SBCL and LangUtils - lisp

Has anyone got langutils working with sbcl?

Definitely, my email is eslick # the mit media lab. I have a deployed web application using langutils to stem text into Montezuma for full text indexing running on SBCL. I've been meaning to take another stab at packaging and releasing langutils; perhaps in a month or two. Love to have your feedback.

Why not ask Ian Eslick himself? He'd probably be quite interested in any problems you experienced with langutils and SBCL.

Related

Common Lisp; Paradigms of AI Programming examples not working

I got Paradigms of Artificial Intelligence Programming because I've read tons of good reviews about it, but every example have errors, I tried using SBCL and Lispworks on windows 7.
For example, I'm trying to write a scheme interpreter, this file has undefined functions, this file has undefined operators, and this file, there's something wrong with REQUIRE.
Google didn't help, it seems that I'm the only guy having these errors. The book is really amazing but I'm dying to see a working code. would you please copy/paste and see if it works for you? Is it only me who has these kind of errors?
Should i try a Clisp or something? Should i contact the author asking him to correct his code? but he is a director at google now, he won't reply to a stupid guy like me, therefore i'm coming here for help, you can't find lisp experts everywhere
Your problem is that you are treating each file as a stand-alone system. They are not independent, but build on each other.
You need to follow the instructions in the README file you link to instead of loading the individual files.

emacs: python auto-complete just doesn't want to work

I couldn't find a specific answer to my question here, so I dare asking it at the off-chance of being directed to the answer ;)
since about two days, I am trying to get the auto-complete-mode to work for python, but I am out of luck :( I have been following the direction from this website and from this website which are both fairly outdated it would appear.
To give a brief description for those unfamiliar: According to the websites, it is apparently necessary to install first rope, rope-mode, pymacs and ropemacs before attempting to install auto-complete. I am reasonably confident, that I have done so successfully (the given tests, if any, in each manual gave the expected result), but once I install auto-complete, it just doesn't, y'know, autocomplete...
Lastly, it is entirely possible that I have done a newbie mistake, so please don't be shy and ask the blatantly obvious question here.
Thanks a lot for your help.
I am using Ubuntu 12.04 and emacs23 and I would like to use it with python2.7
bzr branch lp:python-mode/components-python-mode
includes auto-completion based on company-mode
That's the development branch of python-mode.el, no release so far.

Favourite Colour: Myrtle

At work we use confluence.
On occasion it crashes, resulting in an error 500 page being generated.
This page includes some interesting reference information, including:
System Information:
favouriteColour: Myrtle
javaRuntime: Java(TM) SE Runtime Environment
jvmVersion: 1.0
operatingSystem: Linux 2.6.18-92.1.13.el5
...
Myrtle?
Many thoughts raged through my head. What's Myrtle? Why is it my favourite colour? Is it my favourite colour? Why does that particular tidbit of information require its own system property?
At first I assumed it was just something that someone at work had done. A remnant of a project long forgotten, an old April Fools joke perhaps?
It seems I was mistaken. In fact, even Atlassian acknowledges the colour, though they give no reason for its existence.
Now, I know what you're thinking? Who cares?
I do, gentle reader, I do. And you should too. It's little mysteries like this that make life worth living.
So, is there one among you who knows the secret of The Mysterious Myrtle uh.. Mystery? At least one inquiring mind wants to know..
Your Confluence System Favourite Colour (Australian spelling FTW) is also available from Admin -> System Info.
And if I told you any more, I'd have to kill you.
This looks like an obfuscation of their version or the error number.
You're not supposed to give your exact version away when selling corporate enterprise systems - it supposedly makes it easier for hackers to know what vulnerabilities to hit.
However they still want to be able to know what version you're on or what error id you got when you call for technical assistance, so they have a code word for each release.
I suppose the colours run out pretty quickly.
Maybe it's something incredibly fundamental to programming - like pi for math, 42 for the universe, or the L-unit for space travel (as we all know, without it, space travel is but the fevered dream of a madman).
We can only guess.
FWIW, I've frequently put "interesting" information in error dumps -- I've found it's easier to get people to report it in bug reports, and they're more likely to accurately report something like "myrtle" as opposed to "error #47" or whatever...

Can you review my Perl rewrite of Cucumber?

There is a team working on acceptance testing X11 GUI application in our company, and they created a monstrous acceptance testing framework that drives the GUI as well as running scenarios.
The framework is written using Perl 5, and scenario files look more like very complex Perl programs (thousands of lines long with procedural-programming style) than acceptance tests.
I recently learned Ruby's Cucumber, and generally have been using Ruby for quite a lot of time. But unfortunately I can't just shove Ruby to replace Perl because the people who are writing all of this don't know Ruby and it's quite certain that they wont want "this" kind of interruption.
So to bring Ruby's Cucumber a bit closer to their work, I rewrote it using Perl 5. Unfortunately I am really not a Perl programmer, and would love to get a code review and to hear suggestions from people who both know Perl and Cucumber.
Hi Perl/Cucumber StackOverflow users - please help me create this "open source" attempt to re-create Cucumber for Perl! I would love to hear your comments and will accept any acceptable help.
The minimal source code is here:
http://github.com/kesor/p5-cucumber
Thank you for your attention.
For those not familiar with cucumber - please take just one small moment to take a look at this one small little page: http://cukes.info/
You might get good help on Perlmonks. That site is geared toward this sort of post, whereas Stackoverflow is more about direct and answerable questions.
Good luck, :)
There's a rather fuller implementation of Cucumber on Perl in the form of Test::BDD::Cucumber.
Can you review my Perl rewrite of
Cucumber?
Answer: Yes.
I've read through the github code you posted. It looks pretty good.
I'm curious about the following snippet from Parser::Parser.yp::ERROR. Why did you chose to use "and do {}" over "if( ){ }"? Style? scoping? other?
exists $parser->YYData->{ERRMSG}
and do {
print $parser->YYData->{ERRMSG};
delete $parser->YYData->{ERRMSG};
return;
};
I haven't used cucumber and am still trying to wrap my head around the usage model. How would someone go about verifying/running your calculator example?
If you plan to release your code to CPAN (and you should), then you'll want to bundle this git source into an installable tarball. You'll want to include one of the module installing tools: ExtUtils::MakeMaker, Module::Build, etc. I've recently started using Dist::Zilla, and am really happy with it. It got me over the activation energy of releasing my first cpan module.
20 second tour of Dist::Zilla:
download and install Dist::Zilla
and its requirements (there are
many, as it's a developer only tool)
create a dist.ini file in the top
level of your package ( 'dzil new MyPackage' )
run
'dzil test', 'dzil build', and 'dzil
release',to test, (build and test),
(build, test and release)
See the Dist::Zilla::Tutorial
Does Test::A8N fit your requirement?
If not there are plenty of other Test:: modules on CPAN which may do what you're after.

Word suggestion

Duplicate of how does the google did you mean algorithm work
does anyone know how to write a word suggestion feature like Google's "did you mean?" feature? or know where I can find code samples.
Thanks
Peter Norvig (Google's head of research) has a simple one here which actually works pretty effectively: http://norvig.com/spell-correct.html
It's in python but is fairly simple to translate to other languages (note that there are some translations at the bottom of the page, but the C# version at least has a number of bugs in it and is very inefficient, so it would be worth doing the translation to whatever language you need yourself - it's also a good way to understand the code).