Travis-CI and Rakudo seem to be a bit unstable

[Gabor Szabo]

Hi,

In addition to me breaking my own code, it seems that Travis-CI and
Rakudo are rather fragile together.

Occasionally my tests fail due to failures in the prerequisites.

This failed due to LWP::Simple failing
https://travis-ci.org/szabgab/Perl6-Maven/jobs/215147392
even though 20 min earlier it worked:
https://travis-ci.org/szabgab/Perl6-Maven/builds/215143736

Yesterday Bailador failed:
https://travis-ci.org/szabgab/Perl6-Maven/jobs/214981400
even though it worked a few minutes earlier and a few minutes later.

As far as I can tell neither LWP::Simple nor Bailador changed during that time.

Have you encountered similar issues?  Am I misreading something?


regards
   Gabor

Re: Failed to open file .. too many open files

[Gabor Szabo]

I still get the same error.
I just found another case of slurp-rest which might have been the
cause but it might be good time to
look at the few other cases of "open" in my code. (And later maybe
also the code of Bailador itself.)
I tried to read more about how I am supposed to close file handles in
Perl 6, but the explanation I found didn't help me.
I opened this ticket https://github.com/perl6/doc/issues/1258 asking
for further clarification.

Anyway here is another code snippet from my code:

for open("$.source_dir/authors.txt").lines -> $line {
 ...
}

Do I need to close this myself? Can I rely on Perl closing it?

If I open a file for reading like this:

sub f {
my $fh = open $file, :r;
LEAVE $fh.close;

for $fh.lines -> $line {
}
}

Is that the correct way to put the LEAVE in? Should it be immediately
after the open statement?

And for writing?

sub write {
my $fh = open $file, :w;
LEAVE $fh.close;
$fh.print("text");
}

regards
Gabor

Re: Perl 6 docs

[Shlomi Fish]

On Sat, 25 Mar 2017 16:49:04 +0100
Gabor Szabo  wrote:

> https://github.com/perl6/doc/issues/1257

Thanks for clarifying!

Re: Failed to open file .. too many open files

[Gabor Szabo]

On Sat, Mar 25, 2017 at 8:42 PM, Elizabeth Mattijsen  wrote:
> $file.IO.slurp and slurp($file) are basically the same.
>
> $handle.slurp-rest does *not* close the handle, as another process might 
> still be writing to it, so you could do another .slurp-rest.
>
>
> To get back to your original code:
>
>get '/atom' => sub {
>my $path = $.meta ~ request.path;
>return open($path).slurp-rest;
>}
>
> I would write that as:
>
>get '/atom' => sub { slurp $.meta ~ request.path }
>
> Should you wind up with an opened handle, could could use a LEAVE phaser to 
> make sure the handle gets closed:
>
>get '/atom' => sub {
>LEAVE $.meta.handle.close;
>return $.meta.handle.slurp-rest;
>}
>

Thanks.

I've converted all those slurp-rest calls so slurp calls.
I am not sure why did I have the slurp-rest in there.
That code seems to be at least 2 years old.

Anyway, thanks for the explanation.

Gabor

Re: Failed to open file .. too many open files

[Elizabeth Mattijsen]

$file.IO.slurp and slurp($file) are basically the same.

$handle.slurp-rest does *not* close the handle, as another process might still 
be writing to it, so you could do another .slurp-rest.


To get back to your original code:

   get '/atom' => sub {
   my $path = $.meta ~ request.path;
   return open($path).slurp-rest;   
   }

I would write that as:

   get '/atom' => sub { slurp $.meta ~ request.path }

Should you wind up with an opened handle, could could use a LEAVE phaser to 
make sure the handle gets closed:

   get '/atom' => sub {
   LEAVE $.meta.handle.close;
   return $.meta.handle.slurp-rest;
   }

HTH,


Liz
=
> On 25 Mar 2017, at 17:12, Gabor Szabo  wrote:
> 
> Oh so you say that's indeed a bug in my code. That's a relief. Thanks!
> 
> 
> As I can see I had some $file.IO.slurp and some of the slurp-rest ones.
> 
> What is the difference between $file.IO.slurp and slurp($file) ?
> Is the latter just an alias for the former?
> 
> Gabor
> 
> 
> On Sat, Mar 25, 2017 at 4:54 PM, Timo Paulssen  wrote:
>> i highly suggest you slurp instead of open + slurp-rest, because that
>> will automatically close the file for you, too.
>> 
>> other than that, you can pass :close to the slurp-rest method and it'll
>> also close the file.
>> 
>> if you're not closing the files you're opening, you'll be relying on the
>> garbage collector to do file handle closing for you, which is
>> nondeterministic and a bad idea in general.
>> 
>> HTH
>>  - Timo

Re: Can this OR be shortened?

[Brandon Allbery]

On Fri, Mar 24, 2017 at 10:45 PM, Brad Gilbert  wrote:

> Basically use | in regexes unless you need ||.


There's been some discussion, here and in IRC, of the | form interacting in
ways that are correct but which people often don't expect. This can result
in match failures or unexpected slowness. My takeaway is there's no
simple-minded "do this" or "do this if/unless" rule as yet.

-- 
brandon s allbery kf8nh   sine nomine associates
allber...@gmail.com  ballb...@sinenomine.net
unix, openafs, kerberos, infrastructure, xmonadhttp://sinenomine.net

Re: Failed to open file .. too many open files

[Gabor Szabo]

Oh so you say that's indeed a bug in my code. That's a relief. Thanks!


As I can see I had some $file.IO.slurp and some of the slurp-rest ones.

What is the difference between $file.IO.slurp and slurp($file) ?
Is the latter just an alias for the former?

Gabor


On Sat, Mar 25, 2017 at 4:54 PM, Timo Paulssen  wrote:
> i highly suggest you slurp instead of open + slurp-rest, because that
> will automatically close the file for you, too.
>
> other than that, you can pass :close to the slurp-rest method and it'll
> also close the file.
>
> if you're not closing the files you're opening, you'll be relying on the
> garbage collector to do file handle closing for you, which is
> nondeterministic and a bad idea in general.
>
> HTH
>   - Timo

Re: Failed to open file .. too many open files

[Timo Paulssen]

i highly suggest you slurp instead of open + slurp-rest, because that
will automatically close the file for you, too.

other than that, you can pass :close to the slurp-rest method and it'll
also close the file.

if you're not closing the files you're opening, you'll be relying on the
garbage collector to do file handle closing for you, which is
nondeterministic and a bad idea in general.

HTH
  - Timo

Failed to open file .. too many open files

[Gabor Szabo]

The Perl 6 Maven site runs on Bailador. I've just updated the Rakudo
underneath to 2017.01 and
it seemed to be working fine, but after a while it started crashing
with this error message:



Failed to open file /home/gabor/work/perl6maven-live.com/main.json:
too many open files
  in sub  at /home/gabor/work/Perl6-Maven/lib/Perl6/Maven.pm6
(Perl6::Maven) line 26
  in block  at 
/home/gabor/rakudo-star-2017.01/install/share/perl6/site/sources/A80D3EE41ED647143B6A367AAE142A5967A850D7
(Bailador::Route) line 57
  in method recurse-on-routes at
/home/gabor/rakudo-star-2017.01/install/share/perl6/site/sources/A80D3EE41ED647143B6A367AAE142A5967A850D7
(Bailador::Route) line 56
  in method dispatch at
/home/gabor/rakudo-star-2017.01/install/share/perl6/site/sources/06F75D046213CAD92B84F66E5F161C084878D0C3
(Bailador::App) line 84
  in block  at 
/home/gabor/rakudo-star-2017.01/install/share/perl6/site/sources/06F75D046213CAD92B84F66E5F161C084878D0C3
(Bailador::App) line 77

I've tried to reproduce the problem by running curl against the local
version of the application,
but so far, after a few hundred requests, I have not encountered the problem.


line 26 in the Maven.pm6 file looks like this:

get '/atom' => sub {
my $path = $.meta ~ request.path;
return open($path).slurp-rest;
}

I wonder if you have any idea what could be the source of this problem?

The full source of the application is here:
https://github.com/szabgab/Perl6-Maven/

regards
Gabor

Re: Using Rakudo Start on OSX using the .dmg

[Will Coleda]

These notes are already available in the README.txt file that is in the .dmg

Regards.

