[Koha-devel] Re: accented characters in templates

[Ambrose LI]

Hi,

In article <200311102219.hAAMJRM07452 at alma.athenscounty.lib.oh.us>
you write:

>Forgive my ignorance of these multi-lingual matters: If I'm
>making HTML corrections to a non-English template, what do I
>need to do to ensure that accented or non-english characters
>are re-saved correctly?

If your editor does not touch the encoding at all (e.g., plain
non-multilingual vi) and does not word-wrap automatically, the
accented characters should be safe.

However, if your editor somehow interprets the characters (e.g.,
if it uses Unicode internally, and then re-encodes the file
when you save it -- probably all the new graphical editors),
you'll need to ensure that the editor uses the correct character
set. If there is a "meta" line with a charset in the HTML, that
would be the correct character set you should use; otherwise,
there might be an included file with the "meta" tag. If you
can't find the "meta" tag, then it's iso-8859-1 (not utf-8).

I find that doing a diff between the old and new versions
is helpful in such cases. The diff should only show what is
corrected; if anything else shows up, then something is wrong.
If you have the "wdiff" command installed, "wdiff" output is
even more useful as an additional check, though it's much less
readable than "diff" output.

-- 
Ambrose LI Cheuk-Wing  <a.c.li at ieee.org>

http://ada.dhs.org/~acli/