I boiled the cases where the compilation triggered in the code server stalls 
(see my previous email) down to three errors (the other dozen or so I was able 
to fix), but what do these last error messages try to convey to me?
 
[INFO]             [ERROR] Line 193: Rebind result 
'ch.zh.ksta.zhstregisterjp.client.ui.components.ZHStRegisterJPThreeListField.ZHStRegisterJPThreeListFieldAppearance'
 must be a class
[INFO]             [ERROR] Line 141: Rebind result 
'com.sencha.gxt.data.shared.LabelProvider' must be a class
 
What is this «rebinding» about? Is there some explanation somewhere? And why 
does it fail here? Esp. the latter error bothers me since that is in some 
library we use and I can’t change that.
 
And the third one:
 
[INFO]             [ERROR] Line 85: Only string constants may be used as 
property name in System.getProperty()
 
The code causing the above is from a spring library:
 
    public static boolean getBoolean(final String name) {
        boolean result = false;
        try {
            result = Boolean.parseBoolean(System.getProperty(name));
        } catch (IllegalArgumentException | NullPointerException e) {
        }
        return result;
    }
 
Does that REALLY mean that GWT can’t handle a System.getProperty(…)-calls with 
an argument that’s passed into the method? That seems completely brain-damaged 
to me. Why is that so???
 
 
 
From: [email protected] 
<mailto:[email protected]>  
<[email protected] 
<mailto:[email protected]> > On Behalf Of Michael Moser
Sent: Friday, February 10, 2023 2:08 PM
To: [email protected] 
<mailto:[email protected]> 
Subject: Can one convince code server to continue with compilation even if 
certain source files are missing?
 
Hi all
I am trying to set my GWT development environment up such that I deploy to a 
local Tomcat (v8.5 in my case) from my IDE (Eclipse and IntelliJ – I am setting 
this up for an entire team) and then start CodeServer to supply the JS code, 
the code maps and what not.
 
I managed to get things starting up, the application gets deployed to Tomcat 
from the IDE and then I manually start the code server via mvn gtw:codeServer 
in a command window. Note: I am NOT using any GWT plugin, neither Eclipse’s nor 
IntelliJ’s. Maybe I will give them another  try later, but in my last 
installation neither worked.
 
I can then connect to the WebPage, I get the login-form and the initial webpage 
and the code-server starts compiling the JS-code. So far so good.
 
However, then the compilation in the code server always fails due to some 
missing source files. I analyzed the errors and there indeed bugs in our code. 
Earlier developers added code that should be server-side only into shared DTO 
classes ☹. I will need to re-arrange and fix that but I don’t have the time 
right now.
 
The offending methods are never used on the client side and thus apparently 
they are ignored and do no harm. Amazingly the GWT compiler – when compiling 
the application for the version that gets packed up as war file – ignores those 
issues and continues with the compilation and the deployed application then 
runs just fine.
 
But when the very same application is compiled in code-server the compilation 
fails:

…
[INFO]       Compiling module ch.zh.ksta.zhstregisterjp.ZHStRegisterJPWeb
[INFO]          Ignored 8 units with compilation errors in first pass.
[INFO] Compile with -strict or with -logLevel set to TRACE or DEBUG to see all 
errors.
[INFO]          [ERROR] Errors in 
'file:/D:/Projects/KStA_ZH_RegisterJP/code/application/zhstregisterjp-common/src/main/java/ch/zh/ksta/zhstregisterjp/shared/security/SecurityUtils.java'
[INFO]             [ERROR] Line 36: No source code is available for type 
org.springframework.security.core.context.SecurityContextHolder; did you forget 
to inherit a required module?
…
[INFO]       [ERROR] Compiler returned false
[INFO]       [WARN] recompile failed
[INFO]       [WARN] continuing to serve previous version


Since there *is no* previous version that’s it and the application then 
hangs/crashes.
It almost seems as if code server compiles in strict mode here, even though I 
explicitly have specified <failOnError>false</failOnError> in the plugin config.
Can one somehow “fix” (or rather circumvent) this behavior?
 
I need to be able to run and debug this application soon and have to postpone 
the cleanup to later. Is there a way to get the code server ignoring those 
errors and continue compiling that code – as the GWT-compiler apparently does 
when generating the code for the .war file?
 
Regards,
Michael
 
 
-- 
You received this message because you are subscribed to the Google Groups "GWT 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected] 
<mailto:[email protected]> .
To view this discussion on the web visit 
https://groups.google.com/d/msgid/google-web-toolkit/00fe01d93d50%24a9751180%24fc5f3480%24%40gmail.com
 
<https://groups.google.com/d/msgid/google-web-toolkit/00fe01d93d50%24a9751180%24fc5f3480%24%40gmail.com?utm_medium=email&utm_source=footer>
 .

-- 
You received this message because you are subscribed to the Google Groups "GWT 
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/google-web-toolkit/004001d93d6a%24cf35fd70%246da1f850%24%40gmail.com.

Reply via email to