Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2015-07-02

| Channels | #marpa index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:48 ilbot3 joined #marpa
01:48 Topic for #marpa is now Start here: http://savage.net.au/Marpa.html - Pastebin: http://scsys.co.uk:8002/marpa - Jeffrey's Marpa site: http://jeffreykegler.github.io/Marpa-web-site/ - IRC log: http://irclog.perlgeek.de/marpa/today
03:12 jeffreykegler joined #marpa
06:26 ronsavage joined #marpa
09:00 lwa joined #marpa
10:33 ronsavage joined #marpa
13:56 Cheery I'm facing a kind of problem with how I'm traversing the parse tree
13:58 Cheery the problem is that I'm traversing the parse tree in pre+post order.
13:58 Cheery "for" symbol "in" statement block
13:59 Cheery the symbol is traversed first, then statement, then block.
13:59 Cheery so by the time postorder pass for the 'for' -clause starts, the block has been built, the statement has been compiled in the current environment
14:00 Cheery I'm getting a chance to bind the variable only after the block has been compiled
14:01 Cheery well. I can hack around here..
14:01 Cheery but then..
14:01 Cheery statement "for" symbol "in" statement
14:02 Cheery this is doing nearly the same thing, but requires completely different code, because traversing order matters
14:03 Cheery so either I somehow specify the traversing order when it's different..
14:04 Cheery or I'm changing how my code gets generated.
14:05 Cheery I'm steering towards changing how my code gets generated.. because I keep feeling it keeps the thing easier to reason about.
14:06 Cheery at the same time it makes some more challenges. particularly late binding.
14:09 Cheery I'll end up with local variables that need to be "bound" if they appear.
14:10 Cheery I also potentially end up with block break & block continue -links.
14:15 Cheery as well as speculative iteration stop -instructions.
14:17 Cheery but traversal stays simple, and evaluation order isn't bound to traversal order.
14:23 jeffreykegler joined #marpa
14:45 Cheery the functional model would avoid the pre-order functions entirely.
14:49 Cheery but it'd be filled with 'fill later' -slots.
15:12 sivoais joined #marpa
18:21 ernimril joined #marpa
18:27 lwa joined #marpa
21:50 jeffreykegler joined #marpa
22:28 ronsavage joined #marpa
23:26 jeffreykegler joined #marpa

| Channels | #marpa index | Today | | Search | Google Search | Plain-Text | summary