You are add line
<META HTTP-EQUIV="Content-Type" CONTENT="text/html;
charset=ISO-8859-1">
into generated html files. In my comments for commits I
use national charachters (it in cp1251) and when You
include it comments into html with defined charset
ISO-8859-1 it shows in browser incorrectly.
Logged In: YES
user_id=584620
The charset declaration is mandated by the HTML spec.
Removing it would invalidate the HTML. If it was removed, the
local chars would appear correct in *your* browser, but not in
*mine*, because yours assumes a default charset of cp1251
and mine assumes windows-1252. This wouldn't solve the
problem.
A possible solution might be to add a -charset parameter to let
the user specify the charset of the log file. StatCvs could then
translate the log into something like UTF-8 encoded Unicode,
or output the HTML using the specified encoding.
But one point is valid: The assumption that most logs are
ISO-8859-1 encoded might be wrong. I need more input on
this one.
The hassles of dealing with pre-XML tools :-(
Logged In: YES
user_id=584620
A recent experience: Log messages committed from the Mac OS X
command line and written with vim will default to UTF-8. This means
CVS logs are not unlikely to contain several different encodings, with no
means of telling which is which.
This makes me think that there's not much we can do here. The -charset
switch would be nice for advanced users, though.
Logged In: YES
user_id=584620
I'm moving this to the Feature Request tracker with a new summary.
The parameter should be called "-encoding", not "-charset".