On Mon, Jun 10, 2002 at 10:37:24PM +0100, Nick Ing-Simmons wrote: > Nicholas Clark <[EMAIL PROTECTED]> writes: > > > >I like this best. To me, unadorned binmode means as raw as possible. > > Arrrrgl - :raw is an abomination of a name please don't use that term. > > "As raw as possible means > read(2), write(2) - no buffering so no buffer snooping,... we don't > want that normally. > > I think :crlf(+) and :crlf(-) or somesuch would have been so much nicer > if Camel-III had not legitimized :raw
Hmm. Yes, raw and cooked ttys also use the term raw to mean something. (partly LF -> CR/LF related) What stops us using :crlf(+) and :crlf(-) in preference to what the Camel currently says? (And issuing a patch (is that called a "plaster"?) for the Camel) The Camel doesn't say that they *don't* work. > >Can we write that > > > >open(FH,">:encoding(Xxxx):lf",$file) > > > >or something like that, with a second layer to say "no crlf" ? > > Except that should be :lf:encoding(Xxxx) - which can be > written :perlio:encoding(Xxxx) Ah. Right. I was trying to mean that the open should be saying in one go "I don't want CRLF translation, but I do need the specified character set translation" rather than have the open describe the character set translation and then use binmode to request the line endings stuff. Nicholas Clark -- Even better than the real thing: http://nms-cgi.sourceforge.net/
