|
weekly meeting at 18:30 UTC, Tuesday Set by moderator on 3 September 2008. |
|||
|
12:21
gaz joined
12:24
Wknight8111 joined
16:27
moritz joined
17:20
jhorwitz joined
17:30
cotto_work joined
17:38
pmichaud joined
|
|||
| Tene | early report magic ACTIVATE! | 18:06 | |
| [cardinal] | |||
| * Waiting on dmknopp's CLA to arrive. He mailed it a week ago | |||
| * Added some initial support for modules | |||
| * Added support for calling functions in another namespace. Only supports one level deep until I'm less lazy. | |||
| * Fixed some parsing bugs in cardinal. | |||
| * Added xlibtest.rb | |||
| [parrot] | |||
| * Removed the string param from exception handlers | |||
| * Implemented my ExceptionHandler proposal in the exceptionmagic branch (review plz) | |||
| KTHXBAI | |||
|
18:07
kj joined
18:18
NotFound joined
18:28
chromatic joined
|
|||
| cotto_work | HAI (ENOREPORT) | 18:30 | |
| moritz | hi | ||
| NotFound | H | 18:31 | |
| jhorwitz | OH HAI | ||
| Tene | I already reported. It' sup there^. | ||
| I need to leave now. | |||
| chromatic | Hello everyone. | ||
| Wknight8111 | hello | 18:32 | |
| kj | hello | ||
| jhorwitz | yo | ||
| pmichaud | hello. | ||
|
18:32
cotto_work joined
|
|||
| chromatic | Let's go in alphabetical order. | 18:33 | |
| I haven't done much, but I'm going to investigate any reproduceable crashes anyone wants, so send me mail or ticket numbers. | |||
| cotto_work? | |||
| cotto_work | no report | ||
| chromatic | japhb? | 18:34 | |
| jhorwitz? | 18:35 | ||
| jhorwitz | merged mod_parrot's hll-modules branch back into trunk. HLLs are now complete Apache modules (written in PIR or the HLL) that can manage their own directives, configurations and hook registration. converted the PIR layer and mod_perl6 to the new model. | ||
| taking notes on implementing a basic mod_cardinal and will post as a tutorial on writing a mod_parrot HLL layer. | |||
| getting ready for the next release of mod_parrot. i want it to be patch free, so i'm waiting on namespace interpolation in rakudo before i put out a new release. | |||
| EOR | |||
| chromatic | kj? | 18:36 | |
| kj | work on pirc: | ||
| * implement computation of local label offsets | |||
| * implement argument-direction check: "$S0 = print" is now illegal | |||
| * added a hashtable for string-reuse, and possibly to replace all linked lists | |||
| * cleanup and refactorings of code | |||
| * trying to understand Packfile stuff, probably only way to learn this stuff is trying. | |||
| .end | |||
| chromatic | moritz? | ||
| moritz | * Applied patches from Ontolog and masak | ||
| * trying to give fast feedback on patches | |||
| * I'd like to get a commit bit for Carl Mäsak. I applied quite a few Rakudo patches from him, they were all good. | |||
| .EOR | 18:37 | ||
| chromatic | NotFound? | ||
| NotFound | * Fixing some bugs | ||
| * Some morw work on Xlib and Mysql | 18:38 | ||
| * Added pirric, PIR Retro basIC, to examples | |||
| EOR | |||
| chromatic | pirric sounds like a victory to me. | ||
| particle? | |||
| particle | ~ added a few makefile targets for hll fakecutables | ||
| ~ questions, answers, general snarkiness | |||
| .end | |||
| chromatic | pmichaud? | 18:39 | |
| pmichaud | ** Rakudo spectest_regression: 169 files, 3676 passing tests (+16/+306 from last week!) | ||
| NotFound | Forgot to say: fixed loss of precission in complex sqrt | ||
| pmichaud | == Parrot stuff | ||
| : released 0.7.1 last Tuesday, no road bumps or difficulties | |||
| : spent some time yesterday writing up design notes for a | |||
| lexicals (re)implementation | |||
| -- current draft available at www.pmichaud.com/perl6/lexical.txt | |||
| -- new design is vastly simpler internally, but also more flexible | |||
| -- comments welcome | |||
| -- I may start a branch to try implementing it (but will | |||
| gladly hand off to someone else) | |||
| : awaiting comments or suggestions on RT #59006 (stringifying floats) | |||
| == Rakudo stuff | |||
| : fixed Str.perl method | |||
| : provided guidance to others on :multi and various interfaces for builtins | |||
| : last week I raised potential problem with container PMCs in aggregates | |||
| -- wrote a document outlining how Rakudo will store elements to | |||
| avoid building lots of container PMCs in aggregates | |||
| -- jonathan++ and I discussed on thursday, we think this approach will be | |||
| workable for quite a while | |||
| -- should also make it easy/easier to implement list assignment | 18:40 | ||
| -- design available at www.pmichaud.com/perl6/references.txt | |||
| -- will be cleaned up a bit and moved to languages/perl6/docs | |||
| == Plan for this week | |||
| : update container and value type handling in Rakudo according to new design | |||
| : may even get list assignment working | |||
| : experiment with new lexicals design and implementation | |||
| : work on symbol table exports in P6object, Rakudo, and PIR | |||
| : work on PCT and HLL mapping (time permitting) | |||
