Does VOLT Help in Developing Arabic Fonts?!

Primary tabs

218 posts / 0 new
Last post
AzizMostafa's picture
Offline
Joined: 18 Apr 2006 - 10:43pm
Does VOLT Help in Developing Arabic Fonts?!
0

Peace be on All Friends,
In VOLT, Marks Positioning is carried out Independently for each letter and for each component of every ligature?!
So, to avoid the Collision of Marks with Marks and/or Letters in Arabic Fonts:
1 The thinest letter or ligature should be at least a bit wider than the widest mark. Just like asking 3 cowboys to stand side by side without allowing their Hats to collide. And this can be done at the expense of Beauty.
2 Marks should be horizontally centered on top of or below each letter and each component of every ligature. And this is not always possible even if Beauty is sacrificed.
Samples speak louder than words?!
Daringly

Nick Curtis's picture
Offline
Joined: 21 Apr 2005 - 8:16am
0

Karl? A link to INDD1? What a total disaster that was: it totally screwed up font-handling with every other Adobe program, and I had to uninstall it, then rebuild Illustrator and Photoshop to undo the damage.

No wonder I can't get any help with my spacer problem…although, IF I ever get the problem solved, I probably ought to look into a cloud-based subscription model. Why sell each set outright for $3, when I could charge users $1 each time they use a set? A perpetual money machine! SWEEEEEET!

Karl Stange's picture
Offline
Joined: 17 Sep 2009 - 10:07am
0

Karl? A link to INDD1?

Oops! Looking at fixing that now...

Karl Stange's picture
Offline
Joined: 17 Sep 2009 - 10:07am
0

I think it should be fixed now, I am hit and miss with the internal linking to posts, which could have something to do with an intake of red wine. As for InDesign 1, I only used it briefly and was transitioning from Quark 4.11, so there was a whole world of hurt involved.

erwindenissen's picture
Offline
Joined: 31 Aug 2007 - 1:12pm
0

VOLT fails to compile it with no good reason, so I'm afraid you've run into a bug (an internal limit). Basically it is now too complex. I hope you can make some changes (simplify things a bit) which allows you to successfully compile it again.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

I am work on Saleem Nastaleeq Font dot less How I Auto justified word kashida like سـ ،بـ I am work on dot less word bay_init+tattwel = بـ and seen_init+tattwel=سـ MS word is not present these words if you suggest any way get these position which feature is support.
Bay ب dotless put nuqta by mark positioning like ب، پ ، ت، ٹ ث nuqta placing is good but when I bay_init+tattwel = بـ is not auto present with justify. When I type Bay_init with tattwel type kashidaبـWhat Problem and some time ligature place nuqta file is not compiled failed message. For example glyph 1-100 between sequence forgot and add between forgot glyph place and then put nuqta file is not compiled what reason when I did not ad missing figure like 67 and other number of glyph, file is compile best way. When I ad missing glyph file is compile failed message. What happened. My font is near to ready. Mane problem is kashida word is not get how to get? Any one help me. Plz check my New Saleem Nastaleeq Font and give some fruit full suggestion I will be appreciate to you. Second thing is ligature nuqta less when I diacritic mark file is not compile and compiled failed massage. Plz give me support and very easy way to learn that I understand I wait your kindness.

Jan Żurawski's picture
Offline
Joined: 3 Sep 2005 - 12:21pm
0

I would try to do that in this way:
Save your VOLT project ( e.g. AVolt.vfp )
Remove your {R_L_6} group from mark feature. Try to compile - if it compile, save your project as AVolt_1.vfp.
Open again project AVolt.vfp. Remove all features but mark and remove all groups but {R_L_6}, {AA}, {BB}. Try to compile - if it compile, save your project as AVolt_2.vfp. If not find errors and repeare. When you have compiled AVolt_2.vfp project generate AVolt_1.ttf and AVolt_2.ttf files and extract feature files from them AVolt_1.fea and AVolt_2.fea. Merge these two .fea files and make your final font.

