manoj r patil

Well, the basic difference between JavaScript/Python and Java is that the first one is interpreted language whereas the latter one is compiled.

So its very easy to add support for interpreted languages out of box due to less complexity and JavaScript being the standard (not based on ECMAScript), you will find many browsers supporting it.

For java, one point in time (around 2003-09), people were making lot of interesting stuff in Applet which is HTML embedded client side java class. But then there was client dependency of installing JRE. So it was thick but bulky client. Now when JavaScript itself is being enriched with tons of frameworks like jQuery, AngularJS, Node3JS, etc. even Adobe flash is in the downtrend.

Also interesting person stuff can have many diffent connotations... for example, i keep doing many personal things in Excel!

About learning, JavaScript has derived much of its syntax from Java (and its name also!), so for Java person, it should be very easy to learn JavaScript and eventually you will understand the subtle differences.

All the best for your stint!

Dear all,

Here is a great "hands on" book Pentaho for Big Data Analytics talking about big data solving its problems using Pentaho's tools.

I am running a giveaway contest for this book and you can visit my blog and enter the contest to win a free copy of this book. You can leave your comments on what do you feel about this book at the bottom.

Thanks for reading and wishing you most and more.

-manoj

If your both applications are independent of each other, I don't think you can do this implicitly.

One approach to address your problem could be to persist the old state before making any changes from the second application and if you found it is failed, you can restore the old state from the first application. I believe your roll back trigger is generating in the first application here.

I have httpd running on SSL with load balancer configured with two JBoss app servers running behind. Now my incoming request from the external world would be on https:// layer whereas I want the internal redirection from httpd to jboss should happen on http://. If not, at least I should be able to use the same openSSL certificate which I am using for httpd.

Can I get some views on this proble? Any pointers would be more than welcome!

I think you have tailored the exception in the log. Can you please put the detailed stack trace of the exception which might give some hint?

In your remote interface, your method call is facade.sendStatus(starDate,endDate);

whereas in your log, its saying status = facade.sendStatus(startDate, endDate);

Is this typo or what? This may not be related to your actual issue but thought to point it out

I want to expose web service which is running on the app server through apache web server. In the current framework, I have web server redirecting requests to 2 app servers in a round robin fashion and am planning to deploy web service on both of them.

What is the best practice to expose web service in this typical environment? Any pointers are welcome.

I can ask the client to update the browser settings with proper reasoning (like add respective policy for some applet, etc.). But then can not really install anything on his machine.

You are absolutely right that over http, chances of transferring large file error-less are not very high.

I was knowing about applet option but not really java web start. Thanks Bill for pointing this out, now I can google on it.

Ulf Dittmer wrote:What is a "web client"?

Web client what I meant here is basically any web based page through which user wants to upload the file.

So it can be Safari running on iPad or IE running on windows ...

William Brogden wrote:I would certainly look into "resumable ftp" - google search found some interesting products.

Surely you want to avoid the frustration of trying to upload a GB file in the presence of interruptions.

Bill

This would be best choice when the client is desktop. But then in case of web client, we have lesser flexibility esp. when it is not RIA.

Ulf Dittmer wrote:What are you trying to achieve by doing this? In other words, how would a web service be better than straight HTTP file upload?

Actually I want to process this file and publish the results to our site. But then the interface should be in such a way that client should be able to push his data from the desktop application. ...so was wondering if web service could be better idea or how.

Hi,

I am currently analyzing on porting one file upload program to web service.

The file size can be upto 1GB and we want this process asynchronous. So the call can be really stateless.

Questions I have are:

Would RESTful service be better choice?

For REST, is Jursey API the best?

Hi Ranchers,

I was wondering if anyone has used Vertica database?

Actually we are using MySQL for different OLAP analysis and frequently runs ETL to push the data to PALO cube. Now are in a process to analyze different options for scalability and found Vertica benchmarks really impressive. But then I would like to know from some real user about its usefulness.

Any pointers are welcome.

I want file names and this will give me path along with file name like input/file1.CHA input/file2.CHA

Tim Holloway wrote:Just out of curiosity, was it necessary to be so complicated?

Try this instead:
for F in `ls input/*.CHA` do ...

I have one script which processes files one by one. I want to automate this to process all the files in one subdirectory. So I tried the following script buts its giving me error 'ls: invalid option -- 2'

#!/bin/bash FILES="ls input/*.CHA | cut -d '/' -f2" for f in `$FILES` do echo "converting $f file ..." cmd="./runimportcha.sh $f" # unrar each file echo $cmd >> $f.log echo mv input/$f done/ done

Any idea what am I doing wrong here?