2017.45 Uplink Established

Elizabeth Mattijsen got carried away trying to devise an API for getrusage. The result is a new set of modules available with use Telemetry. And a helper module for easy activation: snapper. The simplest use case: perl6 -Msnapper yourscript.pl6. This will start up a separate thread that will take snapshots every 0.1 seconds. Once it is done, it will display a report much like this on STDERR:

Telemetry Report of Process #72908 (2017-11-06T21:26:26Z)
Number of Snapshots: 7
Supervisor thread ran the whole time
Initial Size:        67676 Kbytes
Total Time:           0.58 seconds
Total CPU Usage:      0.59 seconds

wallclock  util%  max-rss
   105476  13.28    17680
   105220  13.27       12
   105178  12.73     1024
   105149  12.85      152
   105141  12.42        4
    49525  12.98       12
--------- ------ --------
   575689  12.92    18884

Legend:
wallclock  Number of microseconds elapsed
    util%  Percentage of CPU utilization (0..100%)
  max-rss  Maximum resident set size (in Kbytes)

Even this hot off the press, Wenzel P. P. Peppmeyer took this new development to actually create a distribution that runs an httpd daemon in your program, so that you can look at usage data of your (long running program) in your browser!

But that’s only the tip of the iceberg: there are also a number of ways to get at system information ad-hoc, such as:

$ perl6 -MTelemetry -e 'say T<wallclock cpu max-rss>'
(140878 264269 84756)

Or plan your snapshots and reporting more precisely:

while $working {
    snap;
    LAST snap;
    # do a lot of work
}
say report(:csv)  # output in CSV format

Other features of Telemetry include a Telemetry::Instrument role for creating your own “instruments” for plugging into this framework (with two Instruments already supplied: Telemetry::Instrument::Usage (basically giving all the data of getrusage + wallclock in microseconds) and Telemetry::Instrument::ThreadPool (giving all the data pertaining to actions of the $*SCHEDULER, such as being able to see when new worker threads are started). Exciting times!

LPW Rakudo Perl 6 Presentations

Well, a program as such hasn’t been decided yet, but it looks like the following Rakudo Perl 6 related presentations will be given at the London Perl Workshop on 25 November 2017 at the University of Westminster:

Hope to see you there!

Optimizing Code

Jonathan Worthington published another part of his MoarVM Specializer Improvements series: Optimizing Code. A long read, but if you fancy understanding MoarVM capabilities better, it is highly recommended reading indeed!

Building a Single Page Application with Cro

This seem to have slipped through the cracks the past weeks. Jonathan Worthington describes how to set up a simple Single Page Application using Cro as the backend, and webpack, ES6, React and Redux as the frontend (no prior knowledge needed). It shows a SPA for a (food/beer) festival where people could leave their tips about what’s hot and what’s not, and being able to see them in real time. It supports:

  • Submitting new tips (a POST to the backend).
  • Having the latest tips appear live (delivered over a web socket).
  • Being able to agree to disagree with a tip (also a POST).
  • Being able to see a list of the tips sorted most agreeable to most disagreeable (obtained by a GET).

A must read if you are looking into building fully interactive web-applications, specifically using all of the asynchronous functionality that Rakudo Perl 6 offers today!

Other Core Developments

  • Although not strictly a core feature, but in the eyes of yours truly as important to Rakudo Perl 6 as Rails was to Ruby: a new release of Cro (version 0.7.1). New features are: cro web starting a web interface for stubbing, running, tracing & viewing logs from services in development, the router having include/delegate support for composing routes. A large number of issues was fixed, especially in HTTP/2.0 support.
  • Brian S. Julin‘s Pull Request to support the $.^name syntax has been finally merged. I guess it got a bit Warnocked 😦
  • Elizabeth Mattijsen fixed the multi sub foo(:@c is copy) syntax, which apparently was overlooked during the GLR. She also made certain combinations of attribute initializations throw an X::NYI (Not Yet Implemented) exception with a workaround suggestion, rather than silently not initializing.
  • Samantha McVey sped up joining of large strings (like more than 300 graphemes) up to 10x.
  • Daniel Green implemented an nqp op to quickly create an Int from another Int, but without any mixins. He also sped up various types of interpolation into regular expressions.
  • Timo Paulssen improved the “no such attribute” error on a failing nqp::getattr. This has helped yours truly already a lot in the past week!
  • And many more smaller fixes and improvements!.

Other Blog Posts

Meanwhile on Twitter

Meanwhile on StackOverflow

Meanwhile on perl6-users

Winding down

Not a small Perl 6 Weekly again this week. It’s almost become a dayjob! But a fun one! So keep all the new stuff coming so yours truly can report them again in the next issue of the Perl 6 Weekly. See you then!

Advertisements

Got something to note?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s