ANSI Encoding with Umlauts leads to compile errors and javadoc errors about jollyday HOT 8 CLOSED

Hey @HoffmannTom,
thanks for your issue. I tried to recreate this here https://github.com/focus-shift/jollyday/runs/6684578321?check_suite_focus=true based in windows and java 17. I cannot see any of these errors. Could you take a look at it? Maybe something differs.

from jollyday.

hm... I updated to the same java version and also used git clone instead of downloading the zip file.
It looks like there are different default charactersets used on github.
Maybe it would be better to specify the charset of the source files via the parameter "-encoding" and thus being independent of the platform default.
javadoc
I also see that the default charset differs by OS and OS language:
https://stackoverflow.com/questions/9312816/platforms-default-charset-on-different-platforms
On my system, UTF-8 seems to be the default but the java files are ANSI encoded.

from jollyday.

Btw... the same problem occurs at compiling the java sources:

[INFO] --- maven-compiler-plugin:3.10.1:compile (default-compile) @ jollyday-jaxb ---
[INFO] Changes detected - recompiling the module!
[INFO] Compiling 44 source files to E:\EBS\jollyday\jollyday-jaxb\target\classes
[ERROR] /E:/EBS/jollyday/jollyday-jaxb/target/generated-sources/jaxb/de/focus_shift/jaxb/mapping/EthiopianOrthodoxHoliday.java:[4,4] unmappable character (0xC4) for encoding UTF-8
[ERROR] /E:/EBS/jollyday/jollyday-jaxb/target/generated-sources/jaxb/de/focus_shift/jaxb/mapping/EthiopianOrthodoxHoliday.java:[18,20] unmappable character (0xFC) for encoding UTF-8

Using the -encoding parameter would also help to get platform independent:
https://docs.oracle.com/javase/8/docs/technotes/tools/windows/javac.html
Thanks in advance!

from jollyday.

Hello @derTobsch
short update:
I changed the pom.xml with:
<project.build.sourceEncoding>Cp1252</project.build.sourceEncoding>
It compiles successfully after this change.

Currently there is UTF-8 configured.
When I clone the project, the Umlauts are ANSI encoded (single-byte) but UTF-8 is multibyte.
Could you check, whether your sources are UTF-8 or ANSI encoded?
Somehow there is currently a mismatch when I clone the repo.

Thanks!

from jollyday.

Intellij says that everything is UTF-8 encoded. By builds are fine. This is also the preferred way to configure it, see https://maven.apache.org/plugins/maven-resources-plugin/examples/encoding.html. Need some time to look at it. Maybe you can take a look at your IDE? Maybe there is something wrong? Or do you build it on a shell?

@HoffmannTom can you check your git encoding? Maybe there is something configured.

I also provided two branches. Please clone the repo for each branch again and try to build #112 and #111

from jollyday.

Hello @derTobsch ,
I just saw that all files with encoding issues are in the subfolders of "jollyday-jaxb/target/generated-sources/"
So the generation of these files seems to happen with the wrong encoding in my case.
Any option to configure the encoding to utf-8 ?

from jollyday.

I think I got it
I changed the pom.xml within the jaxb folder:

              <jvmArguments>
                <jvmArgument>-Xms32m</jvmArgument>
		<jvmArgument>-Dfile.encoding=UTF-8</jvmArgument>
              </jvmArguments>

This worked for me.

from jollyday.

I think I got it I changed the pom.xml within the jaxb folder:

              <jvmArguments>
                <jvmArgument>-Xms32m</jvmArgument>
		<jvmArgument>-Dfile.encoding=UTF-8</jvmArgument>
              </jvmArguments>

This worked for me.

That sounds good. If you want to provide a pr, please feel welcome to provide one. I would be happy to see the first pr from another person ☺️

from jollyday.