|
Sun 12 Feb 2006 09:29:23 PM UTC, comment #16:
No, CSS is not mandatory. It is only style, it has nothing with contents (well, in the perfect world anyway.) Ideally, a page must be perfectly comprehendable without any CSS at all, and Savane, BTW, does a pretty good job in this respect currently.
That said, I'm not trying to push strongly the duplication of "visibility: hidden". I personally think it's a good thing to have, but it's not a strong opinion.
So, let's consider this thread closed then.
|
Sun 12 Feb 2006 04:02:37 PM UTC, comment #15:
I forgot to mention:
the bottom line is that style should be in CSS, not in XHTML. Putting it in both just makes one of them useless and duplicating code with the exact same purpose just makes things harder to apprehend whenever we tried to understand how it works.
|
Sun 12 Feb 2006 04:01:10 PM UTC, comment #14:
Anyway, if a browser print a page before taking into account all its CSS first, I'd say it's a bug in the browser that make erroneous assumptions (for instance, assuming that parts of the style are not mandatory).
|
Sun 12 Feb 2006 03:36:20 PM UTC, comment #13:
The reason to duplicate it in HTML though it is present in CSS is that the style sheet is a separate file, which is loaded separately from the page. In principle, there could be some delay when the page is shown, but the style sheet is not yet applied. In this case this can hardly ever happen, though, since feedbacks are not shown on pages you see when first visit the site.
|
Sun 12 Feb 2006 03:27:57 PM UTC, comment #12:
...
because it did not seem necessary (already in the CSS; maybe your CSS is out of date, you may regenerate it by typing "make" in the top dir of the trunk, and make sure your browser is not using an outdated CSS in his cache by reloading the page with CTRL-R)
|
Sun 12 Feb 2006 03:26:31 PM UTC, comment #11:
Ok, thanks for this new patch, it works nicely now with both Konqueror and Firefox.
I just skipped the addition of "style="visibility: hidden;"
|
Sat 11 Feb 2006 02:01:05 PM UTC, comment #10:
Ping. Does the patch work?
|
Mon 06 Feb 2006 08:38:50 PM UTC, comment #9:
Here is a patch against the branch. I also think that `Show feedback again' is less clumsy than `Get back feedback' (which is probably not even correct, I think it should have been `Get feedback back'.)
|
Sun 05 Feb 2006 10:38:46 PM UTC, comment #8:
The branch is named dev-2006.02-FormattingText
(see http://svn.gna.org/viewcvs/savane/branches/)
> Yes, it is more or less installed now, but I can never log in (it accepts the
> password, redirects me to the account page, but prints that I'm not logged in
Do you use a fully qualified domain name on the server? For some reason, it causes login problems when it is not the case.
|
Sun 05 Feb 2006 10:15:40 PM UTC, comment #7:
Yes, it is more or less installed now, but I can never log in (it accepts the password, redirects me to the account page, but prints that I'm not logged in there.) I can try getting the branch, what was its name again?
|
Sun 05 Feb 2006 09:46:46 PM UTC, comment #6:
But as I said, I improved it a bit and it does not work :(
I havent tested the original version with firefox. To do tests, did you installed savane on a test machine? If so, you may want to get the current development branch where the current thing has been commited.
|
Sun 05 Feb 2006 09:39:06 PM UTC, comment #5:
Weird. I use Firefox myself (1.0.4 from Debian Sarge) and tested the script using it. Now also checked with Konqueror, both work.
|
Sun 05 Feb 2006 06:07:00 PM UTC, comment #4:
I applied your patch and enhanced it so a small box "get back feedback" shows up, in case someone wants it back.
I work nicely... but suddenly I gave it a try with mozilla firefox and nothing happen.
|
Sun 05 Feb 2006 05:36:43 PM UTC, comment #3:
It is not really necessary. But if Gna! uses JavaScript at least somewhat often, I'd suggest that you add it to the header unconditionally, whether there is a script on the page or not. I guess browsers just default to JavaScript, but I think some additional 100 bytes per page is not that high a price.
|
Sun 05 Feb 2006 05:21:28 PM UTC, comment #2:
Hello,
In your first mail about this, you mention a line to add in the html head. How strong is this requirement? Without, it works and the xhtml validator does not complain (but there are other xhtml mistakes on this page; stuff from the html 4.0 time, so it is hard to exactly be sure it is okey).
|
Sun 05 Feb 2006 02:48:46 PM UTC, comment #1:
Thanks for your contribution.
|
Sun 05 Feb 2006 01:49:21 PM UTC, original submission:
This patch makes feedback messages go away when a user clicks on it. It uses JavaScript, but provides additional features, so for users with JavaScript disabled nothing will change. I have tested that Lynx doesn't mind the script on a page i.e. it doesn't cause wrong page rendering or some other side-effect.
|
Related Recipes:
Savane - Bugs: bug #5202, a simple patch for hiding feedback...