On Sat, Nov 29, 2008 at 11:14:26PM +0100, Johan Vromans wrote: > Chip Salzenberg <chip@pobox.com> writes: > > Objections? > [...] > So my alternative approach would be to first sort this out, making it > possible for a user module to completely take over <> functionality. > Issues like error handling and safe inplace editing can then easily be > handled with appropriate modules. That's a clever thought. Adding -Meditsafe (just to pick a nice pragma) to a command line isn't as catchy as -j~ or -i'**~' but it does offer an easy upgrade path for backward compatibility: eval "use editsafe"; # error OK And you just know that Ingy will find a clever use for overloaded <>. > = What didn't work? IIRC there were issues with not being able to > override eof (eof, eof() and eof(FILE)). Also, overloading <> > only does scalar IO, since it ignores context. So if we had tied eof() processing and READLINE was responsive to context, it could work? -- Chip Salzenberg twitter:chipsalz "UTOPIA PLANITIA IS IN ORBIT is the new HAN SHOT FIRST" - Crisper Than ThouThread Previous | Thread Next