Bug #42553 for Font-TTF: maxp version number parsed incorrectly

Tue Jan 20 04:52:28 2009 pjt47 [...] cam.ac.uk - Ticket created

Subject:	maxp version number parsed incorrectly
Date:	Tue, 20 Jan 2009 09:52:02 +0000
To:	bug-Font-TTF [...] rt.cpan.org
From:	Philip Taylor <pjt47 [...] cam.ac.uk>

http://www.microsoft.com/typography/otspec/otff.htm says: "When a Fixed number is used as a version, the upper 16 bits comprise a major version number, and the lower 16 bits a minor. ... For example, the version number of 'maxp' table version 0.5 is 0x00005000." Font::TTF::Maxp does not take this into account. 0x00005000 is parsed as 0.3125 (i.e. 5/16), and then compared against 0.5, which is wrong and breaks. The same issue applies when writing the version number out again. I haven't checked if any other tables have fractional version numbers that will break in the same way. -- Philip Taylor pjt47@cam.ac.uk

Mon Jan 26 03:11:22 2009 martin_hosken [...] sil.org - Correspondence added

Subject:	Re: [rt.cpan.org #42553] maxp version number parsed incorrectly
Date:	Mon, 26 Jan 2009 15:09:48 +0700
To:	bug-Font-TTF [...] rt.cpan.org
From:	Martin Hosken <martin_hosken [...] sil.org>

Dear Philip, Thank you for all these bug reports, it's great to nail them. Fixed by adding a new TTF_unpack('v', $dat) type parsing. Can't say I'm entirely sure about the FontRevision in the head table, but I think this is still of type 'f'. I'll get a new version out with these fixes shortly. If you want to monitor the pre-release code there is svn available at http://scripts.sil.org/svn-public/utilities/Font-TTF/trunk On Tue, 20 Jan 2009 04:52:29 -0500 "Philip Taylor via RT" <bug-Font-TTF@rt.cpan.org> wrote: Show quoted text

> Tue Jan 20 04:52:28 2009: Request 42553 was acted upon. > Transaction: Ticket created by pjt47@cam.ac.uk > Queue: Font-TTF > Subject: maxp version number parsed incorrectly > Broken in: (no value) > Severity: (no value) > Owner: Nobody > Requestors: pjt47@cam.ac.uk > Status: new > Ticket <URL: http://rt.cpan.org/Ticket/Display.html?id=42553 > > > > http://www.microsoft.com/typography/otspec/otff.htm says: "When a Fixed > number is used as a version, the upper 16 bits comprise a major version > number, and the lower 16 bits a minor. ... For example, the version > number of 'maxp' table version 0.5 is 0x00005000." > > Font::TTF::Maxp does not take this into account. 0x00005000 is parsed as > 0.3125 (i.e. 5/16), and then compared against 0.5, which is wrong and > breaks. The same issue applies when writing the version number out again. > > I haven't checked if any other tables have fractional version numbers > that will break in the same way. > > -- > Philip Taylor > pjt47@cam.ac.uk >

Mon Jan 26 03:11:24 2009 The RT System itself - Status changed from 'new' to 'open'

Mon Jan 26 04:16:29 2009 martin_hosken [...] sil.org - Correspondence added

Subject:	Re: [rt.cpan.org #42553] maxp version number parsed incorrectly
Date:	Mon, 26 Jan 2009 16:15:29 +0700
To:	bug-Font-TTF [...] rt.cpan.org
From:	Martin Hosken <martin_hosken [...] sil.org>

Fixed in 0.46

Mon Jan 26 05:35:00 2009 pjt47 [...] cam.ac.uk - Correspondence added

Subject:	Re: [rt.cpan.org #42553] maxp version number parsed incorrectly
Date:	Mon, 26 Jan 2009 09:47:48 +0000
To:	bug-Font-TTF [...] rt.cpan.org
From:	Philip Taylor <pjt47 [...] cam.ac.uk>

martin_hosken@sil.org via RT wrote: Show quoted text

> <URL: http://rt.cpan.org/Ticket/Display.html?id=42553 > > > Dear Philip, > > Thank you for all these bug reports, it's great to nail them.

Thanks for the quick response! Show quoted text

> Fixed by adding a new TTF_unpack('v', $dat) type parsing.

Minor nitpicking: 'eval("0x$frac")' looks a bit nasty - 'hex($frac)' would be safer and more efficient. Show quoted text

> I'll get a new version out with these fixes shortly.

In case it's relevant for release planning, I have a few more bug reports I intend to file within the next few days (hopefully today if I get time). -- Philip Taylor pjt47@cam.ac.uk

Tue Jan 07 20:38:32 2014 MHOSKEN [...] cpan.org - Correspondence added

On Mon Jan 26 05:35:00 2009, pjt47@cam.ac.uk wrote: Show quoted text

> martin_hosken@sil.org via RT wrote:

> > <URL: http://rt.cpan.org/Ticket/Display.html?id=42553 > > > > > Dear Philip, > > > > Thank you for all these bug reports, it's great to nail them.

> > Thanks for the quick response! >

> > Fixed by adding a new TTF_unpack('v', $dat) type parsing.

> > Minor nitpicking: 'eval("0x$frac")' looks a bit nasty - 'hex($frac)' > would be safer and more efficient. >

> > I'll get a new version out with these fixes shortly.

> > In case it's relevant for release planning, I have a few more bug > reports I intend to file within the next few days (hopefully today if I > get time). >

Tue Jan 07 20:38:32 2014 MHOSKEN [...] cpan.org - Status changed from 'open' to 'resolved'

Tue Jan 07 20:38:33 2014 MHOSKEN [...] cpan.org - Fixed in 0.47 added