[Gs-devel] RE: about OCF features

[Igor V. Melichev]

Dear Mr. Toshiya Suzuki, Mr. Hideyuki Suzuki,

> From:   Hideyuki Suzuki [hideyuki at sat.t.u-tokyo.ac.jp]
> Sent:   12 ?????  2001 ?. 08:42
> To:     mpsuzuki at hiroshima-u.ac.jp
> Cc:     igor at artifex.com; gs-devel at ghostscript.com
> Subject:        Re: [Gs-devel] RE: about OCF features

> I think Mr. Igor is talking about emulation of OCF with CID fonts,
> which is not implemented yet and allows us construct OCF dictionaries
> from CID fonts.  I agree that the support of OCF is necessary, but I
> don't think the emulation is necessary.

Right you are.

Loading CIDFont-CMap, CPSI converts some of them into OCF format,
i.e. emulates OCF with CID font. It has a list of fonts to be simulated.
Other CID fonts are being loaded as usual.

I believe that GS supports both CID fonts and OCF fonts
(except few features, such as rearranged fonts), but it doesn't
provide the emulation of OCF with CID, i.e. "OCF compatibility"
(the last term is used by Adobe in some manuals).

My question was whether do we need to implement the emulation.
Mr. Hideyuki Suzuki says that we don't. I'm glad to know this,
because it is complicated thing.

> From:   Hideyuki Suzuki [hideyuki at sat.t.u-tokyo.ac.jp]
> Sent:   12 ?????  2001 ?. 10:44
> To:     mpsuzuki at hiroshima-u.ac.jp
> Cc:     gs-devel at ghostscript.com; igor at artifex.com
> Subject:        Re: [Gs-devel] RE: about OCF features

> I guess support of OCF and CMap/CID are implemented separately from each
> other.  I think "OCF compatibility" Mr. Igor V. Melichev is talking about
> is conversion of structure from CID fonts into the OCF structure which is
> needed for support of old PS documents (which depend on the OCF structure)
> when only CID fonts are available.  In other words, with the feature,
> interpreters only with CID fonts can load the CID fonts as if there exist
> OCF fonts.

Right you are. I only can add that Adobe's CSL includes
both CMap/CID and "OCF compatibility". If you have CPSI with
CJK font support, you can find CIDInitC and CIDInitN procsets in there.
'N' means 'native', 'C' means 'compatible', i.e. "OCF compatibility".

> From:   mpsuzuki at hiroshima-u.ac.jp
> Sent:   12 ?????  2001 ?. 09:46
> To:     hideyuki at sat.t.u-tokyo.ac.jp
> Cc:     gs-devel at ghostscript.com; igor at artifex.com;
mpsuzuki at hiroshima-u.ac.jp
> Subject:        Re: [Gs-devel] RE: about OCF features

> Hmm, the OCF on PS Level 3 is implemented as
> an emulation by CMap/CID operators?

OCF compatibility is not level 3 feature.
It appeared roughly in same time as CSL did.

> From:   mpsuzuki at hiroshima-u.ac.jp
> Sent:   12 ?????  2001 ?. 09:18
> To:     igor at artifex.com
> Cc:     gs-devel at ghostscript.com; mpsuzuki at hiroshima-u.ac.jp
> Subject:        Re: [Gs-devel] RE: about OCF features

> Thanks for your detailed comment on our patch. I will refine it.

OK.

> I've ever found CSL files in recent PS priner's resource
> (I'm sure on 2015 - supporting CID natively, and possibly
> on 30xx), but I couldn't check whether it's used /or not.

Check for CIDInitC and CIDInitN in Resource/ProcSet/ .

> Possibly, most modern softwares are independent on the font
> format, and not designed to use OCF.

This is good news.

> Please let me ask - will GS stop the support of OCF?

No, we don't suppose to stop OCF support.
I asked about a different thing. I believe that now we clarified this.
Ask more questions, if not.

Igor.