[Bug 226795] Review Request: sdcc - Small Device C Compiler

bugzilla at redhat.com bugzilla at redhat.com
Wed Feb 28 02:47:08 UTC 2007


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: sdcc - Small Device C Compiler


https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=226795





------- Additional Comments From rc040203 at freenet.de  2007-02-27 21:46 EST -------
(In reply to comment #27)
> Looks good,
> 
> A few last issues:
> 1) The changelog entry (and your last comment) about "Disable creation of 
>    debuginfo package" is plain wrong. sdcc contains native binaries (the 
>    compiler, linker et all). for which we want a debuginfo package, thus the
>    debuginfo is a good thing. The problem was that it was an empty package.
>    The changelog should read something like: "Disable stripping of binaries,
>    so that we get a proper debuginfo package"
Right, the changelog is wrong.

> 2) Remove the empty %doc from the "%files src"

> 3) The descripion of the -src subpackage is a bit vague, try explaning that
>    these are the actual sources of the c-library for the devices and that these
>    sources are meant for reference of how the c-library works.
> 
> Also I see that you need a sponsor, that is not a problem I can sponsor 
> you, but before doing that I would like todo one more package review with 
> you, so can you submit another package for review and post the bugzilla 
> id here, then I'll reviw it and assuming that goes well then sponsor you.
I don't feel able to sponsor anybody, because the ACL issues disable me from
being able to fulfil the tasks I consider to a sponsor's obligations :(

> Ralf, do you agree with the modifications I've requested for the package? And
> what do you think of the sdcc-src subpackage?
Well, I'd not have requested a 'src' package, because I don't see any use for
the sources anyway, ... but this is an issue upstream should take care about.

In same boat, is this package shipping the a target's library's *.o's in
parallel to libraries (*.lib, *.a). Normally this doesn't make any sense, 
... but this is an issue upstream should take care about.

> Maybe sdcc-sources or
> sdcc-libc-sources is better?
Hmm, I'm not sure. sdcc-libc-sources sounds like the most "self-explanatory"
package name to me, but this is a matter of personal preference.

Technically, I see directory ownership issues between *-src and the main package
(IMO, *-src must require the main package).

Finally, I don't think the "BR: byacc" is right. It probably should be "bison".
AFAIS, the toplevel configure seems to be wanting to enforce bison, but seems to
fail on this.

-- 
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.




More information about the package-review mailing list