John Hudson's picture
Offline
Joined: 21 Dec 2002 - 11:00am
0

I believe the problem is that your cursive connection GPOS feature has the marks flag set to something other than NONE, probably all. Every OpenType Layout lookup has a flag that indicates whether all marks, no mark or a class of marks should be processed as part of the lookup. For features that involve interaction between adjacent letter shapes, such as ligature GSUB features and cursive attachment GPOS features, you want to set the marks flag to NONE. This means that marks in the glyph sequence will be ignored by this lookup. [Note that, obviously, this means that mark glyphs need to be classified as Marks in the VOLT glyph data.]

Karl Stange's picture
Offline
Joined: 17 Sep 2009 - 10:07am
0

Oldnick + Karl Stange

Tracking the Old(nick) sounds St(r)ange?!

I am not at all familiar with other forums but I like the way that relationships develop on this one, it feels like a local (pub) where everyone knows your name, or at least your handle. I have yet to decide whether that is a good or a bad thing, it generally feels good though.

Pushing this back to the topic (at least a little bit), are there any clearly established guides to using VOLT, independent of this and Microsoft's site?

John Hudson's picture
Offline
Joined: 21 Dec 2002 - 11:00am
0

Re. the lam_mim ligature:

I'm guessing that there are a couple of different situations in which you want this ligature not to form; when any mark follows the mim but perhaps also when a mark occurs below the lam?

What you need to do in this case is create some separate lookups for this ligature, and a class for lower marks called e.g. 'marks_below', and another class for 'all_marks'.

Your first lookup will form the ligature only if no below mark follows the lam. This is done by putting the 'marks_below' class name in the Process Marks field:


___

The second lookup decomposes the ligature when the mim is followed by a mark. This requires the Process Marks field to be set to ALL, so that the 'all_marks' class will be recognised in the context string:

John Hudson's picture
Offline
Joined: 21 Dec 2002 - 11:00am
0

Not a full VOLT guide, but perhaps a useful: my step-by-step FontLab-to-VOLT workflow documentation.

Karl Stange's picture
Offline
Joined: 17 Sep 2009 - 10:07am
0

Thank you, John. There are several very good ristrettos with your name on them the next time you are in London. A good bottle of red if you like too.

Thomas Milo's picture
Offline
Joined: 24 Oct 2007 - 1:58pm
0

John Hudson wrote:

"Although all my own Arabic work has been in OpenType, I’m pretty convinced that only an approach like Tom’s can result in optimal mark positioning. The downside of such a system is that it tends to be very slow: not something you would want to use for, e.g. e-mail. But in the context of book production, it is the best thing available."

This was not based on actual observation. The core code for ACE was written in 1984 for 4Mhz processors :-)
See: http://laits.utexas.edu/rfpit/texts/Mimesis.html

Today it runs on 1Ghz, 2Ghz or even faster computers. ACE has never been slow, and it's certainly faster than OpenType.

Moreover, what's wrong with elegant email. Ruqah, one of our projects, is perfect for (e)mail. Anyway, ACE is technology, not a specific font.

Thomas Milo
DecoType
www.decotype.com

John Hudson's picture
Offline
Joined: 21 Dec 2002 - 11:00am
0

Sounds good, Karl. I should be in London 13-15 November.

erwindenissen's picture
Offline
Joined: 31 Aug 2007 - 1:12pm
0

If you haven't solved the problem yet, do contact me and I'll look into it.

Erwin Denissen

http://www.high-logic.com/

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

John Hudson,

i am waiting for your response. last day mail for your given ID.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

i work on Urdu Nastaleeq font but on working lot of problems any one help me.
My font contextual recognized shape is not working on Ms office.
adobe illustrator working very fine some time unexpectedly quite what happened.
i am very upset lot of work on it. thanks lot.

Karl Stange's picture
Offline
Joined: 17 Sep 2009 - 10:07am
0

It may help if you start a new thread specific to the problems you are encountering.

AzizMostafa's picture
Offline
Joined: 18 Apr 2006 - 10:43pm
0

Saleem Ali > My font is near to ready.

@ Your font needs QB http://typophile.com/files/QB3.3.pdf
Interested? Email FlowerCrosswords@gmail.com

Karl Stange's picture
Offline
Joined: 17 Sep 2009 - 10:07am
0

That works for me, I will email you off site to arrange something.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

How i use single Adjustment in Arabic font Plz describe in detail. prompt response i will be appreciate you.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

Dear John,
Plz above post consider.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

Dear John Hudson,

Thank you very much for your immediate response and your kindness will make me a better font designer in future. Again thank you, good men spread the knowledge. I will get more information about volt feature time to time whenever trouble shoot.

Best Regards,

Saleem Ali Ghalib

HanaTomhenry's picture
Offline
Joined: 2 Feb 2015 - 1:56am
0

very good Huz 6

AzizMostafa's picture
Offline
Joined: 18 Apr 2006 - 10:43pm
0

Oldnick + Karl Stange

Tracking the Old(nick) sounds St(r)ange?!

mjins2's picture
Offline
Joined: 24 May 2012 - 11:39am
0

since its initial idea - which interestingly grew from the sanctum of founder stefan siegel's shoreditch kitchen walls into a working databa -se of over 6,000 designers spread across 88 countries - the dedicated designer platform has certainly done what it says on the tin. a global howroom where trend scouts, fashion insiders and stylists are able to source talent, it isn't just an online shop, it's a veritable fashion spring- board, saturater with the ideas of tomorrow's designers, and eagerly digested in the process too. ladiva attractive layering of colorless enjoy! cody colorless casual style is so attractive. seemed to rock seemed unvarnished. we'll tiresome colorless chic the field is variegat with all kinds of flowering plants. and she worked a skirt and blouse since its initial idea which interestingly grew from the sanctum of ounder stefan siegel's shoreditch kitchen walls into a working database of over 6,000 designers spread across 88 countries - the dedicated designer platform has certainly done what it says on the tin. a global showroom where trend scouts, fashion insiders and stylists are able to source talent,
it isn't just an online shop, it's a veritable fashion spring-board, saturated with the ideas of tomorrow's designers, and eagerly digested in the process too. ladiva attractive layering of colorless enjoy! cody colorless casual style is so attractive. seemed to rock seemed unvar-nished. we'll tiresome colorless chic the field is variegated with all kinds of flowering plants. and she worked a skirt and blouse

Rahmi Tura's picture
Offline
Joined: 28 May 2007 - 9:07am
0
Jan Żurawski's picture
Offline
Joined: 3 Sep 2005 - 12:21pm
0

saleem Ali - have you got any glyph group empty?
After removing all empty groups (and lookups with them) my test font magically compiled.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

My problem is kashida word get in illustrator CS 6 rest of type in illustrator is very nice. I manually bay init tattwel= extend bay. I would auto adjustment extend bay how can I do. I attached trouble shoot pic.

John Hudson's picture
Offline
Joined: 21 Dec 2002 - 11:00am
0

Yeh barree is very difficult in OpenType, because there is no way to kern the yeh barree against the letter with which it is colliding on the right, because of all the intervening letters. OpenType GPOS simply isn't a good format for handling this kind of problem (which is one that I will be discussing in my presentation at TypeCon next week). You can only kern adjacent glyphs, so in order to handle these situations you would need to have a contextual kern between the final letter in the previous word and the following space glyph, in which the context is everything up to and including the yeh barree. It's a nightmare. Also note that some browsers treat each word as a separate glyph run, so context lookups across word spaces might not work even if you made them. As I said, a nightmare. Oh, and if you were to build all of the separate contextual lookups necessary to handle all the variable spacing contexts, chances are that you would overflow the GPOS subtable boundaries and the font wouldn't compile. Did I mention that this is a nightmare?

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

@ erwindenissen,
Any solved these problems for compiler issue. I hope you give good news for this font.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

@ john said: although if the problem is with GPOS data this can be more difficult.
absolutely right bugs in GPOS. how solved these problems?

Thanks for All valuable comments my question is how many GPOS in volt? My ligatures is 1500 to 2000 and may be exceed amount. My volt file consist one GPOS in 1500 ligatures and i put some more add encounter an unexplained compilation failure message as shown above pic,. Are you suggest GPOS make in volt 7 or above and different anchor name that above work flow is best, all work take 15 days and more, any have prior experience plz share me, again thanks for all and kindness.

Bahman Eslami's picture
Offline
Joined: 20 Feb 2007 - 12:47am
0

With the john's comment I think you have to come up with a design solution. That would be designing a shorter Yeh baree when the context behind it is short. So when you have a short word, the yeh should be substituted with a shorter yeh. @john is it possible or it's a nightmare too?

erwindenissen's picture
Offline
Joined: 31 Aug 2007 - 1:12pm
0

I've taken a quick look at your font project, but couldn't find any cause for the compiler issue. I'm afraid there is no easy way to overcome this problem, but if you manage to find a solution, let us know.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

Dear John Hudson,

Please tell me ligature bari yeh before init, medi form join the ligature, joint point, what you know any rule in volt please provide detail, i thankful to you whole life. My imge show above the problem.

Best regards,

Saleem Ali Ghalib

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

I have very difficulty in this situation word is overlaping each other. My problems is to see below how i calculate width in volt Single adjustment menu plz answer this problems: first is" kay inital and bari yeh fina, lam init, middle hamza and yeh fina Meem init, jeem meddle dochashmi hay and bari yeh fina" same as other words so you can see. How i solved and how calculate single adjustment menu table width define. I will be appreciate for your prompt action.
pic

John Hudson's picture
Offline
Joined: 21 Dec 2002 - 11:00am
0

Have you turned on 'Use extension lookups' in the VOLT options? This is the first thing to try when you encounter an unexplained compilation failure, which are most commonly caused by subtable overruns, i.e. your GSUB or GPOS data is too large to be compiled to a single subtable. Using the extension lookups mechanism often solves this, although if the problem is with GPOS data this can be more difficult.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

Dear John Husson,

Diacritic mark is not apply on ligature plz describe solve the problem.
I am marking glyph definition ligature 2 component and use anchor attachment but it is not working on MS word. Happening situation is fig: 2, fig:3

My previous post i learned lot of works in volt and succeed init, medi and fina change the baselineshift and word rising the desired ligature as fig; 1 i thank ful to you but the diacritic mark is not working on it. What happened.

I need to know if there is a way to de-ligate the Lam-Meem ligature when I put a mark on the Meem so that, when I put the mark for the Meem, the ligature is back seperated to Lam (initial) and Meem (medial), so the mark doesn't like it's the Lam's.

Plz describe the detail with screen shot if you don't mind i understand that is not your job it is merly your kindness yours suggestion is very helpful for me, font is very flat that has a great impact for urdu audience, designer support layout font.

Thanks,

Saleem Ali Ghalib

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

Are otf font with arabic feature work on adobe in design and illustrator Mac and PC computer?

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

Dear John hudson, Sergey Malkin
My Volt file is not Compile when insert new ligatures for nuqta placing i use add positioning in anchor attachment this file, put on My Group name is R_Liga-1, R_Liga-2,R_Liga-3,R_Liga-4,R_Liga-5 and R_Liga-6 every group contain 250 and above glyph problem is when i insert R_liga6 file is not compile message i am very afraid lot of work in volt file weight 3mb i several time change anchor name i try tool option use extension look up and other way but file is not compile message. Nuqtas_All_Ligatures up and down but its not compile.

If you are willing my fonts for investigation, John or Sergey thats solve these problems i thankful you whole life.

Thanks,
If i change anchor name file is compile its not good way because, i lot of work above group any one have mention above solution the critic situation of this problems if any quick response i really appreciate. This pic is describe these problems obviously.

Karl Stange's picture
Offline
Joined: 17 Sep 2009 - 10:07am
0

Are otf font with arabic feature work on adobe in design and illustrator Mac and PC computer?

Based on Adam Twardoch's recent thread: Adobe InDesign/Illustrator CS6: get Arabic/Hebrew support and extra languages for free I would think that the answer is yes, but not out of the box, so to speak. Prior to CS6 I think you would need to explicitly purchase the ME version.

John Hudson's picture
Offline
Joined: 21 Dec 2002 - 11:00am
0

Did you try turning on the 'Use extension lookups' option? Did that make any difference?

The subtable size limit is not some particular number of attachment points or lookup entries, but is a byte size limit. So you should try to make your GPOS lookups as efficient as possible, e.g. try to share anchor attachment points for different marks as far as possible.

An Arabic font that relies heavily on ligatures instead of contextual variants is always going to require more GPOS anchor points.

John Hudson's picture
Offline
Joined: 21 Dec 2002 - 11:00am
0

That's certainly an easier solution technically, Bahman. We've done this sort of thing in non-traditional typefaces to avoid having the yeh barree collide with dots or marks below preceding letters. But in the traditional styles, how much do you want to corrupt their norms in order to accommodate technological difficulties?

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

I am working on Nastaleeq urdu font it is not contextual alternate are not get what happened MS work .doc file attached the problems highlighted in circle. insert image error if you want see my problem give your email.

AzizMostafa's picture
Offline
Joined: 18 Apr 2006 - 10:43pm
0

Karl & saleem, this thread is specific to Nastaleeq :
http://typophile.com/node/82934
Stay tuned with developments?!

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

Thank you

Aziz Mostafa

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

Sorry erwindenissen i resend my problematic Project when i put R_L_5 group and show compile failed message.

JanekZ says:
saleem Ali - have you got any glyph group empty?
After removing all empty groups (and lookups with them) my test font magically compiled.

He say absolutely right when i see group some of glyph is empty i remove all empty glyph and then save no compiler issue. when i all bugs remove and recall disappeared glyph and put Nuqtas_All_Ligatures and then compile no compiled. what happening i no understand what issue in this project

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

Dear JanekZ ,
I am remove all glyph group empty successfully compile and then recall missing glyph that's compiler issue whats magic? I think my glyph go to Bermuda triangle that its disappeared. i try couple of time.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

I would try to do that above way: but compiler issue maintain.

m saleem's picture
Offline
Joined: 14 Feb 2010 - 11:17pm
0

I tray 'Use extension lookups' but no such as difference.

How i calculate bite size in volt? You know Any other software like volt where i compiled file.

Distribute in different GSUB and GPOS but not achieved my purpose. I try according by your second paragraph no solved my problems.

Ligatures must be include in this project that the nuqta placement and barri yeh positioning is best way to put nuqtas in ligatures, instead of contextual alternate and nuqta placement is so difficult in urdu Nastaleeq. So i think best way to required some more frequently ligature include my project. I have no idea encounter compiler issue.

If you have some time for my volt project i can send your personal email. you are prompt response i am highly appreciated with your kindness and sympathy.

Sergey Malkin's picture
Offline
Joined: 17 Feb 2004 - 2:07pm
0

You can always do mark positioning in context of neighbor glyphs. So you can move marks higer if there are marks over other letters that can cause collision.

Dealing with ligatures is essentially the same, but may be tricky, though. OpenType model can not distinguish between mark over one or another component, except for mark-to-liga attachments. For example, for sequence of ligature and mark, lookup can say that something should happen only if mark is logically attached to the first component. There could be workaround for that, for example having separate glyphs for mark over first second etc. component.

Thanks,
Sergey