btaz

fatal: You have not concluded your merge. (MERGE_HEAD exists)

I got this message because when I performed a “git pull”. I searched for a solution for this problem on the Internet and it wasn’t until I found this post that I was able to resolve this issue. The problem was that I:

1. Performed a “git pull” and the automatic merge failed and I ended up with merge conflicts
2. I resolved the merge conflicts and added the resolved files back using “git add”
3. Performed a new “git pull” and got the “Fatal You Have not Concluded Your Merge MERGE_HEAD Exists” error

Apparently step 3 overrides MERGE_HEAD, starting a new merge with a dirty index. According to the post this is a common mistake made by programmers that are used to version control systems where the user follows an “update” and “commit” work flow.

So how do we resolve this issue? What worked for me was to follow the instructions for how to “Undo a merge or pull inside a dirty work tree” found here.

1. I used “git reset –merge ORIG_HEAD”
2. I resolved the merge conflicts again and added the resolved files back using “git add”
3. I was then finally able to “push” my changes!

According to the documentation if you run a “git reset –hard ORIG_HEAD” it will let you go back to where you were before you were trying to commit your changes, however you will lose local changes. Most likely not what you want to do. Using “git reset –merge” will let you keep your local changes. You will however have to re-resolve your conflicting merge files.

Some additional information on this topic can be found here.

I used the Oracle SQL Loader to push some data into a table and got the following error: SQL*Loader-522: lfiopn failed for file (loader.log)
This somewhat cryptic error message turned out to be that Oracle SQL Loader didn’t have write permissions in the work directory i.e. in the directory where I executed the sqlldr command. Once I fixed the directory permission everything worked just fine.

A handy list for Java data types and sizes

Integer Data Types

Floating Point Data Types

Boolean Data Types

Character Data Types

Default Values

References:
http://java.sun.com/docs/books/tutorial/java/nutsandbolts/datatypes.html

It’s very useful to be able to see what is included in your class path especially when you’re trying to track down a troublesome dependency. Recently I had an issue with a JDOM dependency; however I had not defined a JDOM dependency in my pom.xml file and I couldn’t quite figure out why this dependency existed. This is where Maven’s excellent dependency plugin comes to the rescue.

In a project you may have dependencies on libraries like Log4J or Freemarker. These libraries in their turn may have dependencies on other libraries, this is also known as transitive dependencies.

Using the following Maven command we’ll be able to see a list of these dependencies:

mvn dependency:resolve

For a nice tree view of the same information use:

mvn dependency:tree

If you want to see a full dependency trail that also show you artifacts that were rejected due to conflicts or other reasons then run maven with the debug flag enabled:

mvn install -X

For more information about the Maven dependency plugin go here: http://maven.apache.org/plugins/maven-dependency-plugin/

I wanted to control how fast SCP copy data between two Linux servers. On a Linux system this can easily be achieved using the “SCP -l” switch. In my case I had to make sure to not exceed 0.5 Mbit/s

Since the “-l” parameter uses Kbit/s the correct value in my case is 500

scp -l 500 big-file.tar someuser@someserver.com:backups/.

Logitech has yet to release updated keyboard and mice drivers for OS X 10.6 / Snow Leopard. I’m quite fond of my Logitech MX 510 mouse and wanted this problem fixed as soon as possible.

I used the instruction from the link below which tells you to download the “Logitech Control Center”, locate the “Logitech Control Center.mkpg” file inside the “Logitech Control Center Installer.app” and execute it. This will by-pass the OS X version check inside the Logitech installer.

I however wanted to use the installer for my mouse MX 510 which uses the “LCC Installer 2.app”. I followed the same instructions, located the “Logitech Control Center.mkpg” and executed it which successfully got the driver installed.

I found the solution to this problem here:
http://www.tuaw.com/2009/09/01/snow-leopard-get-your-logitech-mouse-and-keyboard-working-again/

I wanted to add the Spring 3.0 M4 milestone early access release to a Maven project to test out some of the new features. However Spring  milestone releases are not readily available in the standard public Maven repositories, so what do you do?

For the full explanation check out the link to “Spring/Maven Repositories” link in the bottom. For a quick 1-2-3 step solution add this to the <repositories> section in your Maven pom.xml file so that Maven can find Spring’s early access milestone releases:

<repository>
  <id>spring-milestone</id>
  <name>Spring Portfolio Milestone Repository</name>
  <url>http://s3.amazonaws.com/maven.springframework.org/milestone</url>
</repository>

<dependency>
  <groupId>org.springframework</groupId>
  <artifactId>spring-core</artifactId>
  <packaging>jar</packaging>
  <version>3.0.0.M4</version>
</dependency>

How do you automatically provide an answer to a Unix command for example

cp: overwrite `destination/./b.txt'?

Many Unix/Linux commands that operate on files may stop and ask for confirmation for each file before completing an action. Of course for many Unix/Linux commands there are parameters that allows you to specify the desired behavior, but there are other commands that doesn’t have that capability. Advanced Unix users just look up the “yes” command on how to solve this. Here’s a contrived example where a directory called “destination” contains two files “a.txt” and “b.txt”. We will now copy two files “b.txt” and “c.txt” from a directory called “source”

|-- destination
|   |-- a.txt
|   `-- b.txt
`-- source
 |-- b.txt
 `-- c.txt

We’ll copy the files using the “cp” command

cp source/*.txt destination/.

When we do this we’ll get the following response on the command line where the OS is asking us what it want us to do since one file being copied already exist in the destination directory i.e. “b.txt”

cp: overwrite `destination/./b.txt'?

Unfortunately on Linux the cp command doesn’t have an option to automatically answer this question with a ‘y’ or ‘n’ answer. There’s more than one solution to this problem depending on what you want to do, and one solution is to use the Unix “yes” command. This command will output a string repeatedly until killed. If we want to avoid overwriting any of the files in the destination directory we could use the “yes” command to answer all questions with a no “n”. Below you can see how the “yes” command is used to automatically provide the answer “n” to the “cp” command. Use “man yes” for more information about the “yes” command.

yes n | cp source/*.txt destination/.

This successfully copies all files with the exception of the “b.txt” file which is what we wanted. The “yes” command automatically provided the answer “n” to the “cp” command question whether or not we wanted to overwrite the destination file.

In the case that you mess up when you use the “yes” command for example by typing this:

yes bingo

The “yes” command will output the word “bingo” repeatedly to the screen. You can stop it by typing Control-C or in a different shell type “pgrep yes” which will display the “yes” commands PID (process ID) which you can use to kill the “yes” command by typing “kill <PID>”

pgrep yes

In my case it returned 24940 so I killed it with:

kill 24940

This is a somewhat confusing compilation failure that sometimes happen when you write unit tests using JUnit 4. This is an example code snippet that produces this error (result.getValue() returns an Integer object):

assertEquals(12345, result.getValue());

And when you try to compile your project it produces a compilation error like this:

/projects/myapp/src/test/java/org/myapp/MyTest.java:[88,8] reference to assertEquals is ambiguous, both method assertEquals(double,double) in org.junit.Assert and method assertEquals(java.lang.Object,java.lang.Object) in org.junit.Assert match

Since JUnit offer several methods that are very similar the compiler can not always determine which one to use. In our case we have the number 12345 which is an int and result.getValue() which is an Integer. You’d think that the compiler could figure out to convert the number 12345 to an Integer object which would give us a match on “assertEquals(java.lang.Object, java.lang.Object) method, but instead it throws the above error.

The reason for this is that Java uses best match to try to determine the method that will require the least conversion of the parameters. In this case one method requires boxing, and the other method would require un-boxing. Both of these methods have equal priority and that’s why there is ambiguity since none of the methods are more specific than the other.

The solution is to help the compiler determine what it need to do. We can do that using this:

assertEquals(12345, (int) result.getValue());

or this:

assertEquals((Integer)12345, result.getValue());

to resolve this problem. The second solution that converts 12345 to an Integer object is most likely the better once since Java will auto-box a null value to the number 0 which may not be the correct behavior for the class you’re testing.

If you need to define and end of line character (EOL) like \n in your Spring XML files in a bean property you can’t do this

<bean id="some-bean-id" class="some-class">
    <property name="my-property-name" value="\n"/>
</bean>

This simply wont work. You’ll end up with a backslash followed by the letter n. What you have to do is to use an escape character like this

<bean id="some-bean-id">
    <property name="my-property-name" value="&#10;"/>
</bean>