search

Tarobish / Jomhuria

http://tarobish.github.io/Jomhuria/
SIL Open Font License 1.1
23 stars 6 forks source link

it seems the features of the glyphs are broken #7

Closed graphicore closed 9 years ago

graphicore commented 9 years ago

@Tarobish

please check the font i changed the colors of some characters in light blush and green it seem the feature of the glyphs has been broken when these glyphs are coming together (each one of the light blue in connection with each of the green one ) they have a different sort of connection the engine can read the cods but is something is broken like this

screen shot 2015-06-05 at 4 44 19 pm

zoghal commented 9 years ago

in naskh and Titr style of arabic/persian this is true: demo1 : http://share.pho.to/9SG7m/cr/original demo 2: http://share.pho.to/9SG8y/ty/original

davelab6 commented 9 years ago

@zoghal what are those 2 fonts? :)

zoghal commented 9 years ago

@davelab6 These two new Persian fonts. Diba and Ordibehesht.

davelab6 commented 9 years ago

That you designed? :)

zoghal commented 9 years ago

No, I'm not a designer, I am a developer. I'm working on a startup or project research that will help Iranian designers to design standard fonts for farsi.

davelab6 commented 9 years ago

Ok :) Where are these fonts from? On 12 Jun 2015 10:19 pm, "Saleh Souzanchi" notifications@github.com wrote:

No, I'm not a designer, I am a developer. I'm working on a startup or project research that will help Iranian designers to design standard fonts for farsi.

— Reply to this email directly or view it on GitHub https://github.com/Tarobish/Jomhuria/issues/7#issuecomment-111655419.

Tarobish commented 9 years ago

I know the designer Diba it has been design by damoon khanjanzadeh And ordibehesht it is an old Persian typeset has been developed recently

Sent from my iPhone

On Jun 12, 2015, at 8:26 PM, Dave Crossland notifications@github.com wrote:

Ok :) Where are these fonts from? On 12 Jun 2015 10:19 pm, "Saleh Souzanchi" notifications@github.com wrote:

No, I'm not a designer, I am a developer. I'm working on a startup or project research that will help Iranian designers to design standard fonts for farsi.

— Reply to this email directly or view it on GitHub https://github.com/Tarobish/Jomhuria/issues/7#issuecomment-111655419.

— Reply to this email directly or view it on GitHub.

zoghal commented 9 years ago

Diba : designer : Damon khan janzade

Ordibehesht : orginal designer :unknown designer (main source is letraset) this front has The two designers claim fonts and each chose a different name. ordibehest and todeh :d fallow this link

i am supporting and help to ordibehesht for publishing with ofl license (https://github.com/font-store/Font-Ordibehesht)

graphicore commented 9 years ago

@Tarobish I think the first one (on the right side, with aKaf.medi_KafLam plus the LamAlfFina) is broken because the left side bearing is to far on the left side. See the image below.

The glyph should rather start where the base-line connects. Or you do it with kerning, but either way, I think that's your task.

akaf medi_kaflam at 197 from jomhuria-regular_053

I'm checking the other combination now.

Tarobish commented 9 years ago

No, it has to be like this. the point is, it will be connected with the right side of the next character. please see below. or you think it has to be done by kerning? please let me know screen shot 2015-06-22 at 12 15 21 pm screen shot 2015-06-22 at 12 15 35 pm screen shot 2015-06-22 at 12 15 59 pm screen shot 2015-06-22 at 12 16 20 pm screen shot 2015-06-22 at 12 16 25 pm screen shot 2015-06-22 at 12 19 49 pm

graphicore commented 9 years ago

Please stay at the first example. I think we have a different issue with the second one.

Yes I think it has to be done with kerning. But not with the kern feature, but with the glyph being shorter itself.

graphicore commented 9 years ago

Like this:

akaf medi_kaflam at 197 from jomhuria-regular-_054

graphicore commented 9 years ago

On the other hand. You show ligatures uniFCEB uniFC80 etc. Do you think Amiri used these? The feature must have gone lost in Toronto then.

graphicore commented 9 years ago

The original character, with the sidebearing you gave it, how is it supposed to connect anyways? I can't imagine that there is a single working combination this way.

Tarobish commented 9 years ago

yes, This is from Amiri, i believe we kept this feature. otherwise we lost in Toronto.

