Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2017-05-20

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

All times shown according to UTC.

Time Nick Message
00:01 ronsavage joined #marpa
00:10 idiosyncrat CPANtesters looks OK -- that same "dirty dozen" test on Andreas's platforms fail -- I'm unable to find out why.  The failures don't seem to evidence elsewhere, so I'm currently assuming that Andreas is testing some bleeding-edge modules.
00:12 idiosyncrat Also kind of annoying -- there are 2 sources of test results -- the tail log and the main CPANtesters site.
00:13 idiosyncrat The tail log updates quickly, but only tracks the 1000 most recent results -- currently about 25 minutes of results.
00:13 idiosyncrat The main site updates more slowly, but has *everything*.
00:14 idiosyncrat Lately the main site is very slow -- I just upload Marpa-R3-4.001_046, and the Marpa-R3-4.001_045 (from 8 days ago) are still not available.
00:14 ronsavage JK: Re Marpa-R3-4.001_046. All tests pass here.
00:15 idiosyncrat ronsavage: Thanks!
00:16 idiosyncrat ronsavage: When I do the next "new feature" release there'll be major backward incompatibilities in the recce calls.
00:16 idiosyncrat $recce->read(\$string) will still work as previously in its 1 argument form.
00:17 idiosyncrat But the 3-argument form will no longer exist, and $recce->resume() will take no arguments and behave quite differently.
00:18 idiosyncrat My old I/O interface was idiosyncratic, and allowing multiple read()'s breaks it anyway, so I'm taking the opportunity to make a fairly large break.
00:34 ronsavage JK: Hmmm. That sure sounds like it's going to affect some of my code, but by now I'm not surprized :-).
00:38 idiosyncrat I hope it will be an improvement -- and I hope I'll get your advice in the matter.
00:39 idiosyncrat The new $recce->resume() will be easier to use in a loop -- you will be able to write event loops this way:
00:39 idiosyncrat [ Initialize loop ]
00:40 idiosyncrat while ($recce->resume()) { ... do stuff }
00:41 idiosyncrat In other words, cleaner event loops, and more like those in other interfaces.
01:20 ronsavage I just ran 'ag' (like ack like grep but vastly faster) and I have 7 modules which call resume().
01:23 idiosyncrat Is that above or below fears and/or expectations
01:23 idiosyncrat ?
01:25 ronsavage Dems da facts.
01:28 idiosyncrat I looked at my test suite -- 26 uses of resume()
01:29 idiosyncrat plus 15 problematic uses of $recce->read() -- uses of the soon-to-be-removed use of the 2- or 3-argument form.
01:30 idiosyncrat On the other hand, there are 90 (hopefully) unproblematic uses of $recce->read().
01:30 ronsavage Found a typo in the Marpa::R3::Changes doc. 'little or now usage' => 'little or no usage'.
01:32 ronsavage And another one: 'return result' => (I think) 'return a result'.
01:36 idiosyncrat ronsavage: Fixed in commit 66f0674
01:49 ilbot3 joined #marpa
01:49 Topic for #marpa is now Start here: http://savage.net.au/Marpa.html - Code paste/run: https://f.perlbot.pl/#marpa - Jeffrey's Marpa site: http://jeffreykegler.github.io/Marpa-web-site/ - IRC log: http://irclog.perlgeek.de/marpa/today - Youtube channel: https://www.youtube.com/channel/UCYKVfGBtfTqbs1JdYq-dc5g
07:24 sirdancealot joined #marpa
07:50 ronsavage joined #marpa
16:06 kaare_ joined #marpa
16:40 kaare__ joined #marpa
21:31 idiosyncrat joined #marpa

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