| ugexe | i say you make a SO post about it asking why its not documented | 00:07 | |
| Zoffix | ? | ||
| The policy? | |||
| ugexe | the code that can demonstrate the security issue | 00:08 | |
| Zoffix is confused | |||
| ugexe | it was a snark | ||
| Zoffix | .tell MasterDuke something about '-'.IO.slurp.... No idea. Don't have time ATM to look. File a ticket, I'd think | 00:09 | |
| yoleaux | Zoffix: I'll pass your message to MasterDuke. | ||
| ugexe | and not towards you | ||
| Zoffix | Well, I took some sleeping pills a bit ago 'cause I was planning on calling it a day. Not in condition to fix this bug. Gonna take a stab at it tomorrow morning. | 00:20 | |
| And I guess lacking a security policy, I'll just file a `[SECURITY]` tagged ticket once I have a working fix and lizmat++ can mention it in the weekly. | 00:21 | ||
| but we really should setup a proper channel to report these issues and have some proceedures in place or something (/me never dealt with these kinds of problems before) | 00:22 | ||
| Zoffix spots a pattern in buggy behaviour | 00:24 | ||
| and if it's true, then the bug is much less severe | |||
| Zoffix & | |||
| samcv | gotta write up my grant progress report tonight | 01:10 | |
| Zoffix | k, found the "bug" lol | 01:34 | |
| .mkdir: $chmod sets chmod with $chmod modified by umask | 01:35 | ||
| But didn't know that, so I assumed .mkdir: $chmod was setting a busted mode on dirs :) | |||
| I guess this is really just a doc fix :) | 01:37 | ||
| ZofBot: that de-escalated quickly! | 01:38 | ||
| ZofBot | Zoffix, } The "is native" trait is the main gateway used to access C libraries | ||
| Zoffix returns to vacationing :D | |||
| Geth | rakudo/nom: 90d129de8e | (Zoffix Znet)++ | src/core/IO/Handle.pm Fix IO::Handle.open for '-' paths Fixes RT#131503: rt.perl.org/Ticket/Display.html?id=131503 The observed bug is due to .open failing to set requested encoding. We don't actually .open a handle in the case when '-' is a path, so fix by calling .encoding to set the wanted encoding on the handle. |
02:40 | |
| roast: 60a7a88e47 | (Zoffix Znet)++ | 2 files Test IO::Handle.open with '-' paths sets encoding RT#131503: rt.perl.org/Ticket/Display.html?id=131503 Rakudo fix: github.com/rakudo/rakudo/commit/90d129de8e7ae62 |
02:42 | ||
| rakudo/nom: 7344a3d273 | (Zoffix Znet)++ | src/core/IO/Special.pm Fix IO::Special.perl We don't use Mu's .new, so its .perl is no good |
03:46 | ||
| roast: 99c37a96c8 | (Zoffix Znet)++ | S32-io/io-special.t Test IO::Special.perl Rakudo fix: github.com/rakudo/rakudo/commit/73...d27302c345 |
|||
| rakudo/nom: 3755c0e7c4 | (Zoffix Znet)++ | src/core/IO/Handle.pm Generalize logic for IO::Handle.open with path '-'.IO - Do not assume $*IN/$*OUT will contain opened handles, open them if closed - If $*IN/$*OUT are set to unopened handle that already has '-'.IO as path, then we don't wanna infinilop, so we create a SPESHUL IO handle that uses STDIN/STDOUT and return that. |
03:48 | ||
| roast: 7e4138ea79 | (Zoffix Znet)++ | S32-io/open.t Test IO::Handle.open on '-'.IO path handles… …works on any type of handle being in $*IN/$*OUT Rakudo fix: github.com/rakudo/rakudo/commit/37...e7c44b770e |
03:49 | ||
| [Tux] | This is Rakudo version 2017.05-352-g3755c0e7c built on MoarVM version 2017.05-25-g62bc54e9 | 07:29 | |
| csv-ip5xs 3.091 | |||
| test 13.727 | |||
| test-t 4.651 - 4.673 | |||
| csv-parser 13.994 | |||
| Alexdaniel | Zoffix: oh, thanks for opening that ticket | 09:57 | |
| there was something else I forgot to file… let's see… | |||
| a-ha, there it is | 10:04 | ||
| RT #131507 | |||
| timotimo | moritz: my browser scrolls sideways on perl6book.com because the diagram is so huge | 13:19 | |
