Description
This is not so much a bug as something that it looks like you intend to change, but perhaps didn't know how. In moinmoin/modern/css/screen.css, for example, you have a warning (line ~126 or search for the string XXX) about non-ASCII characters. In particular, you use the right angle quote (HTML », raw: ยป) as a breadcrumb separator. You can replace this non-ASCII character with the Unicode escape sequence \00BB: {{{#pagetrail li:after {
- content: " \00BB "; color: black;
} }}} The only real benefit to doing this is that users editing themes in text editors ignorant of beyond-ASCII characters will not have to worry about accidentally clobbering the special characters. If you need other character codes, you might try something like this, which I maintain for my own personal use, but something more generic may be more bookmarkable.
Steps to reproduce
N/A
Example
N/A
Details
MoinMoin Version |
1.3.5 |
OS and Version |
Gentoo Linux, 2.6.13-ck7 |
Python Version |
2.4.2 |
Server Setup |
Apache 2.0.55 |
Server Details |
No fancy mod_python or anything, just CGI |
Workaround
The description above provides all of the information you should need if you'd like to make this change to your own wiki install.
Discussion
Its true that using unicode escape make it less fragile, but on the other hand it also make this less readable. I don't want to remember the spacial code of each character when I can simply use sane editor that support utf-8 and can simply have the character in the file in a readable way.
If you don't care that your users may be using a UTF-8 incapable editor (you do realize that such editors are fairly common, right?), why do you bother warning them? And there's only one non-ASCII character in there, anyway, so it's not like you'd be forced to memorize the number for every Unicode glyph. ;)
Users should not touch this file, it may be used by themes authors, and its reasonable that they will use decent editor.
- In case this hasn't been completely tossed out the window already, you might also consider explicitly specifying the unusual charset in the CSS file. To me, this would be sufficient warning to make sure that I'm not clobbering multi-byte characters and such. Just a line near the top like this would do:
@charset "UTF-8";
We may improve the encoding notification, which should go actually to all files - all the css files use utf-8 encoding, like all moin output files.
Plan
- Priority:
- Assigned to:
- Status: no bug, closed.