Tarobish commented 9 years ago

It's simple. the red lines are R&L bearing from the "Kaf" and the blue one belong to the next character "lam" easily they will be connected. please let me know if I'm wrong :) 773b10be-18d9-11e5-87f6-b815c6340254

graphicore commented 9 years ago

If these shapes meet each other in text, the red and blue lines in the middle would be aligned. Therefore the gap happens.

graphicore commented 9 years ago

In Amiri, the baseline ends with the left side-bearing:

akaf medi_kaflam at 65802 from amiri-bold_055

graphicore commented 9 years ago

This is what we are doing here: Kerning

The solid block under the glyph shapes, that's the side bearings. Think about your red/blue line example in these terms.

Tarobish commented 9 years ago

I think, the baseline ends with the left side-bearing because of the Khaled's design. if we do that our ligature its going to be like this :( I'm right? or I'm confused :( screen shot 2015-06-22 at 1 29 44 pm

graphicore commented 9 years ago

The thing is that we don't actually show the ligature in the text. The ligature is in this legacy unicode code block, that is not really used anymore. What we do instead is showing uni0643.med_KafLam and then uni0644.medi_KafĹam. But for that, the glyph uni0643.med_KafLam is broken. Now, when you repair the glyph, your Ligature breaks, because the positioning is done relative to the left side of the glyph (which is zero in the coordinate system).

graphicore commented 9 years ago

I just fixed the other issue with the other combination. Now we have the same problem there:

selection_056

graphicore commented 9 years ago

I updated the pdfs and gh-pages, you can observe there.

Tarobish commented 9 years ago

Now I got :) I'll fix it :) I have to redesign these characters Thanks

Sent from my iPhone

On Jun 22, 2015, at 1:57 PM, Lasse Fister notifications@github.com wrote:

I just fixed the other issue with the other combination. Now we have the same problem there:

— Reply to this email directly or view it on GitHub.

graphicore commented 9 years ago

I made two test tables:

http://tarobish.github.io/Jomhuria/#tests/kaf-lam-alf-1 http://tarobish.github.io/Jomhuria/#tests/kaf-lam-alf-2

khaledhosny commented 9 years ago

If you want the kaf to touch the lam, then you should decrease the left side bearing, otherwise you can always have a contextual form for the kaf when followed by lam.

Tarobish commented 9 years ago

Thanks, got it. i have to redesign the “Kaf"

On Jun 23, 2015, at 9:34 AM, Khaled Hosny notifications@github.com wrote:

If you want the kaf to touch the lam, then you should decrease the left side bearing, otherwise you can always have a contextual form for the kaf when followed by lam.

— Reply to this email directly or view it on GitHub https://github.com/Tarobish/Jomhuria/issues/7#issuecomment-114565023.

graphicore commented 9 years ago

@Tarobish Why don't do just as I did here and then repair the ligature glyphs. What do you plan to redesign?

Tarobish commented 9 years ago

thats what i want to do (the way you did it) but i have to work on the top side of the “Kaf” this way its going to be a bit short.

On Jun 23, 2015, at 9:43 AM, Lasse Fister notifications@github.com wrote:

@Tarobish https://github.com/Tarobish Why don't do just as I did here https://github.com/Tarobish/Jomhuria/issues/7#issuecomment-114232972 and then repair the ligature glyphs. What do you plan to redesign?

— Reply to this email directly or view it on GitHub https://github.com/Tarobish/Jomhuria/issues/7#issuecomment-114567193.

graphicore commented 9 years ago

OK

Tarobish commented 9 years ago

I just committed, please check it and let me know

graphicore commented 9 years ago

Looking good :-) check the pdfs and this stuff:

http://tarobish.github.io/Jomhuria/#tests/kaf-lam-alf-1 http://tarobish.github.io/Jomhuria/#tests/kaf-lam-alf-2

Tarobish commented 9 years ago

Perfect :) screen shot 2015-06-23 at 2 34 58 pm

On Jun 23, 2015, at 2:21 PM, Lasse Fister notifications@github.com wrote:

Looking good :-) check the pdfs and this stuff:

http://tarobish.github.io/Jomhuria/#tests/kaf-lam-alf-1 http://tarobish.github.io/Jomhuria/#tests/kaf-lam-alf-1 http://tarobish.github.io/Jomhuria/#tests/kaf-lam-alf-2 http://tarobish.github.io/Jomhuria/#tests/kaf-lam-alf-2 — Reply to this email directly or view it on GitHub https://github.com/Tarobish/Jomhuria/issues/7#issuecomment-114645039.