| moritz | timotimo: I'm replacing the complete frontpage (but I'm slow to write proper CSS :/ ) | 13:20 | |
| timotimo: the new version will contain only a half-sized image inline | |||
| Geth | roast: 4d254198d5 | (Zoffix Znet)++ (committed using GitHub Web editor) | S32-io/open.t Fix typo in description |
13:22 | |
| timotimo | that's probably fair | 13:27 | |
| i bet there are more distinct-looking butterflies than there are camels | 13:28 | ||
| moritz | perl6book.com/new/ new version of the site, no CSS yet | 13:30 | |
| Geth | roast: 65a0dc5db3 | (Zoffix Znet)++ | S02-names-vars/variables-and-packages.t Remove fudge for bogus tests Closes RT#122356: rt.perl.org/Ticket/Display.html?id=122346 The behaviour of autovivifying a variable when post incrementing is infinitely more desirable than returning a type object on the first incrementation. |
13:33 | |
| mr_ron | moritz: perl6book.com/ (index page) s!<h1 class="splash-head">Perl 6 Fundamenatls</h1>!<h1 class="splash-head">Perl 6 Fundamentals</h1>! | 13:50 | |
| moritz | mr_ron: thanks | 13:58 | |
| Geth | nqp/master: 16 commits pushed by pmurias++ review: github.com/perl6/nqp/compare/c4af4...0ea634c2c3 |
14:28 | |
| nine | Is clone() meant to be overrideable? | 15:17 | |
| dogbert17 | commit: 2015.06 say ~$<string><part> if 'abc-abc-abc' ~~ / $<string>=( [ $<part>=[abc] ]* % '-' ) / | 15:35 | |
| committable6 | dogbert17, ¦2015.06: «abc abc abc» | ||
| dogbert17 | commit: 2014.06 say ~$<string><part> if 'abc-abc-abc' ~~ / $<string>=( [ $<part>=[abc] ]* % '-' ) / | 15:36 | |
| committable6 | dogbert17, ¦2014.06: «abc abc abc» | ||
| dogbert17 | m: my $rx2 = m:i/a/ | 16:22 | |
| camelia | Use of uninitialized value of type Any in string context. Methods .^name, .perl, .gist, or .say can be used to stringify it to something meaningful. in block <unit> at <tmp> line 1 |
||
| dogbert17 | commit: 2015.11 my $rx2 = m:i/a/ | 16:23 | |
| committable6 | dogbert17, ¦2015.11: «Method 'match' not found for invocant of class 'Any' in block <unit> at /tmp/66Qv1DBCfi:1 «exit code = 1»» | ||
| Geth | rakudo/nom: 36bc410222 | (Zoffix Znet)++ | src/core/Promise.pm Fix lost dynamics in Promise.then When we cue up a .then we forget to set the $*PROMISE dynvar, so lookups of dynamics inside the .then code can't find 'em. Fixes RT#131509: rt.perl.org/Ticket/Display.html?id=131509 |
17:32 | |
| roast: e5d144ee09 | (Zoffix Znet)++ | S17-promise/then.t Test dynamic vars work in Promise.then Rakudo fix: github.com/rakudo/rakudo/commit/36bc4102225a324 |
17:33 | ||
| ggoebel | pmurias: saw the recent commits regarding building js backend... what is the full set of steps now to build it from a fresh git clone github.com/rakudo/rakudo.git? | 20:22 | |
| .tell pmurias saw the recent commits regarding building js backend... what is the full set of steps now to build it from a fresh git clone github.com/rakudo/rakudo.git? | 20:38 | ||
| yoleaux | ggoebel: I'll pass your message to pmurias. | ||
| Zoffix | hm, I thought Perl 5's $. got reset when next file was being read, say with `-n` switch, but it doesn't | 21:10 | |
| I guess there' | |||
| I guess there's no point in making NL module then | |||
| geekosaur | there is because that one's in perlfaq | 21:26 | |
| (how to get $. reset behavior, that is) | |||
| Zoffix | BenGoldberg: I gave you access to most of perl6 repos. You can accept the invite at github.com/perl6 | 21:59 | |
| You know... if you want to fix typos :} | |||
| BenGoldberg | Zoffix++ | 22:00 | |
| Now all I have to do is figure out this git thing ;) | |||
| AlexDaniel | BenGoldberg: actually, small things you can do directly on github | 22:01 | |
| just click edit, edit it, and save ???? | 22:02 |