If:
- You can't find reasonable documentation backing up the feature in perl 5
- The feature doesn't work in 5.005-style threaded perl
- The coding technique makes you think "Boy, that's really clever" (always a good warning in itself)
odds are you'll have problems with it in perl 6. Not guarantees, but good warning signs. | [reply] |
You could always install 6 on a testbed machine and check perldoc. I'll probably move up just because I like my stuff being up to date.
| [reply] |
Odds are I'd just go re-check the source, if I didn't remember off-hand as it was. (Assuming, of course, that I wasn't just emitting parrot assembly anyway) Joys of being deeply involved. :)
| [reply] |
What you should keep in mind depends a great deal on how much the translator is able to translate current idioms into Perl 6 idioms. On your dynamically scoped $_ issue, it's not clear whether the translator will be smart enough to see that you've invoked length in a context where the
$_ will certainly be undefined. If it is that smart, it could certainly install an argument of the form:
sub len ($_ = $CALLER::_) { length }
which has much the same effect as if $_ were dynamically scoped. | [reply]
[d/l]
[select] |
At the same time, don't worry too much about the compatibility issue, the designers of Perl 6 DO have compatibility in their mind all the time.
If you don't believe me, believe this: Apocalypse 1. | [reply] |
| [reply] |