Bug - Not A Bug Tables in the gCLI mutliplying

Bale

Minion
I'm using u55 as I just recommended and verify that it has no rendering bugs, so the problem was introduced in r60.
 
So the 1.7 version I am using is it work, and due to the fact that we use it as a plugin for web browsers, I can't really downgrade it from u65 to u55. If this is a java bug, did anyone report it? Or has the function in java changed for some reason?
 

Veracity

Developer
Staff member
I just installed both JDK 1.7.0_55 and 1.7.0_65 and I can switch between them at will.

Code:
bash-3.2$ setjdk 1.7.0_65
bash-3.2$ java -version
java version "1.7.0_65"
Java(TM) SE Runtime Environment (build 1.7.0_65-b17)
Java HotSpot(TM) 64-Bit Server VM (build 24.65-b04, mixed mode)
bash-3.2$ setjdk 1.7.0_55
bash-3.2$ java -version
java version "1.7.0_55"
Java(TM) SE Runtime Environment (build 1.7.0_55-b13)
Java HotSpot(TM) 64-Bit Server VM (build 24.55-b03, mixed mode)
I'll see if I can code up a standalone program which will generate an HTML table and display it the way KoLmafia displays tables in the gCLI. If it works correctly in version 55 and fails in version 65, I'll submit it with a Java bug report. Hopefully, that will give it a chance to be fixed, as opposed to saying "we have a large Java program with 100s of source files and we see this problem."
 

Veracity

Developer
Staff member
Got it.

Code:
bash-3.2$ setjdk 1.7.0_55
bash-3.2$ ant
Buildfile: /src/java-html/build.xml

init:

compile:

jar:

BUILD SUCCESSFUL
Total time: 0 seconds
bash-3.2$ java -jar java-html.jar
java-7u55.png


Code:
bash-3.2setjdk 1.7.0_65
bash-3.2$ ant
Buildfile: /src/java-html/build.xml

init:

compile:

jar:

BUILD SUCCESSFUL
Total time: 0 seconds
bash-3.2$ java -jar java-html.jar
java-7u65.png
 

Veracity

Developer
Staff member
I filed a bug report. I didn't submit my build file, since you can simply do the following:

Code:
bash-3.2$ javac Test.java
bash-3.2$ java Test
to compile and execute it; no .jar file is needed.
 

Darzil

Developer
I notice it has a comment asking if it also affects 8 or 9 (so clearly hasn't read the bit where it mentions 8). I know Veracity uses apple, but it affects Windows 7 also in the same way, which isn't noted.
 

Bale

Minion
On two different systems I have experienced it effecting both Windows XP and Windows 7. Yet they have this tagged as a macosx bug.

I'm not actually finding anyplace to create a account so that I can comment on this... I can find the login just fine, but I don't have an account.
 

Winterbay

Active member
I tried by creating an accoutn at java.net but apparetnly that doesn't follow over to openjdk.java.net which is irritating.
 

Veracity

Developer
Staff member
Huh. When I submitted the bug report, they gave me a temporary number to refer to it by and said it'd be given a real number if it was reviewed and found to be legitimate and entered into the system. They didn't tell me when that happened.

I'm disappointed to see that the formatting was gone from my sample program.

When I filed the bug report, I had to choose an OS; there was no place to say "this happens on all platforms". Not that I could have confirmed that, but I think it will be clear to them that the affected code is not in a platform-specific package.
 

Bale

Minion
I just took a look at the progress java has made on this issue. This bug has been handed off to three different people so far. It was even marked "Closed, Won't Fix" for most of a day before it got reopened. No idea what that was about. Fortunately there is some good news:

The second person assigned to this finally identified the bug fix that caused the regression and passed it to the person responsible so he could correct his own error. That was yesterday!
 
Last edited:

Darzil

Developer
Now appears to be marked resolved, as it's considered a duplicate of an infinite loop that hasn't been fixed in the couple of months it's been in the system. I don't think I expect a quick fix!
 
Top