jay vas

Ranch Hand
+ Follow
since Aug 30, 2005
Merit badge: grant badges
For More
Cows and Likes
Total received
In last 30 days
Total given
Total received
Received in last 30 days
Total given
Given in last 30 days
Forums and Threads
Scavenger Hunt
expand Ranch Hand Scavenger Hunt
expand Greenhorn Scavenger Hunt

Recent posts by jay vas

UPDATE : Well - that particular method appears , in a sense, to be "polymorphic",,, which is why I was confused.

In any case, we've made some more progress and almost have it working - with the exception of some fancy concurrent constructs and an "engine" map which
appears to be uncommented, we may have it working soon enough.

12 years ago
Hi guys :

We over at the RudolF world have been working on fixing and commenting a version of conways game of life, in clojure :


Its here :


To run it, you can clone the project and simply type

In particular, I'm not sure what this method does - or how to comment/doc it . Any thoughts ?
This is a pedagogical exersize, so feel free to ramble on if something interesting comes up.

Any hints or contributors ?

PS @Seancorfield - I added you as a colloborator on the github fork already in case your interested.
12 years ago
Hi guys :

Thanks to some help from the vibrant clojure community, namely Sean Corfield and Lee Hinnman, including some very useful conversations on java moose, the a publication on the RudolF project has been accepted to ITNG for this spring !

.... Just to remind , RudolF is the open-source, machine learning / data mining app (a clojur sandbox project to demonstrate the power of the language), which I've been soliciting contributors for..... https://github.com/jayunit100/RudolF .

Here is a link to a copy of it, currently on lee hinmans blog : http://writequit.org/papers/files/RudolF.pdf

Thanks Sean for your advice.

If any of you have thoughts for next steps, or are interested in being on a second, follow up publication to this one by contributing to the code base, let me know
12 years ago
Hi guys (btw thanks shawn for keeping the clojure thread alive !!!) :

I'm totally new to macros and extending clojure, so concrete examples would really help :

My problem : I have 3 types of unit tests - fail, succeed, and warning. However, default Leiningan test only seems to have 2 outputs : failures, and successes.
My happiness : I noted that the clojure.test package has a great API, and is really modular. Thus, I know my problem has a solution !
My new problem : How can I extend my unit tests to have 3 result types, and how can I integrate these result types with my build ?

In particular I want to know

1) How is success/failure defined/determined by clojure.test and how is this communicated to external programs (i.e. like jenkins) which automate clojure builds ??? We are planning on automating our builds, and I need to know how clojure conveys success/failure (for that matter, I don't even know how ANT defines/conveys success/failure, and I use ANT every day, so I have a long way to go as a build engineer) .

2) How can I define a "third" test result for tests that don't pass, but dont fail either (for example, in PHP, you can "skip" a unit test)??? I assume clojure might offer a flexible way of defining new test outputs. In particular I found a success? function uses a map.... But its not quite clear how it is used or extended to me.

Thanks !!!

12 years ago
Hi guys : for those of you that use hadoop --- do you manage your data directly ? Or do you just dump it all in hbase ?

- In general most people I hear of using hadoop are doing it to store millions or billions of records for map/reduce
- The Hadoop m/r api can directly read/write to hbase tables.
- It seems like storing results of m/r jobs in programatically created folders, rather than in a single, machine-managed map (like hbase) is an errand which might be prone to errors in subsequent read and data cleaning stages which might occur.
12 years ago
Thats a reasonable question. The answer is no, it won't cause errors. The reason ? Becuase inheritance means that the actual method on the super class are called. And those methods, of course, compile.

Hi guys : I got vimclojure working, and have been getting a message regarding "nailgun".

Do any of you guys use nailgun ?

Here is the message Im getting

Could not determine the Namespace of the file.

This might have different reasons. Please check, that the ng server
is running with the correct classpath and that the file does not contain
syntax errors. The interactive features will not be enabled, ie. the
keybindings will not be mapped.

Error executing Nail! (127)
12 years ago
Good idea... We will be migrating bioclojure to jan aerts hub soon, the stuff on there is just proof of concept....that biojava, lien, and clojure really do play very nicely together.

I see you have some pedagogical projects to.... Maybe we Can merge then into rudolf or vice versa if your interested.
12 years ago
Thanks Sean ....... Here is the URL. There is some pretty cool Clojure bioinformatics data visualization stuff in there now, and also, some Haskell for dealing with large sets of random data samples in sampling experiments.

12 years ago
Hi guys :

Any one interested in collaboratively working on a large, open-ended clojure-lab on git hub ?

I have trying to get a group of Clojure folks together to collaborate on a project on GitHub that will be both interesting, potentially usefull, and extremely educational.

The idea is that, as programmers learning LISP, we might be able to feed off of one anothers projects and examples -- project euler is simply not usefull or insightful enough. To understand Clojure - we need to be doing real things, in high volume : parsing jsons, connecting to databases, and solving domain-specific problems. If we individually do this in separate repos, then we reduce our learning throughput --- but if we share a repository full of serious, real-world clojure mini-apps and projects, we can begin to feed off of our collective knowledge....

To kick-start things, we (as bioinformaticians) created the BioClojure project on GitHub, and got some serious interest and learned alot. We got to sample the powerful java interop, leiningan for getting maven repositories, and many other niceties of the language, thanks to contributions from Lee Hinnman and others.

Now, to open this up to a broader group of Clojure developers, we are migrating it to a much broader audience (we hope), and naming the project RudolF (many of our day jobs percieve functional / dynamic languages in much the same way santa's reindeer once percieved Rudolph's funny red nose)....

So... If so ping me and we can start carving out some pedagogical projects. The next two days we will be adding alot of Clojure and Haskell stub code to the GitHub Rudolf repository..... Hope to see you guys there.
12 years ago
thats interesting. I guess that means that builders could potentially overwrite each others work . seems like a dangerous feature to implement.

In any case, I find that activating multiple builders (java and ant) , and then switching back to one (java) fixes issues. Maybe because ant needs to "set some things up" before the regular java builds will work....
So what do eclipse "builders" do ? (Attached a screenshot here....) Note that you can use the "ant" builder.... you can check it , or un check it. I usually leave it unchecked. But sometimes checking it helps to fix things for me.
"And whether is is "add class folder" or "add jar", the result is that the item is added to the classpath when the compiler is run. "

Can you elaborate on this? I'm not sure how eclipse compiles classes if you are using ant... Whats confusing is how it is that the eclipse is able to "use" the resources created by ant...? And more interesting, how those resources are intermingled with the niceties of the IDE like code-completion.
Thanks... I'm wondering how this relates to java builds in general though.... I can usually get eclipse to "work" by hacking it.... But I want to understand what it's doing at a more fundamental level.