Win a copy of Machine Learning with R: Expert techniques for predictive modeling this week in the Artificial Intelligence and Machine Learning forum!

Karina Guenther

Ranch Hand
+ Follow
since Sep 16, 2005
Graduated From Metro State College with a BS in Computer & Management Science in 1986. Worked in Cobol and Mainframe programing until 2000 when I learned Java, Have steadily improved and expanded my skills in Java and moved into web development ever since.
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 Karina Guenther

I know I can do that. HOWEVER, I don't want to have to do that. Plus, I still have to change the jre in the build properties. I want it to automatically use the java that I have specified in Eclipse when I create the project. Or even in some settings file somewhere.

Ron McLeod wrote:Have you tried specifying the compiler version properties?

4 months ago
I'm trying to build a maven project in Eclipse using the "create a simple project / skip archetype". I have JDK 1.8xx, JDK 11 and jdk 12. In eclipse I have it pointed to jdk 12 for my installed jre.

java_home is set to: java 12
Maven is 3.6

Right now when I create a empty simple maven project it defaults to java 1.5. I know I can change that in the pom, but I want to have it default to at least java 8 when It is created, so I don't have to do that.
How do I do that? I've not really found an answer any place. It it even possible?
4 months ago
Eclipse IDE has a plugin called EGit which works to sync with github, gitlab or bitbucket.

you may also require to setup ssh keys.
Ok, I thought at 1st that didn't answer my question at all and then I looked at the path in the error msg that you pointed out and realized that it was a combination of 2 directories. So I looked at my preferences in Eclipse and realized that I was using a full path for both the home of my ssh keys and the key file. When I removed the path for the key file, it worked:

So, in Eclipse: Windows/Preferences/NetworkConnections/SSH2 on the General tab: in the SSH2 Home the SSH2 directory: C:\Users\Owner\.ssh
and in Private Keys instead of C:\Users\Owner\.ssh\id_rsaFoxWork reduce it to just id_rsaFoxWork.

Tim Holloway wrote:I use an http git server, so I'm less familiar with git under ssh and especially ssh in Windows, but this:

Looks like a mangled version of a more traditional ssh command:

Allowing for rhe fact that the "C:\Users\Owner" would actually be a remote login ID, not a Windows directory path. And the C:\Users\Owner\mykeyRSA_id would be the public key you were given to access the server without having to login via userid/password.

Check your remote configuration under the "Team" submenu for the project.

6 months ago
I've got eclipse 2018 and upgraded to 2019 and suddenly I cannot connect to my repository.
the error I'm getting is:

What is going on?
6 months ago

Junilu Lacar wrote:I'm not sure how Groovy/Spock is considered to be not Object-Oriented. Groovy is in fact more object-oriented. The only minus will be that you'll need to learn a somewhat new language/syntax with Spock and Groovy. It's also a JVM language so you can test Java code and you have access to all standard Java library classes.

Blame my unfamiliarity with how they work and how I see them being used at my new job. Is there a good place to go for how they should work?
6 months ago

Rob Spoor wrote:Did you check JUnit 4 or 5? Because it got a major overhaul with a lot more possibilities.

That's what I read somewhere that it's closer to testng. I've not really studied it. Does the overhaul make it BETTER than TESTNG? If not then I should stick with what I know or go with Spock.
6 months ago
I'm trying to decide between using testng (which I am familiar with) junit (which I used to know) and spock  which i know little about.
Based upon my research it appears that you use testng / junit for robust testing with Testng having a few features that Junit lacks (but they are coming closer to testng).
Spock utilizes Groovy for more graphical results, but based upon the architecture at that I'm seeing at my new job doesn't lend itself to Object orientation. Could someone summarize the plus and minus of each one? Oh yeah I'm also unfamiliar with groovy.

So, could someone with familiarity of both spock and at least one of the others tell me how  / when you would use each? What are the strengths and weaknesses of each? How hard / easy would it to build an architecture that can be shared among projects for each?

I'm leaning toward testng, but I like spocks gui reporting. I  know that with testng you can export test results to other applications or databases, i don't know that about spock or how to do it.
6 months ago
all of a sudden in my Eclipse when adding a new file it's not being detected as a change for gitlab. it's not in the git / repository view even after i've added and changed the code. in terminal view for the project I've done git ls-file and it doesn't exist

for example:

but when i do the git ls-files command it can't find it:

So, what is going wrong and how do i fix it?
i've already tried in Eclipse doing "add to index" for the file.
1 year ago
I have a SSH object that opens a socket connection and submits commands. with the md5sum tool I don't need my app to calc it, I can just do: md5sum myfile.dat > myfile.md5.
thanks, I didn't know that there was md5sum there until I checked for it.

I have an application that connects via a socket to a unix box and uses ssh commands to copy / move files and do some other shell commands.
However, I need to read one of these flat files, add to it or create a new one and then create an md5 checksum file for it.
What is the best way? it seems as if copying the remote file to where my app is running, do what I need to it, then creating the md5 checksum file and then copying them back down seems tedious.
I could use shell script commands like cat or echo to create / write the data I need to the 1st file, but how about reading it and then calculating the checksum and creating that file?

Is there a better way where I could do all that on the remote box?
Once you feel you have some of the basics of the language, I'd also recommend studying design patterns as that will help you learn how create flexible code. Head First is a good one for that. It will also give you additional practice in Java.
3 years ago
My team has setup several Jenkins boxes using the selenium grid. These include Windows XP, Windows 7 and Win 8. While some tests care about which version of windows they are running on...and the browser and the browser version, most do not. Is there a way I can setup my selenium nodes to be specific but generic as well? I've tried the following but the console returned that it had a connection problem. So, is there a way to make this work? Or can it be only one thing?

4 years ago
Check your auto-commit setting in either the database or in your connection settings. Frankly, I would just add a commit statement after you've done all the processing you need to do rather than change the auto-commit. It leaves you with more control over what happens when.
4 years ago
if you are using selenium grid, there is a rest call you can do to kill the ie driver.

I'm looking if there is something similar for chromedriver as I'm having problems with that not stopping when the browser has been closed unexpectedly.
4 years ago