by two mouse clicks.
---------------------------------------------------
About
---------------------------------------------------
After upgrading my Linux box to Suse 9.2, I had a
big problem: In many text(-based) files there didn't
show up any non-ascii characters (e.g. umlauts: äöüß
any more. Instead there were ugly boxes or even
nothing at all.
It took me quite some googling until I found the
reason and solution: Suse switched their default
encoding to utf-8, so that files from my old system
as well as some new files I encounter from time to
time being encoded in latin-1 don't show up correctly
any more.
The solution is to convert every single of these files
to utf-8 using the command line tool "iconv", which
is part of the c library and therefore should be
present on any Linux system.
To ease things up, I created this KDE service menu which
allows me to convert files with just two mouse clicks
and even works on multiple selected files. So, every
time I encounter another non-utf-8 file, I can continue
working with a corrected file in only a matter of
seconds :-).
As there seem to be many other users with this kind
of problem out there, I decided to share it to you.
So enjoy and drop me a note (kde@benno-dielmann.de)
, if you like it (or if there are problems...)!
---------------------------------------------------
Installation
---------------------------------------------------
Copy the file
iconv-servicemenu.desktop
to
$KDEDIR/share/apps/konqueror/servicemenus/
and you are done!
---------------------------------------------------
Usage
---------------------------------------------------
Select the file(s) to convert and select "convert to
utf-8" from the actions submenu of the right-mouse-
button menu. A new (converted) file with suffix _utf-8
is created in the same directory. The original file is
left untouched, in case anything went wrong ;-).
Ratings & Comments
2 Comments
How do you detect the encoding of the textfile before converting for utf-8? Are you using the encoding detection of Konqueror (http://webcvs.kde.org/kdelibs/khtml/misc/) or Mozilla (http://www.mozilla.org/projects/intl/chardet.html)?
;-) In fact, there is no encoding detection at all. Its just a simple wrapper for the command iconv -f latin-1 -t utf_8 %u -o %u_utf-8 where %u is the selected file. Nothing more. If you have to convert other encodings than latin-1, I suggest you to replace this in the iconv-servicemenu.desktop. If you know of any method to add automatic detection please let me know. The examples you provide all seem to involve programming exceeding the scope of such a simple service menu entry. Correct me if I'm wrong... For me personally, there isn't any need for several different input encodings since I use it mainly for converting my old files from my backups (See the lengthy description above...), on which latin-1 works quite well. If there is anyone out there needing detection of different input encodings, please let me know, so I can think about a way to add it.