Paul de Vrieze posted <[EMAIL PROTECTED]>, excerpted below, on Tue, 22 Nov 2005 11:20:28 +0100:
> On Monday 21 November 2005 12:04, Duncan wrote: >> * Set the base tag. I sometimes save web pages for my own use, and >> like them to work when I do. Adding a <base href= ...> tag would be >> very useful, here. Without it, saving just the HTML to disk breaks the >> page rather drastically, because it can't find the CSS and images, as >> they are relative links. Good to have the links relative; good the >> formatting is separated from the content to the degree the page breaks >> without the CSS; bad that there's no base href tag to "unbreak" things >> when the html page gets viewed on its own. > > As the pages are generated from xml, this might be hard to do. Also > isn't this something that should be done by the downloading program. I > know that IE does (used to do) this. Wget can do something similar. Hmm... I don't remember seeing that in IE back around the 4/5/5.5 era, and as I was a public beta tester and a regular in the public beta groups for those (and in line at midnite for '98, MS screwed up for me when they incorporated the anti-privacy stuff into eXPrivacy... because while I didn't have such a problem with it per se, I knew where it was headed, once that sort of stuff was blessed by the kingpin -- to the era of obiquitous spyware, and rootkits arriving on your CDs from "trusted" compannies... so I went from in line at midnite for 98 to refusing to cross the line MS was demanding I cross for eXPrivacy, and I have them to thank for forcing me to finally defect from the land of proprietaryware, to the land of freedom in software! =8^), I (like to think) I knew more about what was in them than even many geeks. I run Konqueror/KHTML, and obviously it isn't setting it, either. However, I do run thru privoxy, and may be able to whip up a filter to add it... I'll have to think a bit... I've not gotten into wget, much more than being aware it's often used in scripted applications including many I rely on here. Thanks for the useful hint, however -- I'll keep it filed away as it seems likely to be found to be useful at some point! > In any case isn't the idea of downloading a page that you also download > the images and stylesheets, etc. belonging to it. (firefox can also do > this). Well, depending on one's purpose... As I mentioned, I run privoxy. A personal preference is light text on dark background, and one of my biggest uses of privoxy is a set of filters designed to enforce that (FWIW, I turned them off and reloaded the site, for purposes of this thread). It follows that one of the primary reasons I may be using view-source and save-as, is to check the code and create or update the related privoxy filter expressions. For such purposes, I'll start with the html page alone. Sometimes I need the css as well, sometimes that only complicates things. However, if there's a base tag it's simple enough to comment it out if I need to, while it's a bit more work to figure out what that base should be manually, and add it, if necessary. Another purpose for page-only would be if one wants to modify it somewhat for local use, but that use is going to be with a net connection, and due to content/format-rull separation with css, there's no need to have the format rules locally. Under normal circumstances I surf with images off, so those aren't an issue, altho if scripting is required on a page (and it's somewhere I feel reasonably secure turning it on) and the scripts are separate files as well, those can sometimes be required. Konqueror does of course have the "everything" aka "web archive" saving feature as well, but I seldom use it. The only time in recent memory I've used it was when I saved the world reaction to 9/11 pages, something I go back and think about, every once in awhile. One brief moment, but of course it didn't last, nor could it have. -- Duncan - List replies preferred. No HTML msgs. "Every nonfree program has a lord, a master -- and if you use the program, he is your master." Richard Stallman in http://www.linuxdevcenter.com/pub/a/linux/2004/12/22/rms_interview.html -- gentoo-dev@gentoo.org mailing list
