Improve javac error output parsing - Fix #336 #337
Conversation
...rs/plexus-compiler-javac/src/main/java/org/codehaus/plexus/compiler/javac/JavacCompiler.java
Show resolved
Hide resolved
|
It looks ok for me but unit or even IT with different locale should be added |
I am going to add more unit tests. IT coverage is so low, they are nothing more than smoke tests for now. To improve that is out of scope for this PR. BTW, it is somewhat difficult to reproduce internal compiler errors and annotation errors across JDKs 8-21, because those are the ones that get typically fixed in new update releases. I even had to install a very old version of JDK one (u45) to reproduce any compiler error that prompts users to raise a JDK bug. But I did, just to be sure. The unit tests, however, are meant to reproduce error outputs for both cases (compiler and annotation processor errors) in several locales. |
Instead of two separate 'while' loops, one inline and one in an extra
method, parsing stack traces after finding error messages specific to
the English locale of the JDK, now there is generic stack trace parsing,
which works in these two cases independent of locale and also in others
previously not covered.
For backward compatibility, the Javac localized message 'javac.msg.bug'
("An exception has occurred in the compiler ... Please file a bug") is
salvaged in a locale-independent way, matching on error reporting URLs
always occurring in all JDK locales (English, Japanese, Chinese, German
as of JDK 21).
Fixes codehaus-plexus#336.
|
@slawekjaranowski, done. Sorry, yesterday I was busy otherwise. |
|
Thanks for unit test ... is it possible to prepare a IT for it ... |
|
It is always possible to prepare an IT. But I already explained why I won't:
Moreover, in this comment in MCOMPILER-424 I also said, that I have already created an OpenJDK bug report for this particular problem. Chances are, that someone will fix it, and then the IT will no longer run, unless you make a complex setup with toolchains, installing specific older JDK update versions in which the bug still exists, even if in the next update of the same JDK it was fixed already. Why set up an IT with a half life like this, if you cannot rely on it still working in three months? If you want to do that and volunteer to maintain it, always finding new examples of these types of JDK bugs in the future, be my guest. |
slawekjaranowski
requested review from
olamy,
gnodet,
slawekjaranowski and
slachiewicz
slawekjaranowski
changed the title
Fixes #336 and generalises stack trace parsing.
@slawekjaranowski, all unit and integration tests are passing. If you generally like the implementation, I can add some more unit tests.