Re: -classpath switch patch

[Nic Ferrier <nferrier at tapsellferrier dot co dot uk>]

Per Bothner <per@bothner.com> writes:

> I took a look at jcf-path.c, and one thing led to another. 
> Here is my patch to fix path processing.  It supports 
> --CLASSPATH as a synonym for --classpath; --bootclasspath, 
> and --wholeclasspath in place of the current --CLASSPATH. 
> I also fixed a logic problem, where jcf_path_init was 
> called *after* option processing, not before.  Finally, 
> I changed "." to not be part of the "system" or built-in 
> class, but as an default for the non-builtin path, following 
> what JDK does. 
>  
> One question I have is whether this is overkill.  Specifically, 
> is there any real reason to have --wholeclasspath?  I tend to 
> think not, unless there is some signicant difference between 
> --wholeclasspath PATH and --classpath PATH--bootclasspath "". 

I can't see any reason for --wholeclasspath, it just seems to be
another confusing switch.

However, without it the Makefile logic will have to be altered (as
opposed to just a search and replace of -CLASSPATH / -wholeclasspath).



Nic