|
donno20 has asked for the wisdom of the Perl Monks concerning the following question:
Hi all,
I am working a program to generate unicode file and then view by Notepad. All conversion work properly but I cannot make a newline.
I already tried out "\r", "\n", "\r\n", "\r\0", "\n\0", "\0\r", "\0\n", all of these not work.
Thanks for giving me some hint to solve the problem. ^_^ |
Re: New line in Unicode again
by halley (Prior) on Apr 21, 2003 at 12:15 UTC
|
Did you remember to use binmode() on your output file stream?
If you need to specify specific bytes for specific characters on specific platforms, don't rely on "\r" and "\n", which are semantic concepts. Try the octal route to call a spade a spade: "\015\012".
I thought Notepad relied on "\015\012" sequences for hard newlines, and used "\015\015\012" internally for word-wraps. This may have changed. Open up an existing file in a hex editor to glean the right sequence.
--
[ e d @ h a l l e y . c c ]
| [reply] |
Re: New line in Unicode again
by aquarium (Curate) on Apr 21, 2003 at 13:29 UTC
|
Unicode does not assign control characters in its standard, so whatever you do will not be portable. On Win32 the following will do what you want I think:
binmode OUTFILE;
print OUTFILE "\x0\xD\x0\xA";
Chris | [reply] |
|
|
The default encoding for Unicode on Windows is UTF-16LE, i.e. little endian, so it would actually be "\xD\0\xA\0". Strictly speaking, Unicode does assign control characters, but that still doesn't guarantee that the logical end-of-line will be the same between systems.
You should start your file with a byte order marker (BOM), which is the same as a zero-width no-break spaces. It is U+FEFF, which in UTF-16LE is "\xFF\xFE".
By the way, what you could have done is create a Unicode file in Notepad, then use Perl to look at the file and see what it has in it. Also note that Perl 5.8 has support for Unicode. See perlunicode and Encode::Unicode.
| [reply] |
|
|
>it would actually be "\xD\0\xA\0".
Correct! I was wondering why it was "a\0b\0c\0" paired instead of "\0a\0b\0c". But I don't have "\x".
>You should start your file with a byte order marker (BOM), which is the same as a zero-width no-break spaces. It is U+FEFF, which in UTF-16LE is "\xFF\xFE".
I don't know what actually the code is. But I did a little trick by reading the first two bytes of an existing unicode file into a buffer. It was work !.
>By the way, what you could have done is create a Unicode file in Notepad, then use Perl to look at the file and see what it has in it
The first two bytes cannot be read by print(), and hidden in notepad. How do I know it was "\xFF\xFE" ?
Cheers, ^_^
| [reply] |
Re: New line in Unicode again
by donno20 (Sexton) on Apr 22, 2003 at 05:12 UTC
|
YES !! I make it. Thanks monks. I have granted ++ votes to all of you.
I summed up my code:
binmode FILE;
$line = $unicode . "\xD\0\xA\0";
Cheers ^_^
| [reply]
[d/l] |
|
