[Bug 1115881] Review Request: scheme48 - Scheme48 Scheme implementation

[bugzilla]

https://bugzilla.redhat.com/show_bug.cgi?id=1115881

Jerry James  changed:

   What|Removed |Added

 Status|ASSIGNED|CLOSED
 Blocks||201449 (FE-DEADREVIEW)
 Resolution|--- |NOTABUG
  Flags|fedora-review?  |
Last Closed||2016-03-09 17:46:18




Referenced Bugs:

https://bugzilla.redhat.com/show_bug.cgi?id=201449
[Bug 201449] FE-DEADREVIEW -- Reviews stalled due to lack of submitter
response should be blocking this bug.
-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are always notified about changes to this product and component
___
package-review mailing list
package-review@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/package-review

[Bug 1115881] Review Request: scheme48 - Scheme48 Scheme implementation

[bugzilla]

https://bugzilla.redhat.com/show_bug.cgi?id=1115881

Matthias Runge  changed:

   What|Removed |Added

 CC||mru...@redhat.com
 Whiteboard|NotReady|NotReady, StalledSubmitter



-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are always notified about changes to this product and component
___
package-review mailing list
package-review@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/package-review

[Bug 1115881] Review Request: scheme48 - Scheme48 Scheme implementation

[bugzilla]

https://bugzilla.redhat.com/show_bug.cgi?id=1115881



--- Comment #2 from Jerry James  ---
Christopher, do you have any intention of continuing with this review?

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are always notified about changes to this product and component
___
package-review mailing list
package-review@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/package-review

[Bug 1115881] Review Request: scheme48 - Scheme48 Scheme implementation

[bugzilla]

https://bugzilla.redhat.com/show_bug.cgi?id=1115881

Christopher Meng i...@cicku.me changed:

   What|Removed |Added

 Whiteboard||NotReady



-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are always notified about changes to this product and component
___
package-review mailing list
package-review@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/package-review

[Bug 1115881] Review Request: scheme48 - Scheme48 Scheme implementation

[bugzilla]

https://bugzilla.redhat.com/show_bug.cgi?id=1115881

Jerry James loganje...@gmail.com changed:

   What|Removed |Added

 Status|NEW |ASSIGNED
 CC||loganje...@gmail.com
   Assignee|nob...@fedoraproject.org|loganje...@gmail.com
  Flags||fedora-review?



--- Comment #1 from Jerry James loganje...@gmail.com ---
I will ignore the warnings about -devel files being in a non-devel package,
since this really is a -devel package.

Issues:
1. I see warnings in the build log about breaking strict aliasing rules when
   compiling c/r6rs/ieee_bytevect.c.  This means that, to be safe, that file
   must be patched or compiled with -fno-strict-aliasing.  Otherwise, the
   optimizer may emit code that does not behave as expected.  (I have actually
   patched code that looks very much like this in other packages, and had the
   patches accepted upstream.  Let me know if you'd like me to produce a patch
   for this issue.)
2. Consider splitting the documentation out into a -doc subpackage.
3. Multiple files are installed into %{_includedir}, including some that are
   not .h files.  Consider moving those to a subdirectory, say
   /usr/include/scheme48.  (This would also require patching the scheme48
   compiler, of course.)
4. The license is identified as GPLv3+, but I do not think that is correct.
   Here is a breakdown of the various licenses in play, corresponding to the
   list in COPYING:
   - The main license is BSD (New BSD).
   - The bignum code is BSD (New BSD).
   - The code in c/free.c is MIT (Modern Style with sublicense).
   - The autoconf files, mkinstalldirs, INSTALL, and install-sh are not part
 of the binary distribution.  We can ignore them for license purposes.
   - c/srfi-27.c carries no license information that I can find.  This could
 be a problem, unless license information is available elsewhere.
   - Files in scheme/srfi:
 o Identified as part of scheme48, so BSD: packages.scm, srfi-2.scm,
   srfi-4.scm, srfi-5.scm, srfi-7.scm, srfi-13-check.scm,
   srfi-14-char-sets.scm, srfi-14-check.scm, srfi-17.scm,
   srfi-19-check.scm, srfi-39.scm, srfi-66.scm, srfi-74.scm, srfi-95.scm,
   srfi-95-check.scm, test-packages.scm
 o MIT (all Modern Style with sublicense): srfi-11.scm, srfi-14.scm,
   srfi-16.scm, srfi-19.scm, srfi-25.scm, srfi-27.scm, srfi-28.scm,
   srfi-40.scm, srfi-42.scm, srfi-45.scm, srfi-61.scm, srfi-63.scm,
   srfi-67.scm, srfi-71.scm, srfi-78.scm
 o Public Domain: srfi-26.scm, srfi-43.scm
 o srfi-1.scm: I don't know what to call this, maybe check with
   fedora-legal on the name to give it.
 o srfi-13.scm: MIT and BSD
 o srfi-14-base-char-sets.scm: generated file
 o srfi-37.scm: BSD
 o srfi-60.scm: MIT and MIT with advertising
   - scheme/sort code
 o Most are similar to srfi-1.scm; check with fedora-legal on the name.
 o Public Domain: sortp.scm, vbinsearch.scm
 o test-packages.scm: part of scheme48, so BSD
 o vhsort.scm, vqsort2.scm: open source, may need fedora-legal input on
   these ones, too
   - Basis Technology Corp. code, covered by main scheme48 license (BSD)
   - tex2page files: I don't see them in the distribution.

   Whew!  So I think the license needs to be BSD and MIT and MIT with
   advertising and Public Domain and whatever the unknown licenses are
   called, and a license for c/srfi-27.c needs to be identified.
5. The directory %{_libdir}/scheme48-%{version} is owned by this package, but
   the directory %{_libdir}/scheme48 is not.  Are both really needed?  Can
   they be consolidated?
6. See the first item in the SHOULD section.  Maybe that Requires should be
   filtered?
7. Can a %check section be added?  I see a check target in Makefile.in.
8. Can the noarch parts of the package be split out into a noarch subpackage?
   See the note under EXTRA items below.
9. There are various warnings by rpmlint that should be addressed, namely:
   o The package name is repeated in the summary.
   o The word described is hyphenated in the description.  It's probably
 better not to do that, and let consuming packages figure out how to
 hyphenate words if the available horizontal space is insufficient.
   o Check the missing-call-to-setgroups warning on posix.so.  Is that a
 legitimate warning?
   o Some files are not UTF-8.
   o Check the wrong-script-interpreter and non-executable-script warnings.


Package Review
==

Legend:
[x] = Pass, [!] = Fail, [-] = Not applicable, [?] = Not evaluated


Issues:
===
- Large documentation must go in a -doc subpackage. Large could be size (~1MB)
  or number of files.
  Note: Documentation size is 2519040 bytes in 27 files.
  See: