IRC Networks
Irc Logs Stats
Start date: 2007-09-27 02:48:27
Last update: 2008-10-24 20:19:38
Channels: 41
Logged Lines: 6230436
Size: 1834.74 MB
Powered by
Channel Info
Network: freenodeChannel: #html |
Search in www.irclog.org
Log from #html at freenode 2006-08-08
[20:48]<wrxfnn>Dorward: http://en.wikipedia.org/wiki/Percent-encoding
[20:49]<wrxfnn>according to it, & should be percent encoded. Not html encoded.
[20:49]<eyaexj>Sample URL: /?name=value&name=value
[20:49]<mjzwzzm>xahlee: If you have a & as data in a URL then it should be percent encoded. If you have an & as an argument seperator in a query string then it shouldn't be.
[20:50]<mjzwzzm>xahlee: And if you have a raw & in a URL and want to represent that in HTML then it needs to be HTML encoded.
[20:50]<eyaexj>If the amp is part of a value, it should be percent encoded. If it's separating name/value pairs, then it should be &.
[20:50]<eyaexj>That's what I meant earlier.
[20:50]<wrxfnn>riczho: ah i see.
[20:50]<wrxfnn>a bit confusing...
[20:53]<wrxfnn>what's the formal name for the argument in url?
[20:53]<wrxfnn>am trying to find a page about the & business in wikip
[20:54]<mjzwzzm>xahlee: Don't trust wikipedia, look to the specs.
[20:54]<rzujryvrn>How can I specify what mimetype should my docs be? I mean, I want to send XHTML as application/xhtml+xml, but now it is text/html
[20:54]<mjzwzzm>cruocitae: Depends on your webserver
[20:55]<mjzwzzm>xahlee: http://www.gbiv.com/protocols/uri/rfc/rfc3986.html#query
[20:56]<mjzwzzm>xahlee: No, CGI is a way for a webserver to talk to another program to get its data from that program instead of a flat file.
[20:58]<rzujryvrn>Dorward, Apache2.2
[20:59]<mjzwzzm>cruocitae: http://httpd.apache.org/docs/2.2/mod/mod_mime.html
[21:08]<wrxfnn>cruocitae: AddType application/xhtml+xml xhtml
[21:09]<wrxfnn>Dorward: the link you gave doesn't say about &
[21:09]<wrxfnn>the wikip page is here: http://en.wikipedia.org/wiki/Query_string
[21:15]<bzzyyrrgg>when you're using refresh/redirect with meta tags, is there anyway for it to set a target? For example: if the website we are wanting to redirect is in frames/frameset, for it to overwrite that coding?
[21:15]<mjzwzzm>Brriiaann: Don't, and no.
[21:15]<bzzyyrrgg>Don't?
[21:16]<mjzwzzm>Brriiaann: meta refresh is not recomended
[21:17]<bzzyyrrgg>Is there a way for me to redirect a user to a new page that goes to the main target?
[21:17]<bzzyyrrgg>so that it *
[21:18]<mjzwzzm>Brriiaann: No, redirects are current frame only. Avoiding frames in the first place would be a good idea.
[21:20]<bzzyyrrgg>I can't, they are automatically put into frames/frameset, which is something that I cannot control, unless there is a code that forbids frames/frameset, but still allows to show the page?
[21:22]<mjzwzzm>Brriiaann: <script type="text/javascript">if ( top != self ) { top.location = self.location; }</script> IIRC
[21:23]<bzzyyrrgg>Excellent.
[21:23]<bzzyyrrgg>that seems to have worked
[21:24]<bzzyyrrgg>Thank you.
[21:24]<bzzyyrrgg>Though, if they have javascript disabled... :X it render useless
[21:24]<bzzyyrrgg>will*
[21:27]<bzzyyrrgg>I guess this is better than nothing. Many thanks again :)
[21:33]<rzujryvrn>xahlee, Dorward, how can I send text/html to IE and application/xhtml+xml to moderner browsers?
[21:33]<wrxfnn>cruocitae: heh heh, you are in trouble.
[21:34]<wrxfnn>cruocitae: don't think there's a way
[21:34]<mjzwzzm>cruocitae: If you are going to do that then you aren't using anything XHTML adds - so don't. Browsers which support XHTML have rather better HTML parsers, so you just harm them by serving as application/xhtml+xml.
[21:35]<rzujryvrn>I saw on a website (don't remember which, but I got the link to it from here), that "experts send different mimetypes to different browsers"
[21:35]<wrxfnn>cruocitae: i think it'll come up if you search for IE, xhtml, mime
[21:35]<mjzwzzm>cruocitae: Replace "experts" with "fanboys" and you'd be right. A little knowledge is a dangerous thing...
[21:36]<gzznvxzdzds>cruocitae: n'uh-uh. Experts work out if XHTML can actually do anything for them, then realise it usually can't
[21:37]<gzznvxzdzds>"upgrading" to xhtml is like buying a fast car - sure you can show people how flashy you are, but it'll cost you a lot more than your reliable old motor
[21:38]<gzznvxzdzds>this tenuous analogy brought to you by GarethAdams
[21:38]<mjzwzzm>And it won't have the three seats in the back
[21:38]<rzujryvrn>dude, take a look at Renault Espace :>
[21:39]<gzznvxzdzds>you have Internet Explorer, not a Renault
[21:40]<rzujryvrn>lol, well-said!
[21:41]<rzujryvrn>how about PHP's "header"?
[21:43]<mjzwzzm>cruocitae: What about it?
[21:47]<rzujryvrn>it can be used to send different mimetypes, nope?
[21:48]<mjzwzzm>cruocitae: Yes - but why? There's no advantage to serving appendix c conformant XHTML as application/xhtml+xml, and its actively harmful in some cases (such as Firefox)
[22:13]<grgj2jv>;beware of xhtml @ cruocitae
[22:13]<mygy2jv>cruocitae: http://www.webdevout.net/articles/beware_of_xhtml.php
[22:21]<wdrrvs>Is there a way to get an image to be sent in mime-type application/octet-stream?
[22:21]<wdrrvs>so when ppl click on it, they'll get a download box instead of the image?
[22:21]<gzznvxzdzds>not through HTML
[22:22]<gzznvxzdzds>that's a server issue
[22:22]<wdrrvs>something like mod_rewrite?
[22:23]<gzznvxzdzds>there are many ways to do it
[22:23]<gzznvxzdzds>depending on exactly what you're trying to do, and how much you want it done
[22:23]<wdrrvs>lol
[22:23]<gzznvxzdzds>if you're using apache, #apache is a good place to ask
[22:23]<wdrrvs>thx
[22:36]<-- dvxn|syzzzyus xzs>http://www.bagdadsoftware.de")
[22:57]<crlgdrg>I have a question about URLs. I'm programming. Should I write URLs like this: http://thing.net/command?p1=v1&p2=v2&p3=v3 ... is that the only correct way to put parameters p1, p2 and p3 into the URL?
[22:57]<crlgdrg>or is that an old-fashioned way
[22:57]<crlgdrg>?
[22:58]<wg22gg>CapnDan: the very short answer is that's the normal way ... but the real answer is a bit more complex
[22:58]<crlgdrg>ugh :-)
[22:58]<wg22gg>CapnDan: it depends what context you're doing it in
[22:58]<crlgdrg>I notice that Oddmuse uses semicolons: ?action=edit;id=PageName
[22:59]<wg22gg>CapnDan, for example if that URL is going to be output on a page ... you may need to escape the ampersands (that's a HTML issue)
[22:59]<znysyj>CapnDan: http://thing.net/command/p1/v1/p2/v2/p3/v3/
[22:59]<znysyj>CapnDan: or better yet http://thing.net/command/ :p
[22:59]<crlgdrg>ugh!
[22:59]<wg22gg>CapnDan: the spec for what a URI (and hence a URL) is here: http://www.gbiv.com/protocols/uri/rfc/rfc3986.html
[23:00]<wg22gg>reisio: i think those aren't parameters strictu sensu ... but a representation of parameters in terms of a path
[23:00]<znysyj>doesn't really matter
[23:01]<wg22gg>"path segments" to use the rfc's language
[23:01]<wg22gg>reisio: well, it depends on what CapnDan is doing
[23:01]<wg22gg>if he's just trying to send a GET request to something accept normal query parameters
[23:01]<wg22gg>then it's no good sticking them in as path segment
[23:01]<znysyj>if it's something that's going to show up in an address bar, then all those queries are gonna be hard to read
[23:02]<znysyj>but of course, perhaps there's no way to simplify it
[23:02]<crlgdrg>I'm putting things on a page for people to click on.
[23:02]<crlgdrg>Not buttons, just <a href type links.
[23:03]<wg22gg>CapnDan: this is also handy: http://en.wikipedia.org/wiki/Query_string
[23:03]<wg22gg>mentions the W3C thinks CGI should treat ; same as & as a separator in the query string
[23:04]<crlgdrg>beautifully concise, thanks webben
[23:04]<wg22gg>CapnDan: ah ... then stick with & ... but remember to escape it as & within href
[23:05]<wg22gg>CapnDan: http://www.w3.org/TR/html4/appendix/notes.html#h-B.2.2
[23:14]<crlgdrg>ok, thanks!
[23:52]<sjfxrafns>hi all. Im having some issues. on the page: www.aces-limited.com/home the image: ACES_web_logo.gif displays fine in firefox, but does not display in ie6. any help?







