[Bug java/19173] New: problems resolving classes

[Bug java/19173] New: problems resolving classes

[ora dot et dot labora at web dot de]

GCJ appears to have a problem with statement 'import <package>.*' 
in  case  all  files  to  be  imporated are given as compilation 
arguments. The very same works fine using SUN's javac compiler.

{* Upps, how can I upload  something here? Anyway, the whole sub-
ject has been investigated by Tom (tromey@redhat.com) and a small
demo project has been sent. *}

Here's a literal description of the demo project: This
are the project files:

 $ tar zxvf xxx.tgz
 xxx/
 xxx/impl/
 xxx/impl/Echo.java
 xxx/hello.java

This is the file's contents: 

================ hello.java =============================
package greeting;

import greeting.impl.*;

public class hello {
  static Echo myecho;
}
================ Echo.java ==============================
package greeting.impl;
public class Echo {}
========================================================

Now I'm going to compile the whole package in one shot:

 $ cd xxx
 javac -d . -C hello.java impl/Echo.java   # ok

=>> Works fine!

Remove files created by 'javac':

 $ rm -rf greeting

Try to compile the very same using gcj:

 $ gcj -d . -C hello.java impl/Echo.java 
 hello.java:3: error: Can't find default package `greeting.impl'. \
 Check the CLASSPATH environment variable and the access to the archives
 1 error

The problem appears to be related to 

 import hello.impl.*;

in hello.java. When changing to read

 import hello.impl.Echo ;

compilation using gcj works fine. Another option is to 
rename directory "xxx" into "greeting" and to use -I ..,
ie.

 $ cd ..
 $ mv xxx greeting
 $ cd greeting
 $ gcj -d . -I .. -C hello.java impl/Echo.java

-- 
           Summary: problems resolving classes
           Product: gcc
           Version: 3.4.3
            Status: UNCONFIRMED
          Severity: critical
          Priority: P2
         Component: java
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: ora dot et dot labora at web dot de
                CC: gcc-bugs at gcc dot gnu dot org,java-prs at gcc dot gnu
                    dot org,tromey at redhat dot com
 GCC build triplet: i686-pc-linux-gnu
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: i686-pc-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19173

[Bug java/19173] problems resolving classes

[ora dot et dot labora at web dot de]

------- Additional Comments From ora dot et dot labora at web dot de  2004-12-28 11:05 -------
Created an attachment (id=7835)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=7835&action=view)
original demo project for bug 19173

Ahh, here we go. The initial bug reporting page should mention that an
attachment might be added  a f t e r  having commited the bug report. 

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19173

[Bug java/19173] problems resolving classes

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|critical                    |normal
           Keywords|                            |rejects-valid


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19173

[Bug java/19173] problems resolving classes

[phil at mkdoc dot com]

------- Additional Comments From phil at mkdoc dot com  2005-03-29 11:36 -------
This appears to be a duplicate of bug 10894.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |phil at mkdoc dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19173

[Bug java/19173] problems resolving classes

[ora dot et dot labora at web dot de]

------- Additional Comments From ora dot et dot labora at web dot de  2005-03-29 14:12 -------
(In reply to comment #2)
> This appears to be a duplicate of bug 10894.

I think you are correct - looks like I missed entry 19173. Nevertheless, I like
my entry because my entry is short, clear and precise. 

Will this bug ever be fixed?

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19173