| : look at nested namespace issues in Rakudo (time permitting) | |||
| .end | |||
| chromatic | Wknight8111? | 18:41 | |
| Wknight8111 | * More work on /docs/book. | ||
| ** Added chapers for PCT, HLLs, and the standard PIR library (if we truely have such a thing) | |||
| ** Expanded testing chapter to including debugging as well | |||
| * More refactoring work on Parrot_PCCINVOKE and friends in the pdd27mmd branch | |||
| * Deleted pdd09gc branch, it got screwed up in a merge from trunk, and I haven't had time to focus on it. | |||
| EOR | |||
| chromatic | Okay, let's go to questions. | ||
| moritz suggested that Mäsak deserves a commit bit. Any objections? | |||
|
18:42
DietCoke joined
|
|||
| chromatic | moritz, are you willing to mentor him? | 18:42 | |
| NotFound | +1 | ||
| jhorwitz | +1 | ||
| pmichaud | +1 | ||
| moritz | chromatic: aye | ||
| Wknight8111 | +1 | ||
| pmichaud | if moritz doesn't, then I will :-) | ||
| chromatic | +1 here too. Have him send in the CLA and we'll make it so. | 18:43 | |
| moritz | $mentor = any(moritz, pmichaud) ;-) | ||
| DietCoke | ... your barewords are dangling. | ||
| moritz | DietCoke: they are types or subs ;) | ||
| pmichaud | $mentor = any <moritz pmichaud>; | ||
| particle | masak++ sent his cla in today | ||
| chromatic | Good name for a rock band: the dangling barewords | ||
| DietCoke, do you have a report? | |||
| DietCoke | Sure. | 18:44 | |
| Worked on partcl this week. Got tcltest.tcl parsing and running without errors. If I hack up [namespace import] I can probably start running the spec tests natively with no fudging. | 18:45 | ||
| Tcl is still slow and could use a pir-sub level profiler. | |||
| (I thought I could do that, but it's beyond my entry-level C skills.) | 18:46 | ||
| .eor | |||
| chromatic | Are there other questions? | ||
| Wknight8111 | A sub level profiler, would that make for a new runcore? | ||
| or are there better ways to implement such a beast? | |||
| chromatic | Wknight8111, almost undoubtedly. | 18:47 | |
| DietCoke | there's an RT discussion that question. | ||
| chromatic | It needs to track sub entry/exit, for start. | ||
| Wknight8111 will read up on it | |||
| pmichaud | I don't have a question so much as a repeat that feedback on lexicals proposal would be especially welcome. | ||
| DietCoke | My take was that it wasn't wasn't a runcore issue because it's not op based. | ||
| pmichaud | and getting stringification to work more reasonably for floats seems like it would be a quick Win for someone with some C programming skill | ||
| chromatic | pmichaud, if you can point me to a ticket, I may be able to implement that tonight. | 18:48 | |
| pmichaud | chromatic: for floats? #59006 | ||
| chromatic | Thanks. | ||
|
18:48
tewk joined
|
|||
| DietCoke | chromatic: I would love it if your solution helped me with my tcl_precision issue. I can talk to you offchannel. | 18:49 | |
| chromatic | Other questions? | ||
| tewk, did you report earlier? | |||
| tewk | NO report today, I just forgot to join | 18:51 | |
| pmichaud | afk | 18:53 | |
| particle | tewk: folks have had questions about using ncigen | ||
| notably, notfound has not been able to use ncigen for hooking xlib to parrot (and rakudo) | 18:54 | ||
| NotFound | I wasn't even ablr to try it. | ||
| tewk | NotFound: whats breaking, do you have something I can repeat and look at? | 18:56 | |
| DietCoke | I have a question. | ||
| chromatic | Go ahead. | 18:57 | |
| NotFound | tewk: Well, that is the question. There is something I can look to see how it must be used? | ||
| particle | tewk, notfound: continue on #parrot, please | ||
| NotFound | Ok | ||
| DietCoke | tcl has "namespace import" and "namespace export" - you have to declare what is exportable from a namespace in order to import it. But the export declaration is sticky: you can declare a command as exportable before you define it. Parrot doesn't support this model, does it? | 18:58 | |
| kj | gotta go. good night | ||
| DietCoke | kj: ~~ | ||
| (furthermore, you can un-import things.) | |||
|
18:59
cotto_work joined
|
|||
| DietCoke | (www.tcl.tk/man/tcl8.5/TclCmd/namespace.htm) | 18:59 | |
| so each namespace slot would have to keep track of where you imported it from, etc. | 19:00 | ||
| chromatic | Any other questions? | 19:01 | |
| DietCoke | (my question is: is my understanding right that parrot doesn't support that model as of now) | ||
| chromatic | Not natively no. | 19:02 | |
| You'd have to build some sort of thunk. | 19:03 | ||
| DietCoke | Ok - I already have my own .Sub override, I can add more slots on it to keep track of such things for now. Danke. | ||
| chromatic | Let's call it a week. The bug count is going down; good work everyone. | 19:04 | |
| moritz | thanks, bye | 19:05 | |
| Wknight8111 | later | ||
|
19:05
moritz left,
chromatic left
|
|||
| NotFound | Hasta la vista | 19:06 | |
|
19:06
NotFound left,
Wknight8111 left
19:07
jhorwitz left,
cotto_work left
19:21
DietCoke left
|
|||