[Bug 212045] Review Request: eclipse-emf - Eclipse Modeling Framework

[bugzilla at redhat.com]

Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug report.

Summary: Review Request: eclipse-emf - Eclipse Modeling Framework


https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=212045





------- Additional Comments From fitzsim at redhat.com  2006-10-25 11:51 EST -------
Some comments about the spec file:

The fedora and redhat macros don't seem to be needed.  Can't you just define
gcj_support to 0 or 1?  Similarly, the majmin macro is only used once, so it is
not needed.

The fetching instructions are fairly involved.  Does upstream not release
tarballs?  How do you know that build_200609210005 corresponds to the 2.2.1
sources?  If fetching from CVS is the only option (as I've heard it is for some
Eclipse projects) then why don't you include this fetching script as a source in
the SRPM?  That way it can be committed to FE CVS and used to create tarballs of
subsequent "releases".  (I tried to run the commented fetch script but the patch
sections failed.  Somehow the tabs in the original patch have been converted to
spaces in the comment, so patch can't locate the correct context.  Also, try the
'patch -p0 << "_EOF_"' form instead of manually escaping each $ character.)

Should the bootclasspath munging be done as a patch?  Or maybe it can be
eliminated altogether: if you specify bootclasspath, you need to make sure that
libgcj.jar or rt.jar is still on the resulting bootclasspath.  That is one
reason why you'd see "java.lang.Object cannot be found".

One other thing: I think license.html should be marked %doc.

I'll post the MUST/SHOULD checklist as a separate comment.


-- 
Configure bugmail: https://bugzilla.redhat.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.