skids

I'm currently going through S17 to bring it up to date.

The following stuck out as things that may not be intentional behaviors,
so I need to know whether to doc current behavior or file an RT on each:

1) Should two different taps of a a serial supply be run simultaneuosly?

$ perl6 -e 'my $p = Supplier.new; my $s = $p.Supply; $s.serial.say; $s.tap({ sleep 1; "42 {now}".say; }); $s.tap({ sleep 1; "43 {now}".say }); $p.emit(42); sleep 3;'
True
42 Instant:1449454902.847377

skids

bri@atlas:~/git/perl6-Proc-Screen$ cat /tmp/screenrc
debug on
defutf8 off
bri@atlas:~/git/perl6-Proc-Screen$ screen -D -m -c /tmp/screenrc # then do "screen -Q info" from another window
Aborted
bri@atlas:~/git/perl6-Proc-Screen$ screen -ls
There are screens on:
        7478..atlas-queryA      (08/07/2015 09:41:20 PM)        (Dead ???)
        7478..atlas     (08/07/2015 09:41:20 PM)        (Dead ???)
Remove dead screens with 'screen -wipe'.

skids

# We have to use NativeCall here as there is otherwise no way to
# get an fd for Term::termios, as of yet.  And, it is buried so
# deep in the IO object it'll require guts work to unearth.
use NativeCall;
my sub open (str $pathname, int $flags) returns int is native {*}
my sub read (int $fd, Buf $buf is rw, int $count) returns int is native {*}
my sub close (int $fd) returns int is native {*}

my sub get-input {

skids

So, this would seem to be the way to make a Supply from an IO::Pipe useful for waiting for the
pipe to close and then doing something with the pipe output.  But it does not work reliably.
Similar schemes working directly with the Supply also have issues, since any state accumulating
the output in tap callbacks may not be complete inside a :done callback.  Note that sometimes
(maybe when certain parts of the guts schedule on the same thread) it pretty much works -- I killed
statistically improbable runs with ^C below.  Also note failures on run 8 and run 32 are popular.


$ perl6 -e 'sub f { my $cmd = Proc::Async.new(:path<ls>); my $c = $cmd.stdout.Channel; $cmd.start; $'
after 6 times

skids

I think there might be a sliver of performance gain on that test script
with the following, but it is hard to tell.  Oddly, it generates smaller
QAST but larger MAST.


diff --git a/src/core/Array.pm b/src/core/Array.pm
index e4960b0..2e03127 100644
--- a/src/core/Array.pm
+++ b/src/core/Array.pm
@@ -319,7 +319,7 @@ my class Array { # declared in BOOTSTRAP

skids

Copyright (C) 1999-2015 The FreeRADIUS server project and contributors
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE
You may redistribute copies of FreeRADIUS under the terms of the
GNU General Public License
For more information about these matters, see the file named COPYRIGHT
Starting - reading configuration files ...
including dictionary file /usr/share/freeradius/dictionary
including dictionary file /usr/share/freeradius/dictionary.dhcp
including dictionary file /usr/share/freeradius/dictionary.vqp

skids

--- a/lib/Language/functions.pod
+++ b/lib/Language/functions.pod
@@ -155,6 +155,58 @@ type:
+=head1 Conventions and Idioms
+
+While the dispatch system described above provides a lot of flexibility,
+there are some conventions that most internal functions, and those in
+many modules, will follow.  These produce a consistent look and feel.
+
+=head2 Slurpy Conventions

skids

# This shows the cost of generating backtraces when using Str.Numeric
# to test whether a string can be numified, which is a normal usage
# scenario.
#
# Test code: different from the pre-allomorph original in that it will only
# ever generate one Failure per .Numeric, and as long as a :$backtrace is
# provided and the result is not thrown, should not generate any new backtraces
# (I think.)  But -- also different in that it always uses an exception unwind
# instead of letting Failure objects go through the normal return path.

skids

my sub build-cursor-to-template {
        my ($x,$y) = 13,13;
        my $raw = qq:x{ tput cup $y $x };

        my Str sub cursor-template( Int :$x,  Int :$y ) {
            # there may be single digit numbers in the escape preamble
            $raw ~~ s:nth(*-1)[\d+] = $y+1;
            $raw ~~ s:nth(*)[\d+]   = $x+1;
            return $raw;
        }

skids

diff --git a/src/core/Capture.pm b/src/core/Capture.pm
index 80d013c..dacdb85 100644
--- a/src/core/Capture.pm
+++ b/src/core/Capture.pm
@@ -96,8 +96,8 @@ my class Capture { # declared in BOOTSTRAP
         if self.^name eq 'Capture' {
             "\\({
                 join ', ',
-                    (nqp::atpos($!list, $_).perl for ^nqp::elems($!list)),
-                    %hash.sort.map( *.perl )