Perl 6 - the future is here, just unevenly distributed

IRC log for #parrot, 2014-07-22

Parrot | source cross referenced

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

All times shown according to UTC.

Time Nick Message
01:16 FROGGS_ joined #parrot
02:21 kid51 joined #parrot
04:09 estrabd joined #parrot
04:15 perlfan joined #parrot
04:51 bighugedog joined #parrot
06:36 basiliscos joined #parrot
07:31 kjs joined #parrot
08:08 basiliscos joined #parrot
08:32 wagle joined #parrot
09:19 kjs joined #parrot
09:30 Psyche^ joined #parrot
11:05 kid51 joined #parrot
13:29 rurban1 joined #parrot
13:31 bluescreen joined #parrot
14:14 khisanth_ joined #parrot
15:02 Chirag joined #parrot
17:31 rurban1 joined #parrot
18:10 Chirag rurban: https://gist.github.com/ZYROz/2b709dc74bd354143f81
18:10 Chirag But my log.data looks very unreliable
18:32 rurban1 joined #parrot
18:41 rurban 152 vs 67 secs :)
18:42 rurban rather 151617.489213,task-clock vs 152306.055310,task-clock
18:43 rurban so is RELEASE_6_6_0-0-gad40760 static call_obj and RELEASE_6_6_0-1-g71d5afe the direct call? can we have a diff between those two commits?
19:32 Util #ps time
19:33 Chirag no.. I am still working on the direct calls part.. This is after changing pmc_callcontext.h only
19:33 rurban1 joined #parrot
19:41 Chirag diff does not work since the .h file is not monitored, I guess .. But basically all the macros are changed...
19:43 Util To get a diff of an untracked file, build the software in its unmodified form, then copy the untracked file to a new name, then modify, rebuild, and run "diff -u newName.h currentName.h".
19:44 Chirag oh.. thanks for that!
20:10 rurban I see
20:11 rurban Maybe it's better to add this hack and push it. so it's easier to repro
20:11 Chirag hmm..
20:12 rurban we just need to remember to remove it later
20:12 Chirag ok..
20:13 rurban and maybe make it readonly so pmc2c will not overwrite it
20:13 rurban chmod -w
20:18 Chirag pushed
20:18 davidfetter joined #parrot
20:35 rurban1 joined #parrot
20:37 rurban I also prepared a faster callcontext now, just not expanded the vtables calls yet, only ATTR
20:37 rurban all tests pass
20:39 dalek parrot/rurban/pcc-gh1083: 71d5afe | ZYROz++ | src/call/pcc.c:
20:39 dalek parrot/rurban/pcc-gh1083: [pcc] inline pcc_add_invocant and its attr accessor
20:39 dalek parrot/rurban/pcc-gh1083: review: https://github.com/parrot/parrot/commit/71d5afeb1d
20:39 dalek parrot/rurban/pcc-gh1083: 56e2d00 | rurban++ | / (2 files):
20:39 dalek parrot/rurban/pcc-gh1083: [pcc] temporarily add two hand-optimized callcontext pmc files
20:39 dalek parrot/rurban/pcc-gh1083:
20:39 dalek parrot/rurban/pcc-gh1083: see gh #1083
20:39 dalek parrot/rurban/pcc-gh1083: review: https://github.com/parrot/parrot/commit/56e2d00aa4
20:40 dalek parrot/rurban/pcc-gh1083: 7b7a019 | ZYROz++ | src/call/pcc.c:
20:40 dalek parrot/rurban/pcc-gh1083: [pcc] inline pcc_add_invocant and its attr accessor
20:40 dalek parrot/rurban/pcc-gh1083: review: https://github.com/parrot/parrot/commit/7b7a0197e4
20:40 dalek parrot/rurban/pcc-gh1083: 4353c37 | rurban++ | / (2 files):
20:40 dalek parrot/rurban/pcc-gh1083: [pcc] temporarily add two hand-optimized callcontext pmc files
20:40 dalek parrot/rurban/pcc-gh1083:
20:40 dalek parrot/rurban/pcc-gh1083: see gh #1083
20:40 dalek parrot/rurban/pcc-gh1083: review: https://github.com/parrot/parrot/commit/4353c37369
20:50 rurban my load is still >5
20:51 rurban ( +-  9.17% )
21:10 rurban 3% faster
21:10 dalek parrot/rurban/pcc-gh1083: 7faa17b | rurban++ | / (2 files):
21:10 dalek parrot/rurban/pcc-gh1083: [pcc] more hand-edited changes GH #1083
21:10 dalek parrot/rurban/pcc-gh1083: review: https://github.com/parrot/parrot/commit/7faa17b9c3
21:10 dalek parrot/rurban/pcc-gh1083: 970c54b | rurban++ | src/call/ (2 files):
21:10 dalek parrot/rurban/pcc-gh1083: [pcc] actually use now the direct methods GH #1083
21:10 dalek parrot/rurban/pcc-gh1083:
21:10 dalek parrot/rurban/pcc-gh1083: and do not use the CallContext vtable methods anymore.
21:10 dalek parrot/rurban/pcc-gh1083: also inline one Continuation_set_pointer call.
21:10 dalek parrot/rurban/pcc-gh1083:
21:10 dalek parrot/rurban/pcc-gh1083: all tests still passing, 2.98% faster.
21:10 dalek parrot/rurban/pcc-gh1083: review: https://github.com/parrot/parrot/commit/970c54b5e2
21:10 rurban so lets change pmc2c to use the faster CallContext calls
21:11 rurban Chirag: can you try also my changes?
21:13 basiliscos joined #parrot
21:35 rurban1 joined #parrot
