Hi,
Followup to this.
Filename can be fixed by attached script. Can you include this in
example. Also, I have few scripts I used to fix encoding isues.
Osamu
#!/bin/sh
# pst_fix_filename script to fix file name encoding by osamu <at> debian dot org
#
for x in *;
do
# For Japanse pst file on Windows 95/98/Me
# convert file name from shift-jis to UTF-8 (tested)
mv "$x" `echo "$x" | iconv -f shift-jis -t utf-8`
## For Chinese pst file on Windows 95/98/Me (osamu's guess)
## convert file name from GB18030 (GBK code point) to UTF-8
## mv "$x" `echo "$x" | iconv -f gbk -t utf-8`
## For Western European pst file on Windows 95/98/Me (osamu's guess)
## convert file name from iso-8859-1 to UTF-8
## mv "$x" `echo "$x" | iconv -f iso-8859-1 -t utf-8`
##
## Use probably BIG5 for Taiwan, UHC for Korea, ISO-8859-2 for Polish ...
##
done
for x in * ; do
echo convert $x
#
# Here is content
#
# I use nkf here but iconv can be used too. I make UTF-8 and iso-2022-jp
version.
# I found content-type is needed for UTF-8 when importing to MacBook mail
program
# ISO-2022-JP seems to be quite robust and does not need content-type (I dd it
anyway)
#
# For mutt, UTF-8 works better but needs to fix title which was OK initially in
pst.
# Need some idea but not needed now. (Procmail body filtering?)
#
# This will convert mail box content to UTF-8 and
# -S for shift-jis input
# -w for UTF-8
nkf -S -w <$x | \
sed -e 's/[Cc]ontent-[Tt]ype:.*text\/plain.*$/Content-Type: text\/plain;
charset=UTF-8/' \
-e 's/[Cc]ontent-[Tt]ype:.*text\/html.*$/Content-Type: text\/html;
charset=UTF-8/' \
>$x.utf8x.mbox
# -S for shift-jis input
# -k for ISO-2022-JP
nkf -S -j <$x | \
sed -e 's/[Cc]ontent-[Tt]ype:.*text\/plain.*$/Content-Type: text\/plain;
charset=iso-2022-jp/' \
-e 's/[Cc]ontent-[Tt]ype:.*text\/html.*$/Content-Type: text\/html;
charset=iso-2022-jp/' \
>$x.2022.mbox
#
# This is example snip script for Wesetrn european windows pst
# (Example of using "iconv")
#
# iconv -f ISO-8859-1 -t UTF-8 $x | ....
done
