[Bug 533919] Review Request: mplus fonts - The M+ family of fonts designed by Coji Morishita

bugzilla at redhat.com bugzilla at redhat.com
Tue Jan 19 13:15:40 UTC 2010


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


https://bugzilla.redhat.com/show_bug.cgi?id=533919

--- Comment #6 from Igshaan Mesias <igshaan.mesias at gmail.com> 2010-01-19 08:15:33 EST ---
Hi,


Package has been updated to 028,

Here's the link to the new src, and spec file.
http://github.com/imesias/Mplus-fonts.git

Kindly review, and post feedback.

Thanks,
Igshaan


(In reply to comment #5)
> Hi Nicolas,
> 
> Upstream released a new version. I'm packaging that version and have completely
> re-written the spec file, though, still adhering to the points below.
> 
> > Anyway, the review:
> > 
> > 1. (not blocking) you don't include any fontconfig files, they would make your
> > user's life a lot easier (however, CJK is quite a mess in fontconfig, so maybe
> > it's better that way for now. Let's see what CJK users think of it)
> 
> I have yet to come up with a decent fontconfig, but for now I'll be omitting
> it. 
> 
> > 2. (not blocking) rpmlint complains at the licensing string. I know the MPlus
> > license has been approved by spot, so it's not a legal problem. However I fear
> > all the repo-checking scripts will complain at you because of that. Please open
> > a fedora rpmlint bug so the license gets added to its list. (however upstream
> > would have made our life easier by using a standard font license such as the
> > OFL or the GPL with font exception)
> 
> I have filed a bug against rpmlint, and it seems the issue has now been
> resolved ie, I no longer get rpmlint errors about the licensing at version
> 0.91.
> 
> > 3. (not blocking) rpmlint complains some of the lines you used in descriptions
> > are too long. Please break them in smaller (<80) bits
> > 
> > 4. (blocking) same problem for at least one for your summaries
> >
> 
> I've made sure all the lines in the summary and descriptions are less than 80
> characters.
> 
> 
> > 5. (not blocking) it seems rpmlint grew a spell-checker in fedora-devel. Please
> > fix the spelling mistakes it has identified (ie everything which is not a human
> > name)
> 
> For someone who's native language should be english, I hope I fixed all the
> spelling mistakes.
> 
> > 6. (blocking) please make sure your Source0 is a full URL pointing to the
> > source file. That will make your future packager life a lot easier (we have
> > many sf-hosted projects in Fedora, so it can be done)
> 
> I have placed a url pointing to the tarball in the Source tag.
> 
> > 7. (not blocking) you don't need to repeat the "Group:    User Interface/X"
> > line in Fedora releases with a recent rpm
> 
> I have removed the unnecessary Group tags from subpackages.
> 
> > 8. (not blocking) it would be nice if each sub-package had a different summary.
> > What is the difference between 1c and 2c ?
> the one is Type 1 the other is Type 2.
> 
> > 9. (blocking) m++ipa.pe is a build script, it has nothing to do in %doc
> 
> I am excluding the build script for now.
> 
> > 10. (not blocking) you don't need to specify %dir %{_fontdir} in the common
> > subpackage, it will be added automatically to each font subpackage
> 
> removed.
> 
> > 11. (not blocking) it's mostly cosmetic, but it would be nice if the font files
> > cased the style names they declare (ie "Bold" instead of "bold"). Apps are not
> > always smart enough to correct this
> 
> I will try and fix, with some help as I am still newbie when it comes to fonts.
> 
> > 12. (not blocking) a few font files declare an "heavy" weight. The standard
> > qualifier for "heavy" is "Black". That means weight selection won't work as
> > expected in apps not smart enough to try every possible "Black" legacy alias.
> > Please ask upstream to change this (also please have upstream check they did
> > mean "Black" and not some other weight). Standard style qualifiers and their
> > meaning have been described by Microsoft in the following whitepaper:
> > http://blogs.msdn.com/text/attachment/2249036.ashx
> 
> Should I speak to upstream about this?
> 
> > 13. (not blocking) the fonts fall just short of complete coverage for several
> > languages and unicode blocks (we only test if a language/block needs less than
> > 10 glyphs to be finished). Please relay upstream so it gets a chance to
> > complete them (for example, ab is incomplete because the Unicode consortium
> > added two glyphs to it this year IIRC)
> 
> AFAIK the font is not complete as of yet.
> 
> > 14. (not blocking) it would be nice if upstream included the text of their
> > license in the fonts copyright field, and not just "Copyright(c) 2009 M+ FONTS
> > PROJECT" without licensing info
> 
> I have spoke to upstream regarding this, he seems unwilling to change the
> license.
> 
> > 15. (not blocking) it would be nice if upstream released a source archive with
> > source files and build scripts, so we can re-build in Fedora using our own
> > fontforge version (that helps finding bugs in fontforge, and as our fontforge
> > improves, so do the fonts we build with it)
> 
> I have yet to ask upstream.
> 
> 
> Would you kindly confirm whether the above is OK before I post the links to the
> new srpms and spec files.
> 
> Thanks,
> Igshaan

-- 
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.



More information about the package-review mailing list