21:44 rurban FROGGS_: I'll rather fix tools/dev/gen_charset_tables.pl also. just the  u_hasBinaryProperty(codepoint, UCHAR_LINE_BREAK) looks good then
21:44 FROGGS_ yeah, probably makes sense
21:44 rurban -    $ret |= 0x0020 if $chr =~ /^[[:space:]\x85\xa0]$/;    # CCLASS_WHITESPACE
21:44 rurban +    $ret |= 0x0020 if $chr =~ /^[[:space:]\x85\xa0\x2028\x2029]$/;# CCLASS_WHITESPACE
21:45 rurban -    $ret |= 0x1000 if $chr =~ /^[\n\r\f\x85]$/;           # CCLASS_NEWLINE
21:45 rurban +    $ret |= 0x1000 if $chr =~ /^[\n\r\f\x85\x2028\x2029]$/;# CCLASS_NEWLINE
21:45 rurban I'll check
21:45 FROGGS_ rurban: \x2028 does not make much sense there, because this is only used to generate the properties of the first 256 codepoints
21:46 FROGGS_ also, \x2028 will match " 28"
21:46 rurban oops
21:46 FROGGS_ and as I said, >\xFF is a noop there
21:47 FROGGS_ for some reason my libicu does that think that \x2028 and \x2029 are newlines
21:47 FROGGS_ that's why I hardcoded these two
21:47 rurban so we need m//u and properly escape them
21:48 rurban oh, I see 0..255 only
21:48 FROGGS_ well, only if that script would create a table for all codepoints :o)
21:48 FROGGS_ so you can leave out the \x2028 and \x2029 there
21:49 rurban "Create a whole character table" is a bit overly enhusiastic then
21:53 rurban Uh, and Parrot_ascii_typetable is missing, and the generated table deviates wildy
21:54 rurban ascii was changed to fixed8 with 505fdc4c9a48fe90d17
22:04 rurban ok, /^[[:space:]\x85\xa0]$/ is needed, already included. now looking for a POSIX newline class...
22:08 rurban is *not* needed
22:11 rurban FROGGS_: 11 is a newline (VT)? I think not, POSIX thinks neither. only 12 and 13 are
22:12 rurban oh, with Word it served as a newline
22:12 rurban (powerpoint slides are using it)
22:13 rurban LINE TABULATION
22:13 FROGGS_ # Line_Break=Mandatory_Break
22:13 FROGGS_ 000B..000C    ; BK # Cc   [2] <control-000B>..<control-000C>
22:15 rurban so it's     $ret |= 0x1000 if $chr =~ /^[\v\n\r\f\x85]$/;         # CCLASS_NEWLINE
22:15 rurban then
22:15 rurban right?
22:17 rurban So I've got discrepancies with 215 and 247 in [[:punct:]]
22:17 FROGGS_ /^[\v\n\r\f\x85]$/ is correct
22:17 rurban 160,168,176,184,190 also
22:18 rurban and some more
22:18 FROGGS_ :/
22:18 rurban I think I'll rather trust libc for this
22:21 FROGGS_ gnight
22:21 rurban thanks again
22:21 FROGGS_ ohh, I have to thank you :o)
22:36 rurban1 joined #parrot
22:50 dalek parrot/FROGGS/newlines2-gh1087: 7789e9f | (Tobias Leich)++ | src/string/encoding/ (2 files):
22:50 dalek parrot/FROGGS/newlines2-gh1087: RT #122341 treat all of 0x0a-0x0d,\x85,\x2028,\x2029 as newlines
22:50 dalek parrot/FROGGS/newlines2-gh1087:
22:50 dalek parrot/FROGGS/newlines2-gh1087: The script tools/dev/gen_charset_tables.pl was not used because it removes
22:50 dalek parrot/FROGGS/newlines2-gh1087: character properties of chars in the range of 0x81 to 0xFF. Additionally,
22:50 dalek parrot/FROGGS/newlines2-gh1087: u_iscclass now checks for characters with enum_cclass_newline property,
22:50 dalek parrot/FROGGS/newlines2-gh1087: which it did not do at all before.
22:50 dalek parrot/FROGGS/newlines2-gh1087:
22:50 dalek parrot/FROGGS/newlines2-gh1087: Signed-off-by: Reini Urban <rurban@cpanel.net>
22:50 dalek parrot/FROGGS/newlines2-gh1087: review: https://github.com/parrot/parrot/commit/7789e9f9ca
22:50 dalek parrot/FROGGS/newlines2-gh1087: c1d57d2 | rurban++ | / (4 files):
22:50 dalek parrot/FROGGS/newlines2-gh1087: regenerate src/string/encoding/tables.c
22:50 dalek parrot/FROGGS/newlines2-gh1087:
22:50 dalek parrot/FROGGS/newlines2-gh1087: so that tools/dev/gen_charset_tables.pl is kept up-to-date.
22:50 dalek parrot/FROGGS/newlines2-gh1087: (it was not)
22:50 dalek parrot/FROGGS/newlines2-gh1087: Added \v to CCLASS_NEWLINE manually (confirmed).
22:50 dalek parrot/FROGGS/newlines2-gh1087: \x85\xa0 already is in the whitespace cclass.
22:50 dalek parrot/FROGGS/newlines2-gh1087: Removed the creation of the now defunct ascii table.
22:50 dalek parrot/FROGGS/newlines2-gh1087:
22:50 dalek parrot/FROGGS/newlines2-gh1087: Beware: My new up-to-date libc removed the [[:punct::]] class
22:50 dalek parrot/FROGGS/newlines2-gh1087: of several chars > 160.
22:50 dalek parrot/FROGGS/newlines2-gh1087: review: https://github.com/parrot/parrot/commit/c1d57d25fd
22:57 rurban1 joined #parrot
23:50 kid51 joined #parrot

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

Parrot | source cross referenced