Chinaunix首页 | 论坛 | 博客
  • 博客访问: 592591
  • 博文数量: 207
  • 博客积分: 10128
  • 博客等级: 上将
  • 技术积分: 2440
  • 用 户 组: 普通用户
  • 注册时间: 2004-10-10 21:40
文章分类

全部博文(207)

文章存档

2009年(200)

2008年(7)

我的朋友

分类:

2009-04-29 14:26:57

Abstract

In this part of the newsletter we look into how to get started. Where to read the documentation, the spec-test. How they are integrated. We take a look at the IRC channel used by the Perl 6 developeres, the perl 6 evalbot and the pastebot used to share code snippets. We'll aslo check out what is the last element of an infinite list.


This entry was first sent out as part of the . Visit .

Welcome back to the Perl 6 Tricks and Treats

Yesterday was the second part of the Introduction to Perl 6 talk I gave in the Tel-Aviv Linux User Group. Unfortunately there were only 4 people including myself. I wonder if that reflects on the state of Perl 6, the level of the first part I gave two weeks earlier or the others were just lazy and planning their Passover Holidays.

Anyway, maybe because of the small number of participants the meeting was good, there were lots of good questions especially on why and how to start writing Perl 6. So let me address those questions here too.

How to get started with Perl 6

Currently Perl 6 is not yet in the state of easily writing in it. There are lots of features not yet implemented, some of them being quite important. So if you are planning to actually start writing Perl 6 code be prepared that you will bump into all kinds of obstacles.

Why write Perl 6 now?

You cannot expect your Perl 6 code to be used in production yet. It will take a lot more time to write anything in Perl 6 than in any of the languages you actually know, so why start writing Perl 6 now?

The reason is to start learning it now. The learning curve of Perl 5 is considered gentle but long. IMHO The learning curve of Perl 6 can be still gentle but it is way longer than that of Perl 5.

Just in the previous issue I wrote about the subroutines in Perl 6. In Perl 5 I can teach subroutines in 5-6 slides. In Perl 6 I think I would need more than 40 slides to cover it all. Obviously that would be too much for an introduction course. Anyway it just shows that you'll have to take your time learning Perl 6.

As there are not many examples or explanations on how to write Perl 6 yet, a lot of things you will need to learn by reading the specs. That can be hard as it is very dense. Then you go out to experiment and you might quickly find that things don't work as the spec says.

As someone who is just learning Perl 6 it is hard to tell when did you misunderstand the specs and when did you just bump into a bug in the implementation or a feature that has not been implemented yet?

Resources

It can be frustrating but there are a couple of very good resources to help you.

Specs and spec-tests

First of all the specs, the spec-test suit and their integration. Both the specs (the Synopsis documents) and the unit tests live in the Pugs Subversion repository. Using a web browser the specs can be read here: . and the spec-tests are here: organized according to the chapters of the specs which in turn follow the order of the .

Of course you can check then out using the svn command and read the specs using perldoc or your favorite text editor.

The best view though might be the integration of the two on . If you click on the word "Synopsis" in the first clickable column - for some reason all the links are called just like that - you'll reach the a HTML version of the relevant chapter from the specs integrated with the tests.

In every test file there are comments saying which part of the spec are they testing. These pages are build using those smart links so you can click on links such as this one:

    - Show t/spec/S02-literals/quoting.t lines 25\x{2013}58 (no results) -
    

and see the relevant tests.

So you can read the spec and then see examples. You can also see if those tests were successful or not or if they are being skipped. The test suite has comments such as this:

    #?rakudo skip 'nested curlies in q{...}' 

meaning that the next test - a call to ok() or is() or some similar testing function - or the next block of tests should be skipped on Rakudo.

That usually mean Rakudo does not understand that code, sometimes even cannot parse it.

In case you don't find an example or if you find the examples lacking this can be a good opportunity to help the Perl 6 development effort by writing more tests.

Before doing that the best would be to talk to some of core developers. For that the best way is to use IRC.

IRC

The Internet Relay Chat is one of the oldest online chat systems. In order to use it you need an IRC client, then you need to connect to an IRC server and in there to join a channel. A channel is like a room full of people talking at the same time. There are many desktop IRC clients but if you don't have one the best might be to start with a web based IRC client. I just found The Perl 6 people are using the irc.freenode.org server and the name of the channel is #perl6.

