML XHTML ~equivalent) and not an well formed XML standpoint? Not that it really matters all that much as if the result isn't a solid representation of the authors original intent, then theres really no other alternative than to escape it and tag it as HTML. In short, unless the above is incorrect, then both Anthony and the FeedValidator are doing all that they both can and should be doing.Is this a fair statement?-- /M:DM. David Peterson http://mdavid.name | http://www.oreillynet.com/pub/au/2354
Re: Can/Does/Should the FeedValidator catch improperly escaped XHTML?
On Thu, 31 Aug 2006 13:50:56 +0200, Sam Ruby [EMAIL PROTECTED] wrote: Also, there are a number of tools that attempt to produce well-formed XHTML, but don't do so consistently enough to drop the content into an Atom feed in such a manner. But is it possible to check the wellformedness of the markup inside atom:content and suggest that the authors use 'type=xhtml' instead of 'type=html' if it is indeed wellformed X(HT)ML? It's a real stretch in terms of how far the validator should go, but I can definately see its usefulness. -- Asbjørn Ulsberg -=|=-http://virtuelvis.com/quark/ «He's a loathsome offensive brute, yet I can't look away»
Re: Can/Does/Should the FeedValidator catch improperly escaped XHTML?
+1 - Definitely would be useful as a suggestion.On 8/31/06, Asbjørn Ulsberg [EMAIL PROTECTED] wrote: On Thu, 31 Aug 2006 13:50:56 +0200, Sam Ruby [EMAIL PROTECTED]wrote: Also, there are a number of tools that attempt to produce well-formed XHTML, but don't do so consistently enough to drop the content into an Atom feed in such a manner. But is it possible to check the wellformedness of the markup insideatom:content and suggest that the authors use 'type=xhtml' instead of'type=html' if it is indeed wellformed X(HT)ML? It's a real stretch in terms of how far the validator should go, but I can definately see itsusefulness.--Asbjørn Ulsberg -=|=-http://virtuelvis.com/quark/«He's a loathsome offensive brute, yet I can't look away» -- /M:DM. David Petersonhttp://mdavid.name | http://www.oreillynet.com/pub/au/2354
Re: Can/Does/Should the FeedValidator catch improperly escaped XHTML?
Asbjørn Ulsberg wrote: But is it possible to check the wellformedness of the markup inside atom:content and suggest that the authors use 'type=xhtml' instead of 'type=html' if it is indeed wellformed X(HT)ML? It's a real stretch in terms of how far the validator should go, but I can definately see its usefulness. -1 Just because a feed appears to contain well-formed xhtml content today, that doesn't mean it's going to be well-formed tomorrow. Encouraging people to use xhtml when they don't know enough to have made that decision themselves is just asking for trouble. Sooner or later they're going to end up with broken xml which will be completely unreadable in many aggregators. Also, escaped html tends to be better supported by aggregators anyway. Regards James
Re: Can/Does/Should the FeedValidator catch improperly escaped XHTML?
* James Holderness [EMAIL PROTECTED] [2006-09-01 01:30]: Encouraging people to use xhtml when they don't know enough to have made that decision themselves is just asking for trouble. +1 Regards, -- Aristotle Pagaltzis // http://plasmasturm.org/
Re: Can/Does/Should the FeedValidator catch improperly escaped XHTML?
Both points well taken.I would like to change my initial vote to -1, though +1 for both James and Aristotle's follow-up.On 8/31/06, A. Pagaltzis [EMAIL PROTECTED] wrote:* James Holderness [EMAIL PROTECTED] [2006-09-01 01:30]: Encouraging people to use xhtml when they don't know enough to have made that decision themselves is just asking for trouble. +1Regards,--Aristotle Pagaltzis // http://plasmasturm.org/-- /M:DM. David Peterson http://mdavid.name | http://www.oreillynet.com/pub/au/2354
