2009.12.28 18:55 "Re: [Tiff] bug in tiff2pdf, when handling Lab images", by Sergius Bobrovsky

How can the white and black points be the same?

I'm not sure. But it's the result of [different] calculations in tiff2pdf code. Also, Acrobat accepts it (see attached file).

In any case, a pdf dictionary allowed to contain only Name - Value pairs. So, /WhitePoint name followed by two [0.9643 1.0000 0.8251] values is simply incorrect in terms of pdf syntax.

in other words, correct dictionary should be something as following:

/ColorSpace [/Lab <<
/WhitePoint [0.9643 1.0000 0.8251]
/BlackPoint [0.9643 1.0000 0.8251]
/Range [-127 127 -127 127]

>>>

>>> ]

it can be achieved by simply adding written += t2pWriteFile(output, (tdata_t)"/BlackPoint ", 12); before

X_W = 0.3457F; /* 0.3127F; */ /* D50, commented D65 */

On Mon, Dec 28, 2009 at 10:28 PM, Manlio Perillo <manlio.perillo@gmail.com> wrote:

>> -----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi.

I have found a bug in tiff2pdf, when embedding a TIFF images having a Lab color space.

I'm reporting the bug here, since on the libtiff web site

>>> (http://www.libtiff.org/bugs.html), the link to the bug traker is broken:

http://bugzilla.remotesensing.org/buglist.cgi?product=libtiff

By the way, the link to this mailing list is broken, too.

>>> The problem is with tiff2pdf writing two times the WhitePoint entry in

the Lab dictionary:

/ColorSpace [/Lab <<

/WhitePoint

[0.9643 1.0000 0.8251]

[0.9643 1.0000 0.8251]

> /Range [-127 127 -127 127]

>>>>

>>>> ]

>
>
>

>>> Regards  Manlio Perillo

iEYEARECAAYFAks43MgACgkQscQJ24LbaUSGBQCfRx/KllvGYAugYS8Y+B6It6oH 3KsAn07uuSp+2clGShZ+UaKq67ng5inw

=vMdm

-----END PGP SIGNATURE-----

--
Sergius Bobrovsky