|
01:36
tokuhirom joined
01:44
vendethiel joined
02:54
colomon joined
03:29
diakopter joined
03:46
tokuhirom joined
06:31
FROGGS joined
06:35
Ven joined
07:47
zakharyas joined
08:03
Ven joined
08:37
Ven joined
09:47
Ven joined
09:57
Ven joined
|
|||
| dalek | arVM: 41d75f1 | jnthn++ | src/strings/utf8.c: Fix indento. |
10:50 | |
| arVM: f09c782 | jnthn++ | src/strings/ (3 files): Strip UTF-8 BOM when decoding. |
|||
| nwc10 | jnthn: error in | 10:54 | |
| static MVMint32 its_the_bom(const char *utf8) { | |||
| needs to be unsigned char | |||
| jnthn | error? | ||
| nwc10 | IIRC, and from inspection | ||
| it's a bug | 10:55 | ||
| uns_utf8[0] == 0xEF && uns_utf8[1] == 0xBB && uns_utf8[2] == 0xBF | |||
| jnthn | In *both* cases, the things we pass to it are char * | ||
| nwc10 | will not be true if char is signed | ||
| jnthn | Uh | ||
| nwc10 | cannot be true if char is signed | ||
| jnthn | You missed a line | ||
| nwc10 | that's more what I meant | ||
| oh | |||
| you are corerct | |||
| I missed a line | |||
| jnthn | phew :) | 10:56 | |
| nwc10 | in which case, sorry | ||
| jnthn | np | ||
| Would much rather than code review with occasional false positives than no code review. :) | 10:57 | ||
| s:1st/than/have/ | |||
| nwc10 | I'm twitchy because it look me a long time previously to find that bug | 10:59 | |
| code had that bug in it, so the BOM-ness never triggered | |||
| until I built on ARM | |||
| obviously the code wasn't *meant* to trigger | |||
| but the bug had been concealed | |||
| jnthn | urgh | 11:00 | |
| Yeah | |||
| nwc10 | I'm suspicious that I shouldn't have any more coffee today | 11:01 | |
| jnthn only had the one today | 11:04 | ||
| nwc10 | but it was a pint? | 11:05 | |
| jnthn | Normalish size cup :) | ||
| masak | but it was a double espresso? | 11:06 | |
| jnthn | Well, it had that much worth of coffee in, I suspect :) | ||
| nwc10 | the Internet fails me. I can't find an image of a Maß filled with coffee | ||
| (or perhaps, more visually approrpiately, cappochino) | |||
| jnthn | hah...that has to be done! | 11:07 | |
| nwc10 | I am thinking this too | ||
| jnthn | Would be such a fitting image for hacktoberfest | ||
|
11:07
travis-ci joined
|
|||
| travis-ci | MoarVM build errored. jnthn 'Strip UTF-8 BOM when decoding.' | 11:07 | |
| travis-ci.org/MoarVM/MoarVM/builds/83865609 github.com/MoarVM/MoarVM/compare/5...9c7826a6ce | |||
|
11:07
travis-ci left
|
|||
| jnthn | Oh? | 11:08 | |
| ah | |||
| nwc10 | I don't have a Maß, and I don't know anyone from whom I could blag a loan. | ||
| jnthn | The command "sudo apt-get update -qq" failed and exited with 100 during . | ||
| Pretty sure that's not my fault :) | |||
| nwc10 | also, mmm, going to be hard to make a drinkable cappuccino in a heavy glass without either breaking the glass, or having coffee that is lukewarm at best, and cold soon after | 11:10 | |
| and not sure if any microwave I have access to is large enough to pre-warm the glass to avoid this | |||
| but these things can be worked round, if a glass can be procured | 11:11 | ||
|
11:11
Ven joined
|
|||
| jnthn | I suspect putting warm water into the glass first to warm it would help | 11:13 | |
| nwc10 | I was thinking exactly that. And incrementally raising the temperature of the water | ||
|
12:20
domidumont joined
|
|||
| domidumont | Hello. To optimize rakudo debian packages, I need to know if the moar files produced when building rakudo are the same whatever the architecture (i.e amd64, or arm ...) ? Can anyone confirm ? | 12:22 | |
| nwc10 | they are identical | ||
| (or it's a bug) | |||
| they have to be - NQP bootstraps using a copy of the file from git | |||
| which is the same file on all architectures | 12:23 | ||
| domidumont | nwc10: ok. thanks. Then I'll need to split rakudo package on Debian: one for the shared lib and one for all other files | ||
| jnthn | Yes, .moarvm files are architecture-independent | 12:46 | |
| And architecture-unoptimized | |||
| nwc10 | domidumont: would that be easier if Makefiles had 2 install targets? For architecture independent things, and arhcietecture dependent things? (and the ability to configure them to point to different places, keeping the defaults as they currently are?) | 13:18 | |
|
13:22
Ven joined
|
|||
| domidumont | nwc10: 2 install targets: no. because the whole package is built at once and files are then picked to be moved in one or another package | 13:29 | |
| nwc10 | aha right. thanks for that clarification | 13:30 | |
| domidumont | nwc10: on the other hand, rakudo lacks the ability to specify where to install the shared lib (I had to patch Makefile-moar.in) | ||
| nwc10: so the configuration parameter would help | |||
| On a unix system, a shared lib (a .so file) should never land in /usr/share | 13:31 | ||
|
15:34
tokuhirom joined
16:01
synbot6 joined
16:14
Ven joined
17:24
domidumont joined
17:27
flussence joined
17:44
Peter_R joined
17:52
FROGGS joined
|
|||
| FROGGS | o/ | 17:55 | |
| nwc10 | \o | 17:57 | |
|
18:35
tokuhirom joined
18:55
vendethiel joined
19:38
brrt joined
19:39
lizmat joined
|
|||
| brrt | good * #moarvm | 19:58 | |
| re: the coffee discussion, why not preheat the glass in an oven? | 20:02 | ||
| can be as slow as you like | |||
| easy to do, too | |||
| (heating water to heat a porcelain or glass cup is about as inefficient as can be due to water's huge heat coefficient) | |||
| lizmat | brrt: it was more about not cooling down so fast | 20:06 | |
| brrt | well, then a thick glass should do the trick, shouldn't it? | ||
| although you have these ehm, 'vacuum' hole glasses nowadays | 20:09 | ||
| those are better still | |||
|
20:15
nbiiebbiih joined
|
|||
| FROGGS .oO( Schroedingers cup... is it cold inside or hot? - Yes ) | 20:23 | ||
|
20:37
tokuhirom joined
20:39
dalek joined
20:46
camelia joined,
ashleyde1 joined,
psch joined,
tadzik joined,
masak joined
20:50
leont joined
20:51
tokuhirom joined
21:17
colomon joined
21:56
nbiiebbiih joined
23:22
colomon joined
23:51
flussence joined
|
|||