As I am writing this there are 134 people in the channel, apperantly most of them sleeping.

If you have an IRC client installed then should open it to the right place.

Otherwise you can use Mibbit: Go to the web site of and type in the relevant information by yourself or you can click on this link I prepared:

You will be asked to type in a nickname with a suggested random name already filled in then "Click to join chatroom". Once you are in you'll see people talking. Not only can you ask question here, but you can also try you code snippets. just type in

    rakudo: say "hello, my name is foo"

after a few second Rakudo will print its response. The channel is logged so you can look up previous discussions. For example, you can

mib_0gaypc is just the nick I was given when connecting using mibbit from the meeting yesterday. You can follow a few lines and see. One of the people in the class asked me what will happen if if we try to fetch the last element of an infinite list. In other words what will happen if we run this code:

     (1..Inf)[*-1]

So I tried. I was not afraid of crashing the evalbot that runs the code as I did that earlier that day and fixed the resource limit of Rakudo.

As you can see in that log, within a few second three heavyweights of the channel started to talk about our little experiment:

pmichaud, , the project leader of Rakudo,

masak, author of a large chunk of the good Perl 6 code out there including .

jnthn, who is, well, just him. He is one of the key developers of Parrot and Rakudo.

Within 10 lines we got to the point that Masak submitted a Rakudo bug report on a related issue.

So not only have we got an answer in a few seconds but we also helped a tiny bit the development of Perl 6.

Private IRC channel

Of course you might not want to have all your trials and errors on the #perl6 channel so you can open a private conversation with the p6eval bot. Just type in the following command:

     /msg p6eval rakudo: say "hello"

this will send a private message to the p6eval bot which will open a new conversation window - in the case of Mibbit it opens a new "tab" and sends you the answer. Once you switched to this tab it is enough to just type

   rakudo: command
   

to ask for the evaluation.

Perl 6 standard

The p6eval bot can also use other implementation and it can also use the definition of Perl 6 to see if your code is correct syntactically. This one uses the definition as Larry Wall writes it. So I type:

    std: command

And the p6eval bot will check if it is correct syntactically: Let's see two examples:

    : std: say 42
    : std 26098: OUTPUT<>

This shows the time it took to parse that code.

    : std: foreach my $x (1) {}
    : std 26098: OUTPUT<<##### PARSE FAILED ##### Syntax error (two terms in a row?) at /tmp/JTnE5W5NEt line 1:------> 
       ...

This shows that the above code (which is actually a perl 5 construct) cannot be parsed by Perl 6.

That's usually a good indication that the error is on our side.

Pastebot

IRC is a really nice way for quick communication but it is pretty much limited to single lines of code. What if you have a multiline snippet that you would like to show to others?

Pastebots are the tool for this. Basically a pastebot is a web page where you can paste some code snippet. Once you submit the form it will generate a unique page containing your code. You can send the address of this unique page to others.

Even better when the pastebot is connected to an IRC channel. Once you submit the form it will send the link along your chosen nickname and the one-line summary you wrote to the selected IRC channel. People can then click on those links and see your example immediately.

The Perl 6 people use .

Testing and contributing

Now let me return to the testing and documentation for a second. If you found that there are test cases missing from the test suite and if you don't have commit access to the Pugs repository yet then it is very much recommended to join the channel and ask how can you submit new test cases.

You can use the pastebot to show what did you write so the members on the channel will be able to evaluate your suggestion easily.

Mailing lists

There are also mailing lists that can be used but they are usually a lot slower in response time. Try sending a message to

    perl6-users-subscribe@perl.org

to subscribe to the perl6-users list.

The .

Getting Perl 6

I am using and recommending the Rakudo implementation of Perl 6 that runs on top of the Parrot Virtual Machine.

See the up-to-date instructions on how to get them on the

That's it for now.

Comments and Discussion

I am always open to comments and criticism (just have a positive spin to it :-) So if you find any issue with the examples, please don't hesitate to let me know.

If you'd like to ask question about Perl 6, probably the best would be to sign up on the Perl 6 users list by sending an e-mail to

    perl6-users-subscribe@perl.org

You can also look at the archive of the .

Previous issues of this newsletter can be found on my

阅读(843) | 评论(0) | 转发(0) |
给主人留下些什么吧!~~