On Sat, Mar 25, 2017 at 4:54 AM, Gabor Szabo  wrote:
> Hi,
>
> I just tried to use the .dmg version of Rakudo Star.
> I might not be the typical Mac user as I spent quite some time trying to 
> figure
> out what do I need to do in order to start using it after the installation.
>
> In the end I found that it was installed to  /Applications/Rakudo
>
> Then I added these to my .bash_profile:
>
> export RAKUDO=/Applications/Rakudo
> export PATH=$RAKUDO/bin:$RAKUDO/share/perl6/site/bin/:$PATH
> export PERL6LIB=$RAKUDO/share/perl6/site/lib/
>
> reloaded it and then I could use it.
>
> Maybe some notes like this could be added to
> http://rakudo.org/how-to-get-rakudo/
>
> regards
>Gabor



-- 
Will "Coke" Coleda

Re: Perl 6 docs

[Will Coleda]

Please open an issue at https://github.com/perl6/doc/issues so it gets
tracked, thanks!

On Sat, Mar 25, 2017 at 8:36 AM, Gabor Szabo  wrote:
> When I search for %INC at https://docs.perl6.org/ it offers   "%INC (Perl 5)"
> but when I search for the more common @INC
>
> Luckily the former leads to
> https://docs.perl6.org/language/5to6-perlvar which also has
> information on the latter, but it would be nice if that was also
> recognized in the search box.
>
> Gabor



-- 
Will "Coke" Coleda

Re: Can this OR be shortened?

[Timo Paulssen]

On 25/03/17 12:53, Tom Browder wrote:
> On Fri, Mar 24, 2017 at 10:36 PM, Timo Paulssen  wrote:
>> I seem to recall you asked about performance recently
>>
>> the regex engine has a significant overhead.
> ...
>
> Isn't that somewhat mitigated by defining regex or token constants for reuse?
>
> -Tom

The overhead is from starting up and finishing a regex match. Creating
the Match object is quite expensive, for example. all regexes are
compiled into code objects, of course.

Perl 6 docs

[Gabor Szabo]

When I search for %INC at https://docs.perl6.org/ it offers   "%INC (Perl 5)"
but when I search for the more common @INC

Luckily the former leads to
https://docs.perl6.org/language/5to6-perlvar which also has
information on the latter, but it would be nice if that was also
recognized in the search box.

Gabor

Re: Can this OR be shortened?

[yary]

On Sat, Mar 25, 2017 at 1:15 AM, ToddAndMargo  wrote:
> On 03/24/2017 07:45 PM, Brad Gilbert wrote:
>>
>> All of these should work
>>
>> if $Terminal ~~ /xterm/ | /linux/ {}
>> if $Terminal ~~ /xterm | linux/ {}
>> if $Terminal ~~ /xterm || linux/ {}
>>
>> Note that | in a regex tries both sides as if in parallel, and goes
>> for the longest,
>
>
> Hi Brad,
>
> What do you mean by longest?


my $a = 'Bart sez cowabunga!';

# matches cowabunga because that's the longest match
$a ~~ / cow | cowabunga /;
say $/;  # "cowabunga"

# matches cow because it is the first alternation
$a ~~ / cow || cowabunga /;
say $/;  # "cow"

This is useful in writing parsers- when given a choice of keywords
that have the same prefix, longest match will find the keyword people
expect regardless of the order of the alternations. Keeps people from
having to remember to also search for a word boundary at the end,
makes it easier to write bug-free grammars.

-y

Using Rakudo Start on OSX using the .dmg

[Gabor Szabo]

Hi,

I just tried to use the .dmg version of Rakudo Star.
I might not be the typical Mac user as I spent quite some time trying to figure
out what do I need to do in order to start using it after the installation.

In the end I found that it was installed to  /Applications/Rakudo

Then I added these to my .bash_profile:

export RAKUDO=/Applications/Rakudo
export PATH=$RAKUDO/bin:$RAKUDO/share/perl6/site/bin/:$PATH
export PERL6LIB=$RAKUDO/share/perl6/site/lib/

reloaded it and then I could use it.

Maybe some notes like this could be added to
http://rakudo.org/how-to-get-rakudo/

regards
   Gabor