🦋 Welcome to the IRC channel of the core developers of the Raku Programming Language (raku.org #rakulang). This channel is logged for the purpose of history keeping about its development | evalbot usage: 'm: say 3;' or /msg camelia m: ... | log inspection situation still under development | For MoarVM see #moarvm
Set by lizmat on 22 May 2021.
00:02 reportable6 left
Geth DBIish/rbt.pg-column-type-change: 9342a220c4 | (Rod Taylor)++ | lib/DBDish/Pg/StatementHandle.pm6
Allow changing column-type manually again

There is an undocumented and very lightly tested feature which allows the column-type to be specified by the client after the prepare/execute phase of the query but before retrieving rows. Re-enable this again for the PostgreSQL driver by setting up import functions on the first (and only the first) call to row().
00:03
DBIish/rbt.pg-column-type-change: 5828d47363 | (Rod Taylor)++ | lib/DBDish/Pg/StatementHandle.pm6
Allow changing column-type manually again

There is an undocumented and very lightly tested feature which allows the column-type to be specified by the client after the prepare/execute phase of the query but before retrieving rows. Re-enable this again for the PostgreSQL driver by setting up import functions on the first (and only the first) call to row().
00:07
DBIish: 5828d47363 | (Rod Taylor)++ | lib/DBDish/Pg/StatementHandle.pm6
Allow changing column-type manually again

There is an undocumented and very lightly tested feature which allows the column-type to be specified by the client after the prepare/execute phase of the query but before retrieving rows. Re-enable this again for the PostgreSQL driver by setting up import functions on the first (and only the first) call to row().
00:24
DBIish/rbt.mysql-large-result-performance: a960748a66 | (Rod Taylor)++ | lib/DBDish/Pg/StatementHandle.pm6
Fix when a decoding function isn't provided

If a decoding function isn't provided in the map, assume String as was done before refactoring in c656ee5.
00:25
DBIish/rbt.mysql-large-result-performance: d6b8f88c9b | (Rod Taylor)++ | 2 files
Handle type selection once for MySQL driver

This is the same technique as the PostgreSQL driver in c656ee5. Setup type conversion functions once per result set instead of hunting for them once per record.
Iterating on @!import-func and incrementing $col seems to perform the best, though it's marginal due to overhead in the rest of the driver.
DBIish/rbt.mysql-large-result-performance: 58c7456813 | (Rod Taylor)++ | 2 files
Handle type selection once for MySQL driver

This is the same technique as the PostgreSQL driver in c656ee5. Setup type conversion functions once per result set instead of hunting for them once per record.
Iterating on @!import-func and incrementing $col seems to perform the best, though it's marginal due to overhead in the rest of the driver.
00:27
DBIish/rbt.mysql-large-result-performance: 12e72d1074 | (Rod Taylor)++ | 2 files
Handle type selection once for MySQL driver

This is the same technique as the PostgreSQL driver in c656ee5. Setup type conversion functions once per result set instead of hunting for them once per record.
Iterating on @!import-func and incrementing $col seems to perform the best, though it's marginal due to overhead in the rest of the driver.
00:28
DBIish: a960748a66 | (Rod Taylor)++ | lib/DBDish/Pg/StatementHandle.pm6
Fix when a decoding function isn't provided

If a decoding function isn't provided in the map, assume String as was done before refactoring in c656ee5.
00:41
DBIish: 12e72d1074 | (Rod Taylor)++ | 2 files
Handle type selection once for MySQL driver

This is the same technique as the PostgreSQL driver in c656ee5. Setup type conversion functions once per result set instead of hunting for them once per record.
Iterating on @!import-func and incrementing $col seems to perform the best, though it's marginal due to overhead in the rest of the driver.
DBIish/rbt.remove-unused-variable: 51fb8dda6f | (Rod Taylor)++ | lib/DBDish/mysql/StatementHandle.pm6
Handle type selection once for MySQL driver

This is the same technique as the PostgreSQL driver in c656ee5. Setup type conversion functions once per result set instead of hunting for them once per record.
Iterating on @!import-func and incrementing $col seems to perform the best, though it's marginal due to overhead in the rest of the driver.
00:51
DBIish/rbt.remove-unused-variable: 7e07c13dae | (Rod Taylor)++ | lib/DBDish/mysql/StatementHandle.pm6
Remove unused variable
DBIish: 51fb8dda6f | (Rod Taylor)++ | lib/DBDish/mysql/StatementHandle.pm6
Handle type selection once for MySQL driver

This is the same technique as the PostgreSQL driver in c656ee5. Setup type conversion functions once per result set instead of hunting for them once per record.
Iterating on @!import-func and incrementing $col seems to perform the best, though it's marginal due to overhead in the rest of the driver.
DBIish: 7e07c13dae | (Rod Taylor)++ | lib/DBDish/mysql/StatementHandle.pm6
Remove unused variable
DBIish: 2be50ef47c | (Rod Taylor)++ | lib/DBDish/mysql/StatementHandle.pm6
Revert "Handle type selection once for MySQL driver"

This reverts commit 51fb8dda6fedbd46066f1d7631c7db7257b20595.
00:52
01:03 reportable6 joined 02:01 squashable6 left 02:30 Xliff_ joined 02:33 _Xliff_ left 03:03 squashable6 joined 03:14 frost joined 04:38 evalable6 left, linkable6 left, evalable6 joined 05:40 linkable6 joined 06:02 reportable6 left 06:03 reportable6 joined 07:11 frost left 07:27 squashable6 left 08:03 timo left 09:25 linkable6 left, evalable6 left 09:28 linkable6 joined 09:29 squashable6 joined 10:27 evalable6 joined 10:45 frost joined 11:05 Altai-man joined 11:46 lizmat_ joined 11:47 TempIRCLogger left, _Xliff_ joined 11:48 nebuchad` joined, lizmat_ left, lizmat_ joined 11:50 TempIRCLogger joined 11:51 bartolin_ joined, lucs_ joined, Util_ joined, RakuIRCLogger joined 11:55 lizmat_ left, lizmat_ joined, lizmat left, gfldex left, bartolin left, tonyo left, JRaspass left, Xliff_ left 11:56 nebuchadnezzar left, lucs left, Util left, gfldex_ joined, lizmat_ left, lizmat joined 12:02 reportable6 left 12:03 JRaspass joined
Geth rakudo: 1fc8cce31a | (Elizabeth Mattijsen)++ | src/core.c/Distro.pm6
Add knowledge about MacOS Monterey
12:06
12:06 frost left 13:05 reportable6 joined 13:45 Xliff_ joined 13:46 Xliff_ left 13:48 _Xliff_ left 13:56 squashable6 left 13:58 squashable6 joined 15:19 Kaiepi left
[Tux] Rakudo v2021.10-127-g1fc8cce31 (v6.d) on MoarVM 2021.10-113-g66688b941
csv-ip5xs0.849 - 0.859
csv-ip5xs-205.150 - 5.280
csv-parser4.147 - 4.227
csv-test-xs-200.408 - 0.417
test6.967 - 6.996
test-t1.638 - 1.641
test-t --race0.912 - 0.964
test-t-2024.085 - 25.026
test-t-20 --race7.348 - 7.367
15:57
16:08 Kaiepi joined 17:44 notna joined 17:47 Colt left, Colt joined 17:56 Colt left 17:59 Colt joined 18:02 reportable6 left 18:07 Colt left, Colt joined 18:15 notna left 18:38 [Coke] left 18:43 Altai-man left 18:45 [Coke] joined
Geth nqp: usev6++ created pull request #749:
[JVM] Reimplement nqp::encode for utf-16
20:44
¦ nqp: usev6 self-assigned Encoding a string as utf16 to buffer containing 8 bit integers is broken on the JVM github.com/Raku/nqp/issues/250 20:54
21:03 reportable6 joined
Geth rakudo/max-threads-default: 3eb584a445 | (Elizabeth Mattijsen)++ | src/core.c/ThreadPoolScheduler.pm6
Make sure we have at least 64 threads
21:29
22:03 evalable6 left 22:11 [Coke] left 22:13 [Coke] joined
[Coke] 64 seems high for low end machines, no? 22:14
vrurg [Coke]: just don't use all of them. 22:21
22:34 kjp joined 22:45 squashable6 left
lizmat [Coke]: that *is* the current fixed default :-) 23:03
23:31 dogbert11 left, dogbert11 joined 23:53 lizmat_ joined, RakuIRCLogger left, TempIRCLogger left 23:54 Geth left, lizmat left