[from Monkey Patching Thread]1.8 to 1.9 compatibility

[Robert Dober]

As I said in the former thread, I have given some thoughts on this.
But I am not a guru and I would like many more opinions on this.

90% of the possible things are quite trivial or easy, like using
enumerator, defining methods like String#to_bytes or Integer#pred.
I do not really know where the main difficulties are, Fibers cannot be
done easily and I assume that a Thread based fiber implementation
might be ultra slow.

Cheers
Robert

 

[Marc Heiler]

As I said in the former thread, I have given some thoughts on this.
But I am not a guru and I would like many more opinions on this.

90% of the possible things are quite trivial or easy, like using
enumerator, defining methods like String#to_bytes or Integer#pred.
I do not really know where the main difficulties are, Fibers cannot be
done easily and I assume that a Thread based fiber implementation
might be ultra slow.

Cheers
Robert

Maybe someone can do a summary... I myself am struggling with it,
especially the encodings. This gives me headache, for .rb files which
worked flawlessly with ruby 1.8.x now give me US ASCII errors.

 

[Dave Thomas]

Maybe someone can do a summary... I myself am struggling with it,
especially the encodings. This gives me headache, for .rb files which
worked flawlessly with ruby 1.8.x now give me US ASCII errors

If a file contains any characters with the top bit set, you now have =20
to tell Ruby what encoding that file is in. For example, if you =20
include a utf-8 em-dash (=97), you'd need to add the comment

# encoding: utf-8

puts "No way=97Way!"


to the top of the file.

Each file in an application can have its own encoding.


Dave