I'm creating a distressed all caps font and would like to use a chaining context table for the first time to simulate an automated substitution. Each lowercase letter will have 3 alternates. For example (a= a.alt1, a.alt2, a.alt3)
My goal is twofold:
1. For alternate glyphs to cycle in place of letters keyed multiple times. If you were to key the word "Mississipi" for example all 4 lowercase s would be different. s, s.alt1, s.alt2, s.alt3
2. For single letters to cycle sequentially in an automated process. If you were to key "run run run" you would get r/u/n/space/r.alt2/u.alt2/n.alt/2/space/r.alt3/u.alt3/n.alt3
here's a screen shot from a font that does this successfully
I've experimented with an existing DEMO font by examining the code and trying out the subtables and substitutions on my font. My first goal is partially accomplished but stops at the 2nd alternate glyph. So if you were to key "ssss" you'd get s, s.alt1, s.alt2, s. I've also tried adding classes to the existing subtables and that partially solves my 2nd goal but completely ruins the first.
I'm beating my brain trying to figure out how this all works and haven't had any luck searching for anything on the web. I just don't completely understand the process or why certain glyphs are in certain classes.
If there's a good way to do this from scratch I'd love to hear (or even a good tutorial somewhere) Thanks.
question about contextual chaining context
question about contextual chaining context
Last edited by Dennis on Sat Dec 17, 2016 3:08 am, edited 1 time in total.
-
- Top Typographer
- Posts: 9878
- Joined: Tue Oct 29, 2002 5:28 am
- Location: Seven Kings, London UK
- Contact:
Re: question about contextual chaining context
Do you have a font that behaves in this way? It sounds like a complex and uncommon requirement to me. Maybe the Randomize feature would be easier? This was my first guess, and seems to work OK.
- Add a Randomize feature
- Add an alternate substitution for a -> a a.alt1 a.alt2 a.alt3
- I used a -> a b c d
Re: question about contextual chaining context
Bhikkhu,
I'll be happy to share the font with you. Check your email.
Thanks for the info on the randomize feature. I gave it a try and couldn't get it to work. It would only substitute the first alternate I plugged in.
I'll be happy to share the font with you. Check your email.
Thanks for the info on the randomize feature. I gave it a try and couldn't get it to work. It would only substitute the first alternate I plugged in.
-
- Top Typographer
- Posts: 9878
- Joined: Tue Oct 29, 2002 5:28 am
- Location: Seven Kings, London UK
- Contact:
Re: question about contextual chaining context
There's no way that I can see of testing the randomize feature in FontCreator. The Preview Toolbar will let you cycle through the alternates by scrolling the set on the toolbar but I had to test the font in PagePlus.
You can look at the code for the font that you sent me in the code editor, but rather you than me working out how to use that.
You can look at the code for the font that you sent me in the code editor, but rather you than me working out how to use that.
-
- Moderator
- Posts: 11158
- Joined: Fri Oct 04, 2002 12:41 am
- Location: Bilthoven, The Netherlands
- Contact:
Re: question about contextual chaining context
Currently FontCreator doesn't show how the random feature could work, but we can add such support.Bhikkhu Pesala wrote:There's no way that I can see of testing the randomize feature in FontCreator.
Even though you can't see it working in FontCreator, it will work in software that has implemented the Random feature.Bhikkhu Pesala wrote:The Preview Toolbar will let you cycle through the alternates by scrolling the set on the toolbar but I had to test the font in PagePlus.
Does PagePlus use "random" or does it step through the sequence of alternates?
-
- Top Typographer
- Posts: 9878
- Joined: Tue Oct 29, 2002 5:28 am
- Location: Seven Kings, London UK
- Contact:
Re: question about contextual chaining context
It looks fairly random to me. PagePlus offers the four alternate sets to choose from, and each gives a different, but apparently random result. It's obviously not random though, as if you delete and retype the text, it gives the same result. Here is a more extensive test set.Erwin Denissen wrote:Does PagePlus use "random" or does it step through the sequence of alternates?
-
- Top Typographer
- Posts: 9878
- Joined: Tue Oct 29, 2002 5:28 am
- Location: Seven Kings, London UK
- Contact:
Re: question about contextual chaining context
The latest update (build 2265) has added support for displaying the results of the Randomize feature in the Preview toolbar and OpenType Designer.Erwin Denissen wrote:Currently FontCreator doesn't show how the random feature could work, but we can add such support.
How's that for quick service?