Welcome to the main channel on the development of MoarVM, a virtual machine for NQP and Rakudo (moarvm.org). This channel is being logged for historical purposes.
Set by lizmat on 24 May 2021.
00:08 reportable6 left 00:10 reportable6 joined 01:39 evalable6 left, linkable6 left 01:41 linkable6 joined 02:39 evalable6 joined 03:04 Techcable joined 03:17 frost joined 04:33 evalable6 left, linkable6 left, reportable6 left, squashable6 joined, shareable6 joined, reportable6 joined, greppable6 joined, bisectable6 joined 04:34 releasable6 joined 04:35 benchable6 joined, tellable6 joined, committable6 joined 04:36 notable6 joined, sourceable6 joined 05:12 frost left 05:33 linkable6 joined 05:34 quotable6 joined, nativecallable6 joined 06:06 reportable6 left 06:08 frost joined 06:18 kjp left 06:33 bloatable6 joined 06:34 coverable6 joined 06:35 statisfiable6 joined 06:54 frost left 07:24 frost joined 07:34 evalable6 joined 07:35 unicodable6 joined 08:08 reportable6 joined 08:11 frost left
Nicholas good *, #moarvm 09:57
MasterDuke Nicholas: were you looking to make coerce_si faster, or just checking for correctness. github.com/j-jorge/atoi-benchmark might have some faster implementations, but i think some of them are even more simplified, and i don't know how much of the speedup we'd lose adding back in the dropped functionality 10:33
Nicholas have two working commits to get rid of ctype.h 10:35
MasterDuke also, i wonder if that `if (str->body.storage_type == MVM_STRING_GRAPHEME_ASCII)` could be `|| str->body.storage_type == MVM_STRING_GRAPHEME8`
Nicholas but realised that there seemed to be bugs in the end of string handling
I now have written some tests
yet to try them on the JVM NQP 10:36
java.lang.NumberFormatException: For input string: " " 10:37
which was probably a tab
OK, "interesting" 10:39
So previously with the strtoll implementation we *weren't* exposed to locale issues because:
Error encoding ASCII string: could not encode codepoint 128
MasterDuke "The characters in the string must all be decimal digits, except that the first character may be an ASCII minus sign '-' (\u002D') to indicate a negative value or an ASCII plus sign '+' ('\u002B') to indicate a positive value. " so the jvm implementation never matched moarvm 10:40
Nicholas thanks for that definition - that's useful to know. 10:41
12:07 reportable6 left 12:10 reportable6 joined
Geth MoarVM: nwc10++ created pull request #1657:
Inline ascii ctype
14:56 timo joined 18:07 reportable6 left 18:08 reportable6 joined 19:38 Techcable left 20:53 MasterDuke left 21:06 squashable6 left 21:07 squashable6 joined 21:16 MasterDuke joined 21:31 kjp joined 22:32